• 专利附录
  • 专利说明
  • 权利要求
  • 类似技术
  • 同族专利
  • 引用文献
  • 法律状态
  • 优先权
    • 专利摘要:
    Activity detection device/physiological inactivity configured to allow the detection of activity states or physiological inactivity of people, in a non-intrusive way and without direct contact, based on the detection of movements and vibrations using the measure of changes in the pattern speckle produced in a multi-mode optical fiber (14, 24, 34, 44) when conducting coherent light inside, comprising: at least one multi-mode optical fiber (14, 24, 34, 44) as an element of transduction, linked to a coherent light source (12, 42) and to an optical detector (11, 41); a processing block (13, 43) connected to the optical detector (11, 41), configured to implement the processing method necessary to issue a decision; connecting elements (15, 45, 55) configured to connect/disconnect the multi-mode optical fiber (14, 24, 34, 44) to/from the optical detector (11, 41) and/from the coherent light source (12, 42). A processing method configured to detect the activity states or physiological inactivity of people, using the defined device. (Machine-translation by Google Translate, not legally binding)
    公开号:ES2610181A1
    申请号:ES201600626
    申请日:2016-07-22
    公开日:2017-04-26
    发明作者:Luis RODRIGUEZ COBO；Mauro M. LOMER BARBOZA；José Miguel López Higuera；Alberto RODRIGUEZ CUEVAS
    申请人:Universidad de Cantabria；
    IPC主号:
    专利说明:


    ACTIVITY DETECTION DEVICE DEVICE PHYSIOLOGICAL BASED ON OPTICAL FFFIRA FIELD OF THE INVENTION
    The present invention belongs to the field of monitoring vital signs of people, and more specifically, to the methods and apparatus for monitoring in continuous physiological activity / inactivity in patients. BACKGROUND OF THE INVENTION
    There are currently a large number of systems for monitoring vital signs and the physiological state of people. However, the vast majority of these systems, as well as the methods they use, have certain disadvantages since they are either intrusive, or require direct contact with the person's skin, so that continuous monitoring for long periods of time it becomes very uncomfortable for the patient, and therefore it is not usually carried out outside the intensive care units.
    In recent years several systems and technologies for monitoring vital signs and physiological activity without direct contact with the skin of the person have been studied and developed, [Kitsiou S, Paré G, Jaana M Effects 01 Home Telemonitoring Interventions on Patients With Chronic Heart Fai / ure: An Overview 01 Syslematic Reviews. J Med Internet Res 2015; 17 (3): e63]. Among the proposed techniques, the following stand out: balistocardiography techniques, capacitive electrocardiogram, magnetic impedance monitoring or fiber optic-based techniques [Teichmann, D., Bruser, c., Eilebrecht, B., Abbas, A., Blanik, N. , & Leonhardt, S. (2012). Non-contact monitoring techniques -principIes and applicationsj.
    Fiber optic based techniques are becoming especially important because
    to its versatility for various medical environments and some of these techniques can be implemented at a low cost compared to other systems. In the article submitted by Lukasz Dziuda HA review of applicable technologies and relevance I used it during magnetic resonance imaging procedures quot; [Dziuda, L. (2015). Fiber-oplic 5 sensors for monitoring palienl physiological parameters: A review of applicable lechnologies and relevance to use during magnelic resonance imaging procedures. Journal of Biomedical Optics, 20 (l)} includes a set of techniques based on optical fiber that allow monitoring vital signs and physiological activity of people, including the analysis of the SpeckIe pattern generated in an optical fiber.
    1 0 This technique stands out from the set of fiber optic techniques due to its low cost and high sensitivity.
    The Speckle fiber optic interference principle occurs when a coherent beam of light propagates through a multimode optical fiber, projecting a structured light pattern. This pattern consists of a large number of bright spots of light (from English: SpeckIe) on a dark background, produced by a phenomenon of intermodal interference. The pattern of SpeckIe varies slowly due to various factors, but its total intensity remains constant. Any external disturbance made on the fiber affects the SpeckIe pattern. Therefore, an adequate analysis of the changes in the pattern can allow us to extract useful information from the disturbances and give applicability to this system, [Rodriguez-Cobo, L., Lomer, M, Galindez, c., & Lopez-Higuera, J .M (2012). POF vibration sensor based on speckle pattern changes. OFS2012 22nd International Conference on Oplical Fiber Sensor, pp. 84212Y-84212Y-4], [Rodriguez-Cobo, L., Lomer, M, Galindez, c., &
    25 Lopez-Higuera, J. M (2012). Speckle characterization in multimode fibers for sensing applications. SPECKLE 2012: V International Conference on Speckle Metrology, pp. 84131R-84131R-6], [M Lomer. (2013). Fiber Oplic Speckle Phenomenon and lIs Applications Sensors. Tecnia 22 (2), (2012)].
    3 0 This type of technology has already been used several times to measure or monitor disturbances of different origins. In this sense, one of the first patents requested is US4297684A, published in 1979. This patent presents the basic equipment necessary to generate and detect the Speckle pattern, consisting of a coherent light source a multimode optical fiber and an optical detector . This system is intended to protect security perimeters by detecting intrusions that generate small pressures or deformations in the fiber. The system uses the Speckle effect on fiber optics in a simple but effective way, indicating intrusion when there are changes in the pattern of light points and non-alteration states when there are no significant changes in the pattern. In addition, patents published later also proposed the use of the Speckle interference effect for the detection of vibration in buildings, seismographic, medical applications, surveillance systems or fracture detection among others (JP2008309497; US4863270A).
    One of the great advantages of the Speckle phenomenon in fiber optics is its enormous sensitivity to changes in pressure, vibrations and mechanical movements. This sensitivity allows to detect changes in the fiber caused by the vibrations generated by the heartbeat and breathing even if the optical fiber is not in direct contact with the person's skin. For this reason, in inventions after the initially described patent (US4297684A) the use of the Speckle phenomenon for medical applications begins.
    In 1993, the first invention was published in which the Speckle effect is used for monitoring heart rate and respiratory rate, US5212379A (Nafarrate Antonio B; Rawson Eric G). In this patent the traditional generic system is used to generate the Speckle effect, consisting of a coherent or partially coherent light source, a multi-mode optical fiber and a photo detector element. In this case, it uses a photodiode as a photo detector element.
    In the invention proposed by Nafarrate Antorino B or Rawson Eric G, it is proposed to insert the optical fiber inside blankets, sheets, mattresses or sleepwear. This invention has important shortcomings that make it difficult for it to be applied in real situations. The use of a single photodiode implies obtaining a single intensity value. Each point of the Speckle has a certain intensity value that changes when the fiber is subjected to a disturbance, the fact of using a photodiode involves grouping a set of points of the Speckle pattern and analyzing them as a whole, obtaining the total intensity of that set and not that of each of the points individually. This implies that enough information is being lost since the light intensity transfer values that occur within the selected region itself are not taken into account, nor can it be determined what compensation has been made between the intensity that has come out of the modes of the selected region and the one that has entered them. For this reason, the use of a single photodiode as a photo detector element implies a great loss of information and makes the necessary calibrations difficult. This fact, in practice, causes the range of values between the maximum and the minimum to be several orders of magnitude less than with other optical detectors; reason why it causes a great difficulty in the classification and differentiation of the vibrations that are generated by slight movements, by breaths or by beats. For this reason the applicability of this system is more limited.
    On a scientific level, one of the most important publications in which the Speckle effect is used in fiber optics to monitor vital signs is the article "A Smart bedfor patient monitoringquot; published by Spillman et al. in 2004 [Spillman Jr., W B., Mayer, M, Bennett, J, Gong, J, Meissner, K. E., Davis,
    B., al. (2004). A 'smarl' bed for non-intrusive monitoring ofpatient physiological factors. Measurement Science and Technology, 15 (8), 1614-1620.]. This article presents the method and results of pulsation and breath monitoring using the Speckle fiber optic effect. The article shows the measurements for different positions, as well as different graphs with different types of processing. Other research has made more progress in the application of this contactless monitoring technique, applying new pattern processing techniques as well as new geometries and application procedures [Lomer, M, RodriguezCobo, L., Revilla, P., Herrero, G ., Madruga, F., & Lopez-Higuera, JM (2014).
    Speckle POF sensorfor detecting vital signs ofpalients. OFS2014 23rd 1nternationa! Conference on Optics! Fiber Sensors, pp. 915721-915721-4]. Despite the evolution, all reported processing methods focus on signal conditioning, an expert eye being necessary to make the final decision.
    William Spillman himself along with his research team filed a patent that protects the system presented in the article: W02004046869A2, published on 06/03/2004. In the description of the patent they incorporate many technical innovations with respect to previous patents and the Nafarrate patent. Among the elements that are specified can be used, various coherent light sources that may be applicable are listed: laser pointers, laser diodes or others. Various photo detector elements: photodiodes or CCD cameras. It is precisely the CCD camera, the system that provides greater precision when identifying disturbances. Lastly, it also specifies that several types of multimode optical fibers can be used, present at the time in the market, as well as different diameters that provide more or less number of modes. It is also specified that the system can be coupled to a large number of domestic or hospital elements, such as beds, chairs, armchairs, blankets or sheets. However, it is in the way of transmitting the data wirelessly as well as in the signal conditioning where the most novel change occurs. This patent describes two complementary methods for obtaining the temporal signal from the sequence of frames detected by the camera, as well as subsequent methods for reducing noise. Despite the techniques described, the result of the proposed system has to be interpreted by an expert eye to make a judgment.
    In the Spillman et al. A digital system was developed that replaced the previous analog systems. The use of a CCD camera allows a differential analysis of the pixels (Digital Image Unit) of a frame against the equivalent pixels of the immediately previous frame in time. Although the method of comparing the Speckle light pattern with the immediately previous pattern is the foundation of many of the systems that use this technology, the fact of incorporating a CCD camera allows to extract the information of the variation of the light intensity of each point of the Speckle pattern and thus the information obtained from the study of the images of the Speckle pattern is much more accurate than the analysis that can be obtained with a single photodiode.
    Beyond the replacement of the components present in the previous inventions with more modern ones that provide greater precision, Spillman et al. They introduce the most prominent innovation element in the way of sending the data. In this invention, the numerical data directly obtained with the detector is transmitted wirelessly from the measuring equipment to a computer system in which the processing is performed. Once the data arrives at the computer system, various filters and a frequency study are used, implemented to increase the noise signal quality and to be able to distinguish more clearly the pulsations, breathing and movement. This digital processing system is much more accurate than previous analogue processes and greatly improves monitoring accuracy.
    However, despite the use of a wireless data transmission system, the need for an expert eye that interprets the signals conditioned on a computer is not eliminated. This supposes a clear barrier to its applicability in real monitoring situations and increases the cost considerably.
    The patent published by Spillman et al. It was not the last one that used the Speckle effect to monitor people. Other subsequent patents have followed this line of application. However, none of them have solved the underlying problems that were found in the Spillman et al.
    In US7532781 B2 the optical fiber with a predefined geometry between two rigid surfaces is introduced. As indicated in the description, this fiber placement system makes the system suitable for multiple uses. Two of
    The applications mentioned are the detection of vital signs and the detection of states of activity / inactivity in people. However, due to the characteristics of the invention in which rigid surfaces are used, the usefulness in medical application environments by means of long-term monitoring is greatly reduced, since it would generate significant comfort problems and could even generate pressure ulcers in the skin of the monitored person. In addition to that problem, the size of the equipment makes its installation and handling more complex.
    As explained above, monitoring using the Speckle pattern
    1 0 involves monitoring vibrations or deformations of the multimode fiber. The vibrations that occur in the multimode fiber are a consequence of breathing, pulsations and movement. However, the movement, however small, usually generates large changes in the Speckle light pattern in this way when performing the differential analysis of the frames, constant maximum values are generated that saturate the
    15 measuring ranges. For this reason during these periods it becomes impossible to detect movements or vibrations of less intensity. In this way, the use of the Speckle pattern analysis in optical fiber to monitor vital signs, breathing and heart rate becomes very complex when the person is awake and in a conscious state since, usually people tend to perform
    20 voluntary or involuntary movements as usual in these conditions.
    That is why, fiber optic-based devices for monitoring vital signs using Speckle that have been developed or are present today, present great disadvantages. Even the mechanical movements that are generated by
    25 speaking, coughing or swallowing saliva are of greater intensity than those due to breathing and beating so it would not be possible to detect the latter.
    On the other hand, the possibility of detecting states of activity and inactivity of people leads to the possibility of detecting cardiorespiratory arrest. This is especially useful among chronic patients, elderly people and people with a history of heart failure, as it offers the possibility of acting in spaces
    of time in which the person is clinically reanimable [Changzhi Li, Jenshan Lin, & Yanming Xiao. (2006). Robust overnight monitoring ofhuman vital signs by a non-contact respiration and heartbeat]. Although the monitoring of physiological activity is useful in any environment; in the cases of people living alone
    or they are admitted to the hospital or residential environment, this type of monitoring acquires special importance since the quality of care is greatly improved. Therefore, there is a need in hospitals, medical centers, nursing homes and even in domestic settings to detect episodes of cardiorespiratory arrest in sick people or with states of functional impairment.
    However, all existing systems in the state of the art for the detection of physiological activity / inactivity in patients, based on the analysis of the Speckle pattern generated in an optical fibro, present some of the following disadvantages:
    - Need for an expert eye that interprets the signals generated by the measurement system.
    - Need to use an additional computer to condition the signals from the photo-detected raw data, which reduces the applicability of the system and increases its cost.
    - Need for prior calibration. Although in the previous systems the need to be previously calibrated is not mentioned, the reality is that the people who hypothetically could use these monitoring systems are medically very different. For example, elderly people with very low levels of physical activity generally have very mild and difficult to detect pulsations, while young people who regularly practice sports have higher intensity pulsations since in each pulsation they move greater blood volumes For this reason, in the case that the systems have been designed to detect standard pulsation measures, in reality, they could easily confuse the pulsations of greater intensity with body movement, while in older people or with weaker heartbeats, I could confuse the slight variations in the Speckle pattern caused by these beats with the normal noise levels that always occur in the Speckle patterns in fiber optics. In this way and to avoid these errors, the need that exists in previous inventions to calibrate the equipment to adjust it to the parameters of each person becomes evident.
    - Need for proper placement of the optical fiber depending on the physical characteristics (mainly height and weight) of the person being monitored.
    - Constant risk of the system being contaminated by bodily or hospital fluids of all kinds. This fact added to the demanding levels of hygiene in hospital settings mean that the system must be washed quite frequently. SUMMARY OF THE INVENTION
    The present invention seeks to solve the aforementioned drawbacks by means of a physiological activity / inactivity detection device based on multi-mode optical fiber and a processing method, configured to allow the detection of the states of physiological activity or inactivity of people, of non-intrusive way and without direct contact. This device is based on the detection of movements and vibrations using the measurement of the changes in the Speckle pattern produced in a multi-mode optical fiber when coherent light is driven inside. The device is capable, not only to perform the differential calculation of the frames, but also to process and make decisions according to the results obtained.
    Specifically, in a first aspect of the present invention, there is provided a physiological activity / inactivity detection device configured to allow the detection of the states of physiological activity or inactivity of people, in a non-intrusive manner and without direct contact, based on the motion and vibration detection using the measurement of the changes in the Speckle pattern produced in a multi-mode optical fiber when coherent light is driven inside, comprising:
    - at least one multi-mode optical fiber as a transduction element, linked to a coherent light source and an optical detector;
    - a processing block connected to the optical detector, configured to implement the processing method necessary to issue a decision, thus eliminating the need to use an additional computer, such that said processing block and the optical detector form a detection block and socket decision;
    - connection elements configured to connect / disconnect the multi-mode optical fiber to / from the optical detector and to the coherent light source, with sufficient precision so as not to disturb the operation of the system;
    in such a way that during the operation of the device, the light coming from the coherent light source is propagated through the multi-mode optical fiber until the optical detector is reached. Said optical detector detects the Speckle pattern formed at the output of the multi-mode optical fiber, and the processing block analyzes the raw signals of the optical detector, emitting activity / inactivity data with the current state.
    In a possible embodiment, the device further comprises an alarm system connected to the processing block, with its corresponding signaling elements to alert of inactivity.
    In a possible embodiment, the device further comprises a wireless block connected to the processing block, configured to send the information generated and processed in the device to a central computer.
    In a possible embodiment, the device further comprises a battery connected to all the elements of the device that need power.
    In a possible embodiment, all the elements of the device of the invention, except for the multi-mode optical fiber -except its ends- and the connection elements, are located inside a protective box that offers care, in case of bumps and falls. , to the set of elements that confront the device. In a particular embodiment, the protective box is made of plastic and has at least two holes close and aligned with the coherent light source and the optical detector, thus allowing the connection of the multi-mode optical fiber to the coherent light source and to the optical detector.
    In a possible embodiment, the connecting elements are connected by a source-detector part, fixed to both the coherent light source and the optical detector, and by a fiber part, fixed to the multi-mode optical fiber, such that the part source-detector comprises two hollow channels, one of the channels being in direct contact with the coherent light source at one of its ends, and the remaining channel, at its end closest to the coherent light source, in direct contact with the optical detector, and such that the fiber part comprises two hollow channels with an inner diameter substantially larger than the diameter of the multi-mode optical fiber, and with an outer diameter substantially smaller than the inner diameter of the channels of the detector source part, the ends being of the multi-mode optical fiber inside the channels of the fiber part, so that each end is placed in a different channel and such that during its placement they are introduced by r two adjacent holes, the rest of the multi-mode optical fiber outside the fiber part, so that during the operation of the device, to connect the multi-mode optical fiber to the coherent light source and the optical detector , the channels of the fiber part are located inside the channels of the source-detector part, in such a way that the ends of the multi-mode optical fiber come into contact with the optical detector and the coherent light source.
    In a possible embodiment, the fiber part has a plurality of external micro grooves, and the source-detector part has a plurality of internal micro grooves, configured to allow both pieces to form a maximum stop between them and to exert resistance to movement preventing the one piece slip
    5 over the other by bumps and rubs. In another possible embodiment, the connection elements are further formed by a fixing part located attached to the fiber part, at its opposite end where the coherent light source and the optical detector are located, and which allows the whole assembly to be fixed to an optional support that simplifies its installation.
    In a possible embodiment, there is a single multi-mode plastic optical fiber, attached at one end to the coherent light source, and at its opposite end to the optical detector.
    In a possible embodiment, the coherent light source is a semiconductor laser. In another possible embodiment, the optical detector is a CCD camera.
    In another aspect of the invention, a processing method configured to detect the states of physiological activity or inactivity of people is provided, using the device defined above. The method comprises the steps of:
    20 -for each frame of the Speckle pattern obtained in the optical detector, performing a differential analysis of the pixels is of said frame against the equivalent pixels of the immediately previous frame in time, such that the difference in the change of intensity in each pixel of the frame;
    25 -make, for each frame analyzed, the sum of all values differs from the intensity change thus obtaining its value of variation of the Speckle pattern, so that as time progresses and each received frame is analyzed
    in the optical detector a new value of a differential sequence of 30 variations is obtained that summarizes the variations of the multi-mode optical fiber along the
    time, and that are detected by the optical detector at each instant;
    - when an individual begins to be monitored and threshold detection values have not yet been calculated, establish an initial detection threshold value for the differential sequence of variations, below which it is considered that there is no activity, such that said threshold value of Initial detection corresponds to the total intensity value that reaches the optical detector, due to the light that propagates through the multi-mode optical fiber. In a possible embodiment, to obtain the initial detection threshold, the coherent light source is turned on and off to obtain maximum values - when the coherent light source is on - and minimum - when the coherent light source is off - of total intensity received in the initial set of frames, such that the minimum intensity value is subtracted from the maximum intensity value, thus obtaining the total intensity value that reaches the optical detector through the multi-mode optical fiber , said value being scaled according to coefficients dependent on the type of multi-mode optical fiber used.
    - analyze the differential sequence of variations in at least one time window that moves over time - at least one time window for the calculation of detection thresholds and at least one time window for the calculation of indices of temporary activity - as long as the size of each time window is less than the time elapsed in the differential sequence of variations, such that each time window has its origin at the current time and its end a few units of time before, and such that each value of the differential sequence of variations it is understood by each time window in different instants of time as it moves, so that each time window, at each instant of time determined, comprises a fragment of the differential sequence of variations;
    - determine whether the fragment of the differential sequence of variations comprised in each time window intended for the calculation of thresholds is periodic;
    - if the sequence fragment is periodic determine, in each time window destined to the calculation of thresholds, the detection threshold from the maximum and minimum values of the sequence fragment. In one possible embodiment, the detection threshold is equal to the average value of the median of the maximum values and the median of the minimum values of the fragment of the differential sequence of variations;
    - determine the index of temporal activity of the differential sequence fragment of variations comprised in each temporary window destined to the calculation of indices, such that said index is obtained from the last detection threshold obtained, calculating, in each temporal window destined to the calculation of indices , the relationship between the number of times that the fragment of the differential sequence of variations exceeds the detection threshold with the times that it does not exceed it and scaling said value with the duration of the index window;
    - combine the obtained temporary activity indices, following statistical models to offer a judgment, thus determining whether or not activity has been detected and if the inactivity alarm is activated.
    In a possible embodiment, the method further comprises the step of:
    - Perform signal processing of the differential sequence of variations, in order to reduce the existing noise.
    In a possible embodiment, there is a single time window for the calculation of both parameters - thresholds and indices - such that for each fragment of the differential sequence of variations a threshold detection value is obtained - in the event that said fragment is periodic -and an index of temporary activity from the last detection threshold obtained. Alternatively, there are two temporary windows of different duration, such that in each instant of a given time a temporary activity index is obtained from the window of shorter duration, and a detection threshold from the window of longer duration. Alternatively, at least two indices of temporal activity and / or at least two detection thresholds are obtained from each single time window.
    In a possible embodiment, to decide if the inactivity alanna is activated, the last 150 values obtained from activity indices are computed, such that if a decrease in progressive activity is detected, followed by a stabilization in activity values close to O, the inactivity alanna is activated. Alternatively, to decide if the inactivity alanna is activated, a Gaussian filtering is performed on the Activity Indices obtained, using two windows of different duration, such that starting from the window of greater duration, its statistics are calculated dynamically determining a decision point that determines the trend of the detected signal, and based on the statistics obtained, a filter is defined that models the trend of the activity index associated with the longer duration window, thresholding said probability to decide if the alanna is activated of inactivity, so that by applying the shorter window to this filter, instantaneous anomalies can be detected that enable the inactivity alarm to be activated. BRIEF DESCRIPTION OF THE FIGURES
    In order to help a better understanding of the features of the invention, in accordance with a preferred example of practical realization thereof, and to complement this description, a set of drawings is attached as an integral part thereof, whose character is Illustrative and not limiting. In these drawings:
    Figure 1 shows a basic diagram of the fundamental elements comprising the device of the invention.
    Figure 2 shows two possible configurations for the placement of the device of the invention.
    Figure 3 shows two possible embodiments of multimode fiber optic geometry.
    Figure 4 shows a more complete diagram of the device of the invention.
    Figure 5 shows a possible embodiment of the connection elements for connecting the multi-mode optical fiber to both the coherent light source and the optical detector.
    Figure 6 shows a sequence of 4 different frames of the same sequence and consecutive in time, of a capture of the Speckle pattern, as well as a differential processing of that sequence and an example of signal processing.
    Figure 7 shows a periodic fragment of the differential sequence of variations comprised of two temporary windows: a window with a longer duration for calculating the detection threshold and a window with a shorter duration for calculating the indexes of temporal activity.
    Figure 8 shows another fragment in which the final part shows a non-periodic behavior of the differential sequence of variations comprised of two time windows: a window of longer duration for the calculation of the detection threshold and a window of shorter duration for the calculation of the Temporary Activity Indices. DETAILED DESCRIPTION OF THE INVENTION
    In this text, the term quot; comprehequot; and its variants should not be understood in an exclusive sense, that is, these terms are not intended to exclude other technical characteristics, additives, components or steps.
    In addition, the terms "; approximately"; "; substantially"; "" about ";" "some"; they should be understood as indicating values close to which these terms accompany, since due to calculation or measurement errors, it is impossible to achieve with total accuracy.
    In addition, in the context of the present invention, a frame is understood as a two-dimensional capture made with an image sensor, consisting of NxM pixels and collecting the Speckle pattern projected by the measured multi-mode optical fiber.
    Furthermore, in the context of the present invention it is understood that the elements: optical detector, processing block, alarm system or wireless block, further comprise the conditioning electronics necessary for its correct operation, such as capacitors, resistors, etc.
    The following preferred embodiments are provided by way of illustration, and are not intended to be limiting of the present invention. In addition, the present invention covers all possible combinations of particular and preferred embodiments indicated herein. For those skilled in the art, other objects, advantages and features of the invention will be derived partly from the description and partly from the practice of the invention.
    The following describes the physiological activity / inactivity detection device based on multi-mode fiber optics, and the processing method, configured to allow the detection of physiological activity or inactivity states of people, non-intrusively and without contact direct. This device is based on the detection of movements and vibrations using the measurement of the changes in the Speckle pattern produced in a multi-mode optical fiber when coherent light is driven inside. The device is capable, not only to perform the differential calculation of the frames, but also to process and make decisions according to the results obtained.
    The device comprises as a transduction element at least one multi-mode optical fiber, preferably plastic, attached to a coherent light source, such as a semiconductor laser, and an optical detector, such as a CCO camera. In a preferred embodiment, the multi-mode optical fiber is attached at one end to the coherent light source, and at its opposite end to the optical detector.
    The device further comprises a processing block connected to the optical detector, which implements the processing method necessary to issue a decision. The integration of the processing block into the device of the invention eliminates the need to use an additional computer, since the function of processing the data is performed from the device itself, which increases its applicability and reduces its cost. This processing block and the optical detector make up a detection and decision-making block.
    In addition, the device of the invention comprises connection elements, configured to connect / disconnect the multi-mode optical fiber to / from the coherent light source and the optical detector.
    A basic diagram of the fundamental elements comprising the device of the invention is shown in Figure 1: the optical detector 11 and the processing block 13 (which make up the detection and decision block 10), the coherent light source 12, multi-mode fiber optic 14 and connection elements 15.
    Optionally, all the elements of the device of the invention, except for the multi-mode optical fiber 14 - except its ends - and the connecting elements 15, are located inside a protective box, preferably of plastic, which offers protection, before blows and falls, to the set of elements that make up the device. This protection box has at least two holes, such that the coherent light source 12 is located close to one of the holes, and the optical detector 11 close to the remaining hole, allowing the connection of the multi-mode optical fiber 14 to said holes. elements.
    The device can be installed in any location where it is necessary to detect the physiological activity, such as a bed, a cradle, etc ... directly offering an alarm signal when the physiological inactivity is detected.
    Two possible configurations for the placement of the device of the invention are shown in Figure 2. In this case the multi-mode optical fiber 24 is located at a) the upper end of the mattress, at the height of the head of the person when it is lying down b) at the height of the coccyx. The detection and decision-making block (optical detector + processing block) and the coherent light source are located inside the protection box 29.
    In addition, the placement and geometry of the multi-mode fiber optic must be such that there are no parts suspended in the air, because this greatly increases the noise against the desired signal. In Figure 3 two possible embodiments of multi-mode fiber optic 34 geometry are shown: a) in a stretched round-trip shape and b) in the form of a ring.
    Optionally, the device of the invention further comprises an alarm system connected to the processing block, with its corresponding signaling elements (visual, acoustic ...) to alert of inactivity. One skilled in the art will understand that, during the operation of the device, and within the detection and decision block, the optical detector sends a signal to the processing block, such that it decides, thanks to the method of processing of the invention that later It is detailed, activate or not the alarm. Additionally, the device of the invention comprises a wireless block connected to the processing block, and configured to send the information generated and processed in the device, to a central computer. Preferably, the alarm system and the wireless block are inside the protection box.
    In a possible embodiment, the device of the invention further comprises a battery connected to the detection and decision block, the coherent light source, the wireless block and the alarm system, and is configured to provide electricity to said elements. In a preferred embodiment, the battery is inside the protective case.
    A more complete diagram of the device of the invention is shown in Figure 4, comprising: the optical detector 41 and the processing block 43 (which make up the detection and decision block 40), the coherent light source 42, multi-mode fiber optic 44, connection elements 45, wireless block 46, alarm system 47 and battery 48. All the elements of the device of the invention, except for the multi-mode fiber optic 44 - except its ends - and the connection elements 45, are located inside the protection box 49. The solid lines represent the electrical connections (battery 48-alarm system 47; battery 48-wireless system 46, battery 48-block detection and decision making 40 and battery 48-coherent light source 42), and dashed lines data connections (41-block optical detector processing 43; processing block 43-alarm system 47 and processing block 43-wireless system 46).
    The device will be in constant risk of being soiled by bodily or hospital fluids of all kinds. This fact, added to the demanding levels of hygiene in hospital settings, means that it must be washed with relative frequency. Therefore, and as mentioned above, the device of the invention has connecting elements of the multi-mode optical fiber to the input of the optical detector and the input of the coherent light source, such that said connection elements allow washing without damaging the electronic elements of the device. In addition, the connecting elements allow the multi-mode optical fiber of the optical detector to be disconnected from the coherent light source and reconnected, with sufficient precision so that the operation of the system is not disturbed. The incorporation of these connection elements supposes an important innovation with respect to the aforementioned systems, because the range of applicability of this system in real situations is significantly increased.
    A possible embodiment of the connecting elements for connecting the multi-mode optical fiber to both the coherent light source and the optical detector is shown in Figure 5. Preferably, the connecting elements 55 are formed by a source-detector part 55a, a fiber part 55b and optionally a fixing part 55c. The figure also shows the two halves of the protection box 59, in which the electronic components of the system are protected (coherent light source, optical detector, processing block ...).
    The source-detector part 55a comprises two hollow channels. One of the channels is in direct contact with the coherent light source at one of its ends, and the remaining channel, at its end closest to the coherent light source, is in direct contact with the optical detector. Therefore, the source-detector part 55a is fixed to both the coherent light source and the optical detector.
    The fiber part 55b in turn comprises two hollow channels with an inner diameter substantially greater than the diameter of the muIti-mode optical fiber, while the outer diameter is substantially smaller than the inner diameter of the channels of the source-detector part 55a. The ends of the multi-mode optical fiber are located inside the channels of the fiber part 55b, such that each end is placed in a different channel and such that during its placement they are introduced by two adjacent holes. The rest of the multi-mode optical fiber remains outside the 55b fiber part. The multi-mode optical fiber is fixed to the 55b fiber part permanently in the assembly.
    During operation of the device for connecting the muIti-mode optical fiber to the coherent light source and the optical detector, the channels of the fiber part 55b are located inside the channels of the source-detector part 55a, in such a way that the ends of the multi-mode optical fiber come into contact with the optical detector and the coherent light source.
    Furthermore, preferably the fiber part 55b has a plurality of external micro grooves, and the source-detector part 55a has a plurality of internal micro grooves. These micro slots have a double function. On the one hand they allow both pieces 55a, 55b to form a maximum stop between them, which prevents their sliding beyond that point. It is precisely the top point at which this sliding blockage occurs, the point at which the placement of the multi-mode optical fiber with respect to the optical detector and the coherent light source is optimal. On the other hand, these micro grooves, exert a certain resistance to movement preventing the sliding of one piece over the other by knocks and rubs while, however, allow them to be manually connected and disconnected from each other.
    Optionally, the connecting elements comprise a fixing part 55c that is attached to the fiber part 55b, at its opposite end where the coherent light source and the optical detector are located, and which allows the whole assembly to be fixed to a support optional (such as cloth) that simplifies installation.
    In this way, the light coming from the coherent light source is propagated through the multi-mode optical fiber until the optical detector is reached, which must be located in the center of the corresponding channel section, so that only receive the light coming from inside the channel and such that the light outside the channel does not reach the optical detector. A Speckle pattern is formed at the output of the multimode optical fiber, being detected by the optical detector. The raw signals of the optical detector are analyzed directly in the processing block that emits a binary signal (activity / inactivity) with the current state. A statistical decision system is implemented on the processing block that provides the ability to pass from raw signals to activity / inactivity data.
    The following describes the method of processing of the invention, which is divided into three phases: two initials that adapt the signal (differential analysis and signal conditioning) and a final decision phase. This processing method is configured to allow the detection of the states of activity or physiological inactivity of people, avoiding the need for a clinical eye for this purpose.
    Description differential frame analysis
    s 10 l S The first step towards detecting the physiological activity / inactivity is to quantify the variations suffered by the multi-mode optical fiber. For this, the method of the invention performs a differential analysis of the pixels is of a frame of the Speckle pattern versus the equivalent pixels of the immediately previous frame in time, such that the difference in the intensity change in absolute value is obtained in absolute value Each pixel of the frame. Next, the sum of all these partial variations of the intensity of the Speckle pattern zone is performed, and the variation value of the Speckle pattern is obtained. As a result of this differential analysis of frames over time, a sequence is obtained that summarizes the variations of the multimode optical fiber and that are detected by the optical detector at each moment (differential sequence of variations) and whose data are the variations of the Speckle pattern.
    twenty Figure 6 shows a sequence of 4 different frames of the same sequence and consecutive in time, of a capture of the Speckle pattern obtained by a CCO camera when the coherent light beam output that has been introduced has been projected on it. on the other end of a multimode optical fiber.
    Description signal conditioning method
    2S Preferably, the method of the invention performs the signal processing of the differential sequence of variations, in order to reduce the existing noise. Figure 6 shows two images of the differential sequence of variations, a) before and b) after signal processing.
    30 Next, the method of the invention analyzes the differential sequence of variations in at least one time window that travels over time, and which has
    its origin in the current moment and its end a few units of time before, so that each value of the differential sequence of variations is understood by each time window in different moments of time as it moves. That is, each time window, at each given time, comprises a fragment of the differential sequence of variations.
    From each time window, a detection threshold value and / or a temporary activity index can be obtained at each given time. The temporary activity index will be computed to detect the activity / inactivity states, so it can be defined as the summary of the grade; of activity detected in a certain time window. For the calculation of these temporary activity indices, a method is established for an automatic adjustment of the activity / inactivity detection thresholds based on the dynamic range, which eliminates the need to perform initial system calibrations for each person based on their physiological characteristics or their duck lodges.
    First, when an individual begins to be monitored (beginning of the differential sequence of variations), and because there are no defined detection thresholds, the method of the invention establishes an initial detection threshold value for the differential sequence. of variations, below which it is considered that there is no activity. This initial detection threshold value corresponds to the total intensity value that reaches the optical detector, due to the light that propagates through the multi-mode optical fiber.
    To do this, preferably, the coherent light source is turned on and off to obtain maximum (when the coherent light source is on) and minimum (when the coherent light source is off) values of total intensity received in the initial set of frames. The minimum intensity value is subtracted from the minimum intensity value, obtaining the total intensity value that reaches the optical detector through the fiber. Preferably, this value is scaled according to coefficients dependent on the type of multi-mode optical fiber used. That is, the initial threshold is proportional to the difference between the maximum and minimum level of intensity scaled by a coefficient dependent on the type of multi-mode optical fiber.
    Next, and once the time elapsed in the differential sequence of variations is greater than the time analyzed in the window or the temporary windows, the detection threshold and / or the temporal activity index is calculated.
    To do this, first, the method of the invention calculates whether the differential sequence of variations comprised in each time window intended for the calculation of the detection threshold at the time of study time (origin at the present time) is periodic. Said periodic pattern in the differential sequence of variations indicates that some of the patient's constants (heart rate, breathing ...) are being detected. If so, the detection threshold is calculated and subsequently, and regardless of whether or not there is periodicity, the index of temporary activity is obtained.
    The detection threshold in a time window is obtained from the maximum and minimum values of the differential sequence of variations comprised in said window, and preferably from the median of the maximum values and the median of the minimum values, such that in a possible embodiment the detection threshold is equal to the average value of the median of the maximum values and the median of the minimum values of the differential sequence of variations.
    In a possible embodiment there is a single time window, such that a detection threshold value is obtained at each time (in the case that the differential sequence of variations comprised by the time window at that time is periodic) and an index of temporary activity from the last detection threshold obtained. In the event that said differential sequence fragment of variations is periodic, the temporal activity index is calculated from the detection threshold previously calculated for said window. In the event that the sequence was not periodic, the temporal activity index is calculated from the last detection threshold obtained. Calculating the relationship between the number of times that the differential sequence of variations exceeds the detection threshold with the times that it does not exceed it, a value is established that, scaled with the duration of the window, produces a temporary activity index.
    For example, imagine at a given time, a time window of 10 seconds duration that comprises 300 values of the differential sequence of variations (30 frames per second). First, the method of the invention checks whether the signal is periodic, for example by Fourier transform. When periodic signal of the heart rate is detected in the differential sequence of variations, the method obtains the detection threshold which, according to a possible embodiment, is calculated from the median of the maximum values and the median of the minimum values of the sequence differential of variations. Next, the method of the invention, regardless of whether or not there is periodicity in the sequence, calculates the index of temporal activity from the last detection threshold calculated.
    In another possible embodiment, there are two temporary windows of different duration, such that the origin of both windows is the same, and such that a temporary activity index is obtained at each time of determined time from the window of shorter duration, and a detection threshold from the longest window.
    In another possible embodiment, from a single time window, at least two indices of temporal activity and / or at least two detection thresholds are obtained at each given time.
    The great novelty in this method of processing is that the detection threshold is modified automatically when a periodic pattern of the differential sequence of variations is detected in the time window, allowing an efficient activity / inactivity detection to the different conditions of a patient.
    This pattern must be periodic since the slightest vital signs, such as pulsations, are. If, on the contrary, the differential sequence of variations comprising the time window at the time of the study is non-periodic, it means that the patient has moved, for example due to convulsions,
    5 posture changes or muscle spasms.
    For example, suppose there is a single time window that moves over time, and that the differential sequence of variations is periodic. In this case, the method of the invention calculates the detection threshold and the temporal activity index. These values, for the same patient, are very similar between consecutive moments to analyze. Imagine that at a given time, the monitored patient has a seizure. In this case, the differential sequence of variations will no longer be periodic, and their values will be greater. The time window, when moving and understanding these values, does not calculate the threshold, since the signal is not periodic; nevertheless calculates the index of temporary activity from the last detection threshold calculated. Because the movements generate high intensity disturbances in the Speckle pattern, the temporal activity index values will be found in most cases above the last detection threshold value calculated (since the movement is very wide, possibly the 100% 20 of the temporary activity indices exceed this detection threshold value). Once the alteration states are finished, and periodic signals are returned, such as those that cause heartbeats, the method of the invention will detect the new maximum and minimum values, and the thresholds will be calculated automatically.
    That is, the fact that the method of the invention calculates the detection threshold in the time window only in the case that the differential sequence of variations comprised in said window is periodic, prevents new maximum and minimum values from being computed, and therefore obtaining a very high detection threshold, which would imply that on returning to normal (end of
    30 movement) the temporary activity indices were below the thresholds,
    and therefore the method mistakenly considers that there is inactivity.
    This calculation of the detection threshold only by detecting periodicity, is one of the great advantages of the method compared to the existing ones, since the movement of the patient does not disturb the detection of activity or inactivity.
    5 Figure 7 shows a periodic fragment of the differential sequence of variations comprised of two temporary windows: a window of longer duration for the calculation of the detection threshold and a window of shorter duration for the calculation of the indexes of temporal activity. Figure 8 shows another fragment in which the final part shows a non-periodic behavior of the differential sequence
    1 0 of variations comprised of two temporary windows: a window of greater duration destined to the calculation of the threshold of detection and a window of smaller duration destined to the calculation of the indices of temporal activity.
    Figure 7 shows the maximum values and the minimum values from the
    Which method of the invention calculates the detection threshold. In Figure 8, since there is movement, and therefore not being a periodic signal, the method of the invention does not calculate said threshold.
    The differential sequence of variations is a signal consequence of variations
    20 physiological patient, noise and fiber movement; and thanks to the indexes of temporal activity it is possible to determine the existence of physiological activity (heartbeat, breathing, movement of the patient ...).
    By means of temporary activity indices, possible derived problems are eliminated
    25 of the confusion of vital signs and movement. Even people with lower intensity pulsations or breathing in terms of vibrations are efficiently monitored as people who have physiological activity, since the detection thresholds have been dynamically adjusted to their constants. On the other hand people with more athletic physicists that generate
    30 pulses of greater intensity are also monitored efficiently, clearly differentiating states of activity and inactivity, regardless of
    that these pulsations can generate changes in the Speckle pattern of the same
    magnitude than those produced by movement.
    As a result of the conditioning phase, a series of temporal activity indices are obtained that summarize the degree of activity detected in the differential sequence in the at least one time window used, and that moves over time. These indices are used, as explained below, to decide whether the inactivity alarm is issued.
    Description decision method
    Parallel to obtaining the indexes of temporal activity, the method of the invention combines them following statistical models to offer a judgment, determining whether or not activity has been detected. This judgment is the last information provided by the system, which can be communicated by different means.
    In a possible embodiment, to decide if the signal is activated, the method of the invention computes the last values obtained from activity indices (for example: the last 150 values corresponding to the last 5 seconds). If the method detects a decrease in progressive activity, followed by a stabilization at very low activity values, close to 0, the inactivity alarm is activated.
    In another possible embodiment, to decide whether the inactivity alanna is activated, Gaussian filtering is performed on the activity indices obtained, using two windows of different duration, for example, a window of duration 1 second, and a window of duration 10 seconds. One skilled in the art will understand that these windows for the analysis of the index of temporal activity are different from the temporary windows for the calculation of the detection thresholds.
    Starting from the window with the longest duration, its statistics are dynamically calculated (eg mean, standard deviation ...), defining a decision point that stops the trend of the detected signal (temporal activity index signal). Based on the statistics obtained, a filter is defined that models the trend of the activity index associated with the longer duration window. This probability is threshold raised to decide if the inactivity alarm is activated. The determination of these thresholds,
    5 is extracted from the probabilistic distributions of the temporal activity indices associated with each window, which are measured during filtering.
    By applying the shorter window to this filter, instantaneous anomalies can be detected that enable the inactivity alarm to be activated. Other options for deciding 1 0 the result of the decision may be higher level approximations such as machine learning techniques.
    As a final result, the device's processing block issues a high-level judgment summarized in "activity / inactivity"; every certain period of time (eg 10
    15 seconds), so that the need to have a clinical eye to analyze the raw data is avoided. Preferably, a wireless block sends the information generated and processed in the device to a central computer. In addition, preferably, an alert system emits a light and / or acoustic signal in the case where physiological inactivity is detected.
    20 As explained above, the use of Speckle pattern analysis in multi-mode fiber optics to monitor vital signs becomes very complex when the person is awake and in a conscious state since, usually people tend to perform voluntary movements or involuntary so
    25 usual in these conditions. In this sense, the invention presented here avoids this problem by agglutinating all these phenomena (movements, vital signs ...) as a physiological activity, issuing a directly applicable decision, and avoiding the need for an expert eye.
    30 This decision alerts when the monitored person has entered cardiorespiratory arrest. This solution is highly effective and allows monitoring
    Complete 24 hours a day without periods of unknown information as occurs in the
    state of the art devices. This is a very remarkable advantage over the
    traditional vital signs monitors since regardless of the state of
    the person: conscious or unconscious, regardless of the movements that
    s perform and regardless of whether the person makes movements very
    usual: (people with Parldnson's disease) or never perform them: (people
    in comatose states); the device of the invention is always able to distinguish
    when the person maintains his vital functions and when he has lost them (symptom of
    cardiorespiratory arrest).
    10
    The main improvements of the device of the invention are related to the taking
    of decision and with the methods of automatic calibration, that eliminate the intervention
    of any operator for the correct operation of the device. Both improvements
    are strictly related to the processing block incorporated into the system
    lS detection.
    In addition, the device of the invention eliminates the need for proper placement.
    of the multi-mode fiber optic depending on the physical characteristics (height and weight
    mainly) of the person you are monitoring. This is because a
    twenty Incorrect placement of the multi-mode fiber optic can cause a very bad
    signal quality, incorporating a large amount of noise into the measurements, which can
    cause incorrect measures or periods of time in which there is no information.
    The dynamic adjustment of the basic activity thresholds reduces the dependence of a
    Good placement to practically zero. Greatly increasing your chances of
    2S application in real cases in which the variability of situations is very high.  
    权利要求:
    Claims (20)
    [1]
    l. Device for detecting physiological activity / inactivity configured to allow the detection of the states of activity or physiological inactivity of people, in a non-intrusive way and without direct contact, based on the detection of movements and vibrations using the measurement of changes in the pattern SpeckIe produced in a multi-mode optical fiber (14, 24, 34, 44) when coherent light is driven inside, characterized in that it comprises:
    - at least one multi-mode optical fiber (14, 24, 34, 44) as a transduction element, connected to a coherent light source (12, 42) and an optical detector (11, 41);
    - a processing block (13, 43) connected to the optical detector (11, 41), configured to implement the processing method necessary to issue a decision, thus eliminating the need to use an additional computer, such that said processing block (13 , 43) and the optical detector (11, 41) form a detection and decision-making block (10, 40);
    - connection elements (15, 45, 55) configured to connect / disconnect the multi-mode optical fiber (14, 24, 34, 44) to / from the optical detector (11, 41) and to the coherent light source ( 12, 42), with sufficient precision so that the operation of the system is not disturbed;
    the device being configured to, during operation, propagate the light from the coherent light source (12, 42) through the multimode optical fiber (14, 24, 34, 44) until the optical detector (11, 41 ), such that said optical detector (11, 41) detects the Speckle pattern formed at the output of the multi-mode optical fiber (14, 24, 34, 44), and such that the processing block (13, 43) analyzes the raw signals of the optical detector (11, 41), emitting activity / inactivity data with the current state.
    [2]
    2. The device of the preceding claim, further comprising an alarm system (47) connected to the processing block (13, 43), with its corresponding signaling elements to alert of inactivity.
    [3]
    3. The device of any of the preceding claims, further comprising a wireless block (46) connected to the processing block (13, 43), configured to send the information generated and processed in the device to a central computer.
    [4]
    Four. The device of any of the preceding claims, further comprising a battery (48) connected to all the elements of the device that need power.
    [5]
    5. The device of any of the preceding claims, wherein all the elements of the device of the invention, except for the multi-mode optical fiber (14, 24, 34, 44) - except for its ends - and the connecting elements (15, 45, 55), they are placed inside a protective box (49, 59) that offers care, in the event of blows and falls, to the set of elements that make up the device.
    [6]
    6. The device of the preceding claim, wherein the protection box (49, 59) is made of plastic and has at least two holes close and aligned with the coherent light source (12, 42) and the optical detector (11, 41), thus allowing the connection of the multi-mode optical fiber (14, 24, 34, 44) to the coherent light source (12, 42) and to the optical detector (11, 41).
    [7]
    7. The device of any of the preceding claims, wherein the connection elements (15, 45, 55) are formed by a source-detector part (55a), fixed both to the coherent light source (12, 42) and to the optical detector (11, 41), and by a fiber part (55b), fixed to the multi-mode optical fiber (14, 24, 34, 44), such that the source-detector part (55a) comprises two hollow channels, one being of the channels in direct contact with the coherent light source (1 2, 42) at one of its ends, and the
    remaining channel, at its end closest to the coherent light source (12, 42), in direct contact with the optical detector (11, 41), and such that the fiber part (55b) comprises two hollow channels with an inner diameter substantially greater than the diameter of the multi-mode optical fiber (14, 24, 34, 44), and with an outside diameter substantially less than the inside diameter of the channels of the source-detector part (55a), the ends of the optical fiber being multi-mode (14, 24, 34, 44) inside the channels of the fiber part (55b), so that each end is placed in a different channel and such that during its placement they are introduced through two adjacent holes, the rest of the multi-mode optical fiber (14, 24, 34,44) remaining outside the fiber part (55b), so that during operation of the device, to connect the multi-mode optical fiber (14, 24, 34, 44) to the coherent light source (12, 42) and to the optical detector (11, 41), the channels of the fiber part (55b) are located inside the channels of the source-detector part (55a), such that the ends of the multi-mode optical fiber (14, 24, 34, 44) come into contact with the optical detector (11, 41) and the coherent light source (12, 42).
    [8]
    8. The device of the preceding claim, wherein the fiber part (55b) has a plurality of external micro grooves, and the source-detector part (55a) has a plurality of internal micro grooves, configured to allow both parts (55a, 55b) form a maximum stop between them and to exert resistance to the movement preventing the sliding of a piece on the other by blows and rubs.
    [9]
    9. The device claims 7 to 8, wherein the connecting elements (15, 45, 55) are further formed by a fixing piece (55c) located attached to the fiber part (55b), at its opposite end where the coherent light source (12, 42) and the optical detector (11, 41), which allows the whole assembly to be fixed to an optional support that simplifies its installation.
    [10]
    10. The device of any one of the preceding claims, wherein there is a single multi-mode optical fiber (14, 24, 34, 44) of plastic, attached at one end to the
    coherent light source (12, 42), and at its opposite end to the optical detector (11, 41).
    [11 ]
    eleven . The device of any of the preceding claims, wherein the coherent light source (12, 42) is a semiconductor laser.
    [12]
    12. The device of any of the preceding claims, wherein the optical detector (11, 41) is a CCD camera.
    [13]
    13. Processing method configured to detect the states of physiological activity or inactivity of people, using the device according to any of the preceding claims, characterized in that it comprises the steps of:
    - for each frame of the Speckle pattern obtained in the optical detector (11, 41), perform a differential analysis of the pixels of said frame against the equivalent pixels of the immediately previous frame in time, such that the difference is obtained in absolute value of the intensity change in each pixel of the frame;
    - perform, for each frame analyzed, the sum of all values differs from the intensity change thus obtaining its value of variation of the Speckle pattern, so that as time progresses and each frame received in the optical detector is analyzed (11, 41) a new value is obtained from a differential sequence of variations that summarizes the variations of the multi-mode optical fiber (14,24,34, 44) over time, and which are detected by the optical detector (11, 41) at every moment;
    - when an individual begins to be monitored and threshold detection values have not yet been calculated, establish an initial detection threshold value for the differential sequence of variations, below which it is considered that there is no activity, such that said threshold value of Initial detection corresponds to the total intensity value that reaches the optical detector (11, 41), due to the light that propagates through the multi-mode optical fiber (14, 24, 34, 44).
    - analyze the differential sequence of variations in at least one time window that moves over time - at least one time window for the calculation of detection thresholds and at least one time window for the calculation of indices of temporary activity - as long as the size of each time window is less than the time elapsed in the differential sequence of variations, such that each time window has its origin at the current time and its end a few units of time before, and such that each value of the differential sequence of variations it is understood by each time window in different instants of time as it moves, so that each time window, at each instant of time determined, comprises a fragment of the differential sequence of variations;
    - determine whether the fragment of the differential sequence of variations comprised in each time window intended for the calculation of thresholds is periodic;
    - if the sequence fragment is periodic determine, in each time window destined to the calculation of thresholds, the detection threshold from the maximum and minimum values of the sequence fragment;
    - determine the index of temporal activity of the differential sequence fragment of variations comprised in each temporary window destined to the calculation of indices, such that said Index is obtained from the last detection threshold obtained, calculating, in each temporal window destined to the calculation of indices , the relationship between the number of times that the fragment of the differential sequence of variations exceeds the detection threshold with the times that it does not exceed it and scaling said value with the duration of the index window;
    - combine the obtained temporary activity indices, following statistical models to offer a judgment, thus determining whether or not activity has been detected and if the inactivity alarm is activated.
    [14]
    14. The method of the preceding claim further comprising the step of:
    - Perform signal processing of the differential sequence of variations, in order to reduce the existing noise.
    [15]
    fifteen. The method of any one of claims 13 to 14, wherein for obtaining the initial detection threshold the coherent light source (12, 42) is turned on and off to obtain maximum values - when the coherent light source (12, 42 ) is on - and minimum - when the coherent light source (12, 42) is off of total intensity received in the initial set of frames, so that the minimum intensity value is subtracted from the minimum intensity value, thus obtaining the total intensity value that reaches the optical detector (11, 41) through the multi-mode optical fiber (14, 24, 34, 44), said value being scaled according to coefficients dependent on the type of multi-optical fiber mode (14, 24, 34, 44) used.
    [16]
    16. The method of any one of claims 13 to 15, wherein the detection threshold is equal to the average value of the median of the maximum values and the median of the minimum values of the fragment of the differential sequence of variations.
    [17]
    17. The method of any one of claims 13 to 16, wherein there is a single time window for the calculation of both parameters - thresholds and indices - such that for each fragment of the differential sequence of variations a threshold detection value is obtained - in the if said fragment is periodic - and a temporary activity index from the last detection threshold obtained.
    [18]
    18. The method of any one of claims 13 to 16, wherein there are two temporary windows of different duration, such that in each instant of time determined a temporary activity index is obtained from the window of shorter duration, and a detection threshold at from the window of longer duration.
    [19]
    19. The method of any one of claims 13 to 16, wherein at least two indices of temporal activity and / or at least two detection thresholds are obtained from each single time window determined.
    The method of any of claims 13 to 19, wherein to decide if the inactivity alarm is activated, the last 150 values obtained from activity indices are computed, such that if a decrease in progressive activity is detected, followed by a stabilization in activity values close to O, the inactivity alarm is activated.
    [21]
    21. The method of any one of claims 13 to 19, wherein to decide whether the inactivity alarm is activated, a Gaussian filtering is performed on the Activity Indices obtained, using two windows of different duration, such that starting from the Longer window, its statistics are dynamically calculated 15 determining a decision point that determines the trend of the detected signal, and based on the statistics obtained, a filter is defined that models the trend of the activity index associated with the window of longer duration, threshold raising said probability to decide if the inactivity alarm is activated, so that by applying the shorter window to this filter, anomalies can be detected
    20 snapshots that enable the inactivity alarm to be activated.
    FIGURE 1
    FIGURE 2
    FIGURE 3
    one
    FIGURE 4
    FIGURE 5
    to)
    12000 10000 8000 100 105 110 115 12000 10000 8000
    FIGURE 6
    100 110 120 130 140 150 160 170
    FIGURE 7
    12000 9000 6000 3000
    400 410 420 430 440 450 460 470 480 490 500 510
    FIGURE 8
    类似技术:
    公开号 | 公开日 | 专利标题
    Presti et al.2017|Smart textile based on 12 fiber Bragg gratings array for vital signs monitoring
    JP6697985B2|2020-05-27|Biometric information output device
    ES2328205B1|2010-08-30|METHOD FOR OBTAINING CARDIAC FREQUENCY AND RESPIRATORY FREQUENCY IN A PESA-PEOPLE ELECTRONIC SCALE.
    JP6149515B2|2017-06-21|Detection method, detection device, and detection program
    JP6706616B2|2020-06-10|Method and device for measuring intracranial pressure ICP in a subject
    US10110859B2|2018-10-23|System, method and device for monitoring light and sound impact on a person
    CN105491942A|2016-04-13|Monitoring system and method for monitoring the hemodynamic status of a subject
    JP6053804B2|2016-12-27|Device and method for monitoring human movement and orientation, and processor and processing method for use in the device
    KR100847137B1|2008-07-18|Portable Apparatus of Physiology Signal Measurement
    Digarse et al.2017|Arduino UNO and GSM based wireless health monitoring system for patients
    US10856810B2|2020-12-08|Smart carpet systems and methods of using same for monitoring physical and physiological activities
    EP3384827A1|2018-10-10|System for determining a physiological parameter of a subject
    JP2014171574A|2014-09-22|Device, system and method each for monitoring respiration
    Kam et al.2017|Compact and low-cost optical fiber respiratory monitoring sensor based on intensity interrogation
    CN108042108B|2020-12-08|Sleep quality monitoring method and system based on body vibration signals
    JP2015533549A|2015-11-26|Monitoring device for analyzing sleep state
    ES2610181B2|2018-03-27|Fiber optic based physiological activity / inactivity detection device
    TWM521450U|2016-05-11|Device of sleep detection
    KR100767642B1|2007-10-17|Chill and hot judge ment apparatus
    WO2017038966A1|2017-03-09|Bio-information output device, bio-information output method and program
    Motoi et al.2003|Evaluation of a new sensor system for ambulatory monitoring of human posture and walking speed using accelerometers and gyroscope
    KR101951815B1|2019-02-26|Ear wearing type health care monitoring system
    US20180235473A1|2018-08-23|System, method and processor for monitoring a vital sign of a subject
    CN109091127A|2018-12-28|For monitoring the method and its equipment of blood pressure
    Nakasho et al.2016|Implementation of a vital signs monitoring system in combination with a bed-leaving detection system
    同族专利:
    公开号 | 公开日
    ES2610181B2|2018-03-27|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    US5212379A|1991-12-06|1993-05-18|Alamed Corporation|Fiber optical monitor for detecting motion based on changes in speckle patterns|
    WO2004046869A2|2002-11-18|2004-06-03|William B Spillman, Jr|System, device, and method for detecting perturbations|
    WO2008011058A2|2006-07-19|2008-01-24|Fiber Sensys Llc.|Fiber-optic mat sensor|
    JP2008309497A|2007-06-12|2008-12-25|Oputeikon:Kk|Fiber optic sensor|
    US20140355001A1|2013-05-28|2014-12-04|Stratus Devices, Inc.|Measuring Deflection in an Optical Fiber Sensor by Comparing Current and Baseline Frames of Speckle Interference Patterns|
    US20160066790A1|2014-09-10|2016-03-10|Samsung Electronics Co., Ltd.|Laser speckle interferometric system and method for mobile devices|
    法律状态:
    2018-03-27| FG2A| Definitive protection|Ref document number: 2610181 Country of ref document: ES Kind code of ref document: B2 Effective date: 20180327 |
    优先权:
    申请号 | 申请日 | 专利标题
    ES201600626A|ES2610181B2|2016-07-22|2016-07-22|Fiber optic based physiological activity / inactivity detection device|ES201600626A| ES2610181B2|2016-07-22|2016-07-22|Fiber optic based physiological activity / inactivity detection device|
    [返回顶部]