WO2014024104A1 - Dispositif et procédé d'extraction d'informations physiologiques - Google Patents

Dispositif et procédé d'extraction d'informations physiologiques Download PDF

Info

Publication number
WO2014024104A1
WO2014024104A1 PCT/IB2013/056324 IB2013056324W WO2014024104A1 WO 2014024104 A1 WO2014024104 A1 WO 2014024104A1 IB 2013056324 W IB2013056324 W IB 2013056324W WO 2014024104 A1 WO2014024104 A1 WO 2014024104A1
Authority
WO
WIPO (PCT)
Prior art keywords
signal
sub
characteristic
interest
signals
Prior art date
Application number
PCT/IB2013/056324
Other languages
English (en)
Inventor
Gerard De Haan
Original Assignee
Koninklijke Philips N.V.
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Koninklijke Philips N.V. filed Critical Koninklijke Philips N.V.
Publication of WO2014024104A1 publication Critical patent/WO2014024104A1/fr

Links

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7203—Signal processing specially adapted for physiological signals or for diagnostic purposes for noise prevention, reduction or removal
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/0059—Measuring for diagnostic purposes; Identification of persons using light, e.g. diagnosis by transillumination, diascopy, fluorescence
    • A61B5/0077—Devices for viewing the surface of the body, e.g. camera, magnifying lens
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
    • A61B5/024—Detecting, measuring or recording pulse rate or heart rate
    • A61B5/02416—Detecting, measuring or recording pulse rate or heart rate using photoplethysmograph signals, e.g. generated by infrared radiation
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/08—Detecting, measuring or recording devices for evaluating the respiratory organs
    • A61B5/0816—Measuring devices for examining respiratory frequency
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0012—Biomedical image inspection
    • G06T7/0014—Biomedical image inspection using an image reference approach
    • G06T7/0016—Biomedical image inspection using an image reference approach involving temporal comparison
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B2576/00—Medical imaging apparatus involving image processing or analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10024—Color image
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30004—Biomedical image processing
    • G06T2207/30076—Plethysmography
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H30/00—ICT specially adapted for the handling or processing of medical images
    • G16H30/40—ICT specially adapted for the handling or processing of medical images for processing medical images, e.g. editing

Definitions

  • the present invention relates to a device and method for extracting physiological information from remotely detected electromagnetic radiation emitted or reflected by a subject, wherein the physiological information is embedded in a data stream comprising a sequence of signal samples representing a region of interest exhibiting a continuous or discrete characteristic signal including physiological information indicative of at least one at least partially periodic vital signal.
  • Photoplethysmography is a commonly known optical measurement approach which can be used to detect blood volume changes in the microvascular bed of tissue of a monitored subject.
  • Conventional PPG approaches include so-called contact PPG.
  • Contact PPG requires measurement components (e.g., light sources and photodetectors) which basically have to be attached to a subject's skin. Consequently, standard photoplethysmography comprises obtrusive measurements, e.g. via a transceiver unit being fixed to the subject's earlobe or fingertip. Therefore, remote PPG measurement is often experienced as being unpleasant.
  • a standard (or: contact) PPG device includes artificial light sources to be directly attached to an indicative surface, e.g., a skin portion, of the subject to be observed.
  • the receiver or detector e.g. at least one photodiode
  • the receiver or detector is closely fixed to the subject's skin patch of interest.
  • signal quality can be deteriorated as well, e.g. due to undesired tissue compression.
  • remote photoplethysmography utilizes light sources or, in general, radiation sources, disposed remote from the subject of interest.
  • light sources or, in general, radiation sources disposed remote from the subject of interest.
  • ambient light sources rather than defined special-purpose light sources are utilized.
  • artificial light sources and/or natural light sources can be exploited. Consequently, in remote PPG environments, it has to be expected that due to widely changing illumination conditions, the detected signals generally provide a very small signal-to-noise ratio.
  • a detector e.g., a camera, can be disposed remote from the subject of interest for remote PPG measurements.
  • remote photoplethysmographic systems and devices are considered unobtrusive and can be adapted and well suited for everyday applications.
  • the field of application may comprise unobtrusive inpatient and outpatient monitoring and even leisure and fitness applications.
  • remote photoplethysmography is far more susceptible to distortion and noise.
  • Undesired subject motion with respect to the detector and/or the radiation source can excessively influence signal detection.
  • remote photoplethysmographic devices are frequently subjected to varying overall illumination conditions. Therefore, it has to be expected that the detected signals are almost always corrupted by noise and distortion.
  • specular reflections in the region of interest comprising at least a portion of the subject's skin tissue.
  • specular reflection is considered a "mirror-like" reflection of incident radiation at a surface. Specular reflections may also occur at the skin surface of a living being. This applies in particular to greasy skin portions and, generally, to subjects having considerably dark skin (high content of melanin). Since skin portions which are subjected to specular reflections basically mirror incident radiation at the skin's surface, the radiation is prevented from penetrating the skin tissue and, consequently, from being reflected or absorbed by blood vessels and surrounding tissue portions. Therefore, specularly reflected radiation is considered non-indicative of the desired vital signals.
  • remote PPG is still considered to pose major challenges to signal detection and signal processing.
  • the recorded data such as captured reflected or emitted electromagnetic radiation (e.g., recorded image frames)
  • the desired signal to be extracted therefrom always comprises, besides the desired signal to be extracted therefrom, further signal components deriving from overall disturbances, for instance noise due to changing luminance conditions (including specular reflections) and relative movement between the observed subject and the detection sensor, a detailed precise extraction of the desired signals is still considered to pose major problems for existing detection approaches and processing algorithms.
  • the region of interest which has been selected for collecting indicative data to be processed during signal extraction is subjected to locally occurring disturbances.
  • the subject's face can be selected to form the region of interest.
  • a rectangular box more or less fitting the face can be chosen, while the remaining portion of the initial image frame can be disregarded during further processing.
  • the region of interest can be cropped or clipped.
  • the region of interest can be tracked over time over a sequence of image frames. Basically, tracking may comprise recurring pattern detection (e.g., face detection) or motion
  • the region of interest can be kept “stable", at least to a certain extent. Tracking the region of interest via pattern detection, positional shift estimation or a suitable combined algorithm is considered to require huge processing power or, so to say, computational costs. However, varying disturbances within the region of interest cannot be addressed in this way.
  • a device for extracting physiological information from remotely detected electromagnetic radiation emitted or reflected by a subject comprising:
  • an interface for receiving a data stream derived from detected electromagnetic radiation, the data stream comprising a sequence of signal samples representing a region of interest exhibiting a continuous or discrete characteristic signal including physiological information indicative of at least one at least partially periodic vital signal,
  • a partitioning unit configured for selectively partitioning the sequence of signal samples into at least two distinct sub-sequences of defined signal subsets representing spatial sub-regions of the region of interest
  • a signal extraction unit for processing each of a plurality of the at least two distinct sub-sequences and thereby extracting a plurality of distinct indicative characteristic sub-signals, wherein each of the at least two distinct indicative characteristic sub-signals is extracted from a respective distinct sub-sequence, and
  • a data combining unit configured for generating an enhanced characteristic signal by combining the extracted characteristic sub-signals.
  • the present invention is based on the idea that spatially dividing or splitting the sequence representing the region of interest into at least two sub-sequences representing respective portions, portion-by-portion processing the portions, so as to extract sub-signals indicative of the desired vital signal and, eventually, combining the respective sub-signals into a resulting enhanced signal, can significantly improve the signal-to-noise ratio in the resulting extracted signals. This can even be the case under considerably adverse situations suffering from local or regional disturbances in the region of interest which even may drift or move within the region of interest over time.
  • the signal subsets of which the sub-sequences are composed may be configured in an overlapping, disjoined or adjoined pattern.
  • the at least two sub-sequences may represent partially overlapping, adjoined, or disjoined portions of the original signal samples.
  • the at least two sub-sequences may comprise respective subsets which are different in size, shape or orientation. However, preferably the subsets within each of these sub-sequences are equal in size, shape or orientation.
  • the device of the invention provides for a beneficial solution for the above challenges, providing both manageable computational costs and considerable improvement of the signal-to-noise ratio.
  • the extracted characteristic sub-signals can be distinctly treated in the combining unit.
  • the extracted characteristic sub- signals can be weighted or biased when generating the enhanced characteristic signal.
  • greater prominence can be given to characteristic sub-signals which are considered to comprise a satisfying or even better signal-to-noise ratio ("plain” or relatively clean sub- signals).
  • characteristic sub-signals which are considered to show a poor signal- to-noise ratio can be attenuated or even disregarded.
  • the region of interest By partitioning the region of interest into a plurality of subsets, it becomes more likely that disturbance- affected portions of the region of interest are clearly present in one or more of the signal subsets (distorted subsets), while the remaining subsets are far less affected by these disturbances (relatively clean signal subsets). Consequently, during further processing, relatively clean signal subsets can be enhanced, while distorted signal subsets can be attenuated or even skipped. It should be understood that in a given region of interest the at least two defined signal subsets the region of interest is partitioned into, may mutually cover an area which corresponds to the size of the region of interest.
  • the signal subsets may be arranged in an adjoining order, a spaced order, or even in a partially overlapping order.
  • determining the present state of each of the at least two defined signal subsets is not necessarily considered to be a rather black or white situation. There may be further increments or graduations related to a signal subset's quality or signal-to-noise ratio. Furthermore, it is worth noting that determining the actual state of the signal subsets does not necessarily require a distinct operation. By contrast, when combining the characteristic sub-signals, statistical algorithms may be applied, so as to allow for inline- weighting the sub-signals during combination.
  • the enhanced characteristic signal can be generated under consideration of a weighted average of the characteristic sub-signals, wherein the respective weighting factors are proportional to the quality or signal-to-noise ratio of each of the treated characteristic sub-signals.
  • the variance of each of the characteristic sub-signals is calculated and the resulting enhanced characteristic signal can be generated on the basis of a weighted average of the characteristic sub-signals, wherein the weighting factors may correspond to the inverse variance of the respective sub-signal:
  • Argument t denotes time or may also represent a present frame number in a sequence of frames.
  • HBo stands for the (overall) enhanced characteristic signal.
  • HBi denotes a respective characteristic sub-signal.
  • the actual weighting factor is denoted by w(i).
  • the addition parameter ⁇ (delta) may be considered a small bias which may be applied to even further attenuate distorted sub-signals by correcting extreme weights w(i) which are erroneously assigned to signal subsets which are heavily distorted and, therefore, not-indicative of the desired vital signals.
  • sub-signal combination may be performed under consideration of a median of the characteristic sub-signals (at the same time instant or frame):
  • HBo ⁇ t) median(HBi(t)) .
  • the data stream may comprise a sequence of frames or, more precisely, a series of image frames comprising spectral information. For instance, RGB-images comprising color information can be utilized. However, also frames representing infrared (IR) and red (R) information can form the sequence of frames.
  • the image frames can represent the observed subject and further elements.
  • the partitioning unit, the signal extraction unit and the data combining unit are commonly embodied by a processing unit which is driven (or: controlled) by respective logic commands.
  • a processing unit may also comprise suitable input and output interfaces.
  • each of the partitioning unit, the signal extraction unit and the data combining unit can be embodied by separate processing units, controlled or controllable by respective commands.
  • each respective processing unit can be adapted to its special purpose. Consequently, a distribution of tasks can be applied, wherein distinct tasks are processed (or: executed) on a single processor of a multi-processor processing unit, or wherein image processing related tasks are executed on an image processor while other operational tasks are executed on a central processing unit.
  • the partitioning unit is further configured for multiply partitioning the region of interest into at least two groups, each of which comprising at least two defined signal subsets represented by respective distinct sub- sequences, wherein the at least two groups comprise different partition patterns.
  • the region of interest may form a basis for different partition patterns which in turn allows for choosing a suitable group (or: pattern) for a given disturbance in the present region of interest.
  • the higher the number of subsets in a group the better a set of subsets may "match" a distortion-affected area or a highly indicative area.
  • the smaller the number of subsets in the group the broader is the basis for pixel agglomeration within the signal subsets which enables some spatial normalization which may basically result in more robust results, e.g. attributable to noise reduction through pixel averaging.
  • a first group may have just one subset which basically corresponds to the whole size of the region of interest. In this way, also a conventionally derived "sub-signal" can be considered during further signal processing.
  • the at least two groups can be different in size, number, orientation, arrangement or even shape of their respective subsets.
  • the region of interest can be multiply partitioned into four groups each of which having a size corresponding to the size of the region of interest.
  • a first group may have just one subset.
  • a second group may have two subsets each of which covering half the area of the subset of the first group.
  • the third group may have four subsets each of which covering half the area of a subset of the second group.
  • the fourth group may comprise eight subsets each of which covering half the area of a respective subset of the third group.
  • a selection of subsets for processing and combining can be made within each of the groups. Still, however, also an embodiment can be envisaged wherein a selection can be made across some of the at least two groups.
  • a subset of a second group can be combined with four subsets of a fourth group.
  • the combination may involve generating an enhanced characteristic signal by "merging" respective characteristic sub-signals. Therefore, data combining may involve sub-signals of one group or a plurality of groups.
  • each additional group of defined subsets exceeding a first group of defined signal subsets comprises a different number of defined signal subsets than a respective anterior group.
  • the device can be configured such that a further group can be added on demand, for instance, when data processing based on a given group or a given plurality of groups does not result in sufficient signal quality.
  • farther reaching segmentation or splitting of the region of interest can be an appropriate measure allowing for better "matching" distorted areas and high-indicative areas with a set of (considerably smaller) subsets.
  • anterior does not necessarily relate to a temporal order.
  • each additional group of subsets comprises a larger number of subsets, wherein each of the subsets covers a smaller portion than a respective subset of the anterior group.
  • each additional group of subsets exceeding a first group of subsets may comprise a smaller number of subsets, wherein each of the subsets covers a larger portion than a respective subset of the anterior group. It should be understood that within a group the signal subsets do not necessarily have the same size and shape. A group may comprise various different signal subsets.
  • the signal extraction unit is further configured for group-wise processing each of a plurality of the at least two distinct sub-sequences of a respective group, thereby extracting a plurality of distinct indicative characteristic sub-signals per group.
  • parallel processing (not necessarily temporally simultaneous processing, or concurrent processing) can be achieved.
  • characteristic sub-signal combination can be performed "across" more than one group.
  • the region of interest of a single frame is more than once utilized for partitioning, sub-signal derivation and the enhanced characteristic signal generation. So, basically the same input data can be "multiplied" for data enrichment measures. Still, given the different subset patterns in each of the groups, the resulting signal quality can be improved under consideration of a weighting algorithm for combining the sub-signals.
  • the data combining unit is further configured for group- wise generating an immediate characteristic signal by group-wise combining the extracted characteristic sub-signals for each group, and for generating the enhanced characteristic signal by combining the intermediate characteristic signals.
  • This embodiment is beneficial in that a two-stage combination of signals into the finally desired characteristic signal is achieved.
  • group-wise characteristic sub-signals for each of the subsets can be processed, as indicated above. Since more than one group is utilized, consequently several sets of sub-signals can be obtained.
  • sub-signal combination can be accomplished per group. In this way, for each group a respective intermediate characteristic signal can be derived.
  • the intermediate characteristic signals can be combined into the resulting enhanced characteristic signal.
  • appropriate weighting and/or signal enhancing or diminishing algorithms can be utilized. Consequently, the signal-to-noise ratio in the resulting characteristic signal can be further improved.
  • the signal extraction unit is further configured for processing an unpartitioned original signal set basically formed by the whole region of interest, thereby extracting another distinct indicative characteristic sub-signal.
  • the unpartitioned original signal set can be regarded as a group comprising a single subset.
  • the device further comprises a segmentation unit configured for determining an indicative frame section, in particular a skin portion of the subject, the indicative frame section comprising the region of interest.
  • the segmentation unit can be further configured for motion compensation, face recognition and/or feature recognition.
  • the device may further comprise a skin segmentation means for detecting the region of interest in the subject, preferably the skin segmentation means comprises a feature tracker for detecting at least one distinct skin portion, in particular a face pattern.
  • the at least one at least partially periodic vital signal is selected from the group consisting of heart rate, heart beat, respiration rate, heart rate variability, Traube-Hering- Mayer waves, and oxygen saturation.
  • the at least two distinct indicative characteristic sub-signals are associated with a signal space representative of characteristics of the electromagnetic radiation, wherein the signal space is preferably a color signal space comprising at least two complementary channels, wherein the at least two complementary channels are related to defined spectral portions, and wherein the at least two distinct indicative characteristic sub-signals are derived under consideration of at least two main components, each of which being related to a respective complementary channel.
  • the desired vital signal eventually can be calculated.
  • the at least two complementary channels may correspond to a red channel, a green channel and a blue channel, each of which covering a distinct spectral portion of visible light.
  • the desired vital signal extraction can make use of differences in radiation absorption (and, therefore, also in reflection) of the subject's blood vessels and the surrounding skin tissue. In the skin tissue, mainly melanin portions influence incident radiation.
  • Making use of at least two, preferably three complementary channels can allow for disturbance compensation. This may involve defining suitable coefficients of a linear combination of the at least two, preferably three main components of the characteristic sub-signals. It goes without saying that the above composition may also apply to the characteristic signal generated through combining the characteristic sub-signals.
  • the at least two complementary channels may comprise infrared channels, or a combination of visible light and infrared light channels. Also in this way, the desired vital signals can be detected.
  • the device further comprises a sensor means, in particular a camera, configured for remotely sensing electromagnetic radiation and capable of capturing the data stream, preferably the sensor means comprises a spectral sensitivity adapted to spectral characteristics of complementary channels forming a signal space.
  • RGB cameras or cameras configured for sensing red and infrared portions of electromagnetic radiation can be applied.
  • further additive or subtractive signal spaces (as well as their derivates, such as logRGB) can be envisaged.
  • CMYK color space can be used.
  • signal spaces can be converted into each other.
  • expanded signal spaces can be utilized, for instance a signal space comprising red, green, blue and infrared channels.
  • derivative signal spaces can be utilized, for instance a signal space making use of H, S, U and V signals (or their derivates), wherein the channels may represent hue, saturation, and the colour difference signals U and V.
  • the signal extraction unit is further configured for enhancing a signal-to-noise ratio in each of the extracted characteristic sub-signals.
  • Signal enhancing may involve illumination normalization, color normalization and motion compensation measures. Basically, these measures can be applied on the characteristic sub- signal level or on the characteristic signal level, but also on both levels.
  • the partitioning unit is further configured for selectively partitioning the region of interest into an initial partition grid comprising a plurality of initial sub-samples, and for defining the at least two defined signal subsets based on a recombination of the initial sub-samples.
  • partitioning the region of interest can be even more flexible.
  • partitioning is not bound to portions having simple geometrical shapes (e.g., rectangle, triangle).
  • each of the at least two defined signal subsets may have a shape of any desired (polygonal) form.
  • the signal subsets can be suitably adapted or matched to distortions or high- indicative areas in the region of interest.
  • physiological information from remotely detected electromagnetic radiation emitted or reflected by a subject comprising the steps of:
  • the data stream comprising a sequence of signal samples representing a region of interest exhibiting a continuous or discrete characteristic signal including physiological information indicative of at least one at least partially periodic vital signal, - selectively partitioning the sequence of signal samples into at least two distinct sub-sequences of defined signal subsets representing spatial sub-regions of the region of interest,
  • the method can be carried out utilizing the device for extracting physiological information of the invention.
  • the method further comprises the step of:
  • a computer program which comprises program code means for causing a computer to perform the steps of the processing method when said computer program is carried out on a computer.
  • the term "computer” stands for a large variety of processing devices. In other words, also mobile devices having a considerable computing capacity can be referred to as computing device, even though they provide less processing power resources than standard desktop computers. Furthermore, the term “computer” may also refer to a distributed computing device which may involve or make use of computing capacity provided in a cloud environment.
  • FIG. 1 shows a schematic illustration of a general layout of a device in which the present invention can be used
  • Fig. 2a shows a simplified exemplary frame representing an observed subject
  • Fig. 2b shows an enlarged partial view of the frame according to Fig. 2a, representing a region of interest;
  • Figs. 3a, 3b show segmentation patterns applicable to respective regions of interest
  • Fig. 4 shows an illustrative representation of a region of interest which is partitioned into signal subsets
  • Fig. 5 shows a representation of groups having different numbers of subsets which are processed and eventually combined
  • Fig. 6 shows another representation of groups of signal subsets which are processed and combined in a two-stage approach
  • Figs. 7a, 7b show a representation of a region of interest which is partitioned into an initial partition grid based on which two defined signal subsets are formed;
  • Fig. 8a, 8b, 8c depict diagrams, each of which showing a spectrogram comprising physiological information obtained from a subject of interest;
  • Fig. 9 shows an illustrative block diagram representing several steps of an embodiment of a method according to the invention.
  • Fig. 1 shows a schematic illustration of a device for extracting physiological information which is denoted by a reference numeral 10.
  • the device can be utilized for recording image frames representing a remote subject 12 for remote PPG monitoring.
  • the image frames can be derived from electromagnetic radiation 14 reflected by the subject 12. Possibly, under certain conditions, in particular specific luminance conditions, at least part of the electromagnetic radiation 14 could be emitted or transmitted by the subject 12. Radiation transmission may occur when the subject 12 is exposed to strong illumination sources shining through the subject 12. Radiation emission may occur when infrared radiation caused by body heat is addressed and captured. However, for remote PPG applications, a huge portion of the electromagnetic radiation 14 can be considered radiation reflected by the subject 12.
  • the subject 12 can be a human being or an animal, or, in general, a living being. Furthermore, the subject 12 can be part of a human being highly indicative of a desired signal, e.g., a face portion or, in general, a skin portion.
  • a source of radiation such as sunlight 16a or an artificial radiation source 16b, also a combination of several radiation sources can affect or impinge on the subject 12.
  • the radiation source 16a, 16b basically emits incident radiation 18a, 18b striking the subject 12.
  • a defined part or portion of the subject 12 can be detected by a sensor means 22.
  • the sensor means 22 can be embodied, by way of example, by a camera adapted for capturing information belonging to at least one spectral component of the electromagnetic radiation 14.
  • the device 10 also can be adapted to process input signals, namely an input data stream, already recorded in advance and, in the meantime, stored or buffered.
  • the electromagnetic radiation 14 can contain a continuous or discrete characteristic signal which can be highly indicative of at least one at least partially periodic vital signal 20.
  • the characteristic signal can be embedded in an (input) data stream 24.
  • a potentially highly indicative portion of the subject 12 can be selected (or: masked with a pixel pattern).
  • a mean pixel value can be derived from the pixel pattern.
  • the detected signals can be normalized and compensated for overall disturbances to some extent.
  • the characteristic signal is considered to contain a constant (DC) portion and an alternating (AC) portion superimposing the DC portion.
  • the AC portion can be extracted and, furthermore, compensated for disturbances.
  • the AC portion of the characteristic signal can comprise a dominant frequency which can be highly indicative of the subject's 12 heart rate.
  • the mean pixel value can be represented by a characteristic signal.
  • the vital signal of interest 20 can be embedded in slight fluctuations (slight periodic property changes) of the characteristic signal.
  • the captured data stream can be considered a representation of a certain area of interest in the subject 12 which may cover an agglomerated pixel area covering a plurality of pixels.
  • the vital signal 20 may allow several conclusions concerning heart rate, heart beat, heart rate variability, respiratory rate, or even oxygen saturation.
  • the known methods for obtaining such vital signals may comprise tactile heart rate monitoring, electrocardiography, or pulse oximetry. To this end, however, obtrusive monitoring was required. As indicated above, an alternate approach is directed to unobtrusive remote measuring utilizing image processing methods.
  • the data stream 24 comprising the continuous or discrete characteristic signal can be delivered from the sensor means 22 to an interface 26. Needless to say, also a buffer means could be interposed between the sensor means 22 and the interface 26. Downstream of the interface 26, the input data stream 24' can be delivered to a processing module or processing unit 28.
  • the partitioning unit 28 can be considered a computing device or, at least, part of a computing device driven by respective logic commands (program code), so as to provide for desired data processing.
  • the partitioning unit 28 may comprise several components or units which are addressed in the following. It should be understood that each component or unit of the processing unit 28 can be implemented virtually or discretely. For instance, the partitioning unit 28 may comprise a number of processors, such as multi-core processors or single-core processors.
  • At least one processor can be utilized by the partitioning unit 28.
  • Each of the processors can be configured as a standard processor (e.g., central processing unit) or as a special purpose processor (e.g., graphics processor).
  • the partitioning unit 28 can be suitably operated so as to distribute several tasks of data processing to adequate processors.
  • the processing unit 28 comprises a segmentation unit 30 configured for determining an indicative frame section, in particular a skin portion of the subject 12, such that the indicative frame section preferably comprises the region of interest.
  • the indicative frame section and the region of interest can match or correspond to each other.
  • the indicative frame section and the region of interest also may somehow deviate in size or position.
  • the segmentation unit 30 can be adapted for skin segmentation and/or feature tracking measures. Both skin segmentation and feature tracking can be utilized for pattern detection, so as to initially detect the region of interest and to track the region of interest over time. Hence, the segmentation unit 30 can contribute in motion compensation.
  • pattern detection or skin segmentation can be performed manually by a user of the device 10.
  • the user can mask a face portion or a skin portion of the subject 12 in a frame representing an initial frame for determining an initial frame section to be processed.
  • the partitioning unit 28 may further comprise a partitioning unit 32 configured for selectively partitioning the region of interest into at least two defined signal subsets. In this way, a sequence of frames embedded in the data stream can be split into at least two subsequences.
  • splitting or partitioning typically refers to dividing an area in a region of interest.
  • the partitioning unit 28 may further comprise a signal extraction unit 34 for processing each of a plurality of the at least two distinct sub-sequences generated by the partitioning unit 32.
  • a signal extraction unit 34 for processing each of a plurality of the at least two distinct sub-sequences generated by the partitioning unit 32.
  • a signal extraction unit 34 can be configured for processing each of a plurality of the at least two distinct sub-sequences. In other words, the signal extraction unit 34 does not necessarily have to process every sub-sequence generated by the partitioning unit 32.
  • the portion of interest can be locally corrupted by distortions, it can be assumed that some of the at least two distinct sub-sequences (of respective subsets) exhibit heavily distorted signals while others may exhibit far less distorted signals.
  • the partitioning unit 32 and the signal extraction unit 34 can be further configured for multiply partitioning and processing the region of interest such that at least two groups of signal subsets are formed and processed accordingly. In this way, each time instant (frame number) of the region of interest can be processed several times.
  • the provision of at least two groups of defined signal subsets allows for far more flexibility when facing locally occurring distortions and disturbances in the region of interest.
  • the processing unit 28 can comprise a data combining unit 36 configured for generating an enhanced characteristic signal under consideration of the extracted characteristic sub-signals.
  • a data combining unit 36 configured for generating an enhanced characteristic signal under consideration of the extracted characteristic sub-signals.
  • particularities of the characteristic sub-signals can be considered. For instance, statistical measures can be applied so as to attenuate outliers which are supposed to be mainly caused by locally occurring disturbances. Hence, the resulting enhanced characteristic signal can be further improved.
  • local disturbance-related artefacts can be precisely realized and addressed.
  • merely rather untargeted (in terms of local distortions) signal enhancement measures can be applied.
  • processing unit 28 can further comprise a signal
  • the enhancement unit 38 which is configured for further processing the characteristic signal generated by the signal combining unit 36.
  • the signal enhancement unit 38 can be configured to seek for dominant frequencies in the characteristic signal.
  • the signal enhancement unit 38 can be configured for filtering the characteristic signal such that frequency portions which are considered not to be indicative of the desired vital signal 20 can be disregarded or, at least, attenuated.
  • a processed data stream 40 can be generated by the processing unit 28.
  • an (output) interface 42 can be provided to which the processed data 40 can be delivered.
  • Both interfaces 26, 42 can be embodied by the same (hardware) connectors. Via the interface 42, output data 44 can be made available for further analyses and/or for display measures.
  • the processing unit 28 as well as the interfaces 26, 42 can be embodied in a common processing apparatus or housing 48. Reference numeral 48 can also describe a virtual system boundary.
  • the sensor means 22 can be integrated in the common processing housing 48.
  • a potential overall system boundary of the device 10 is denoted by a reference numeral 50.
  • the device 10 also can be implemented as a distributed device. For instance, at least the sensor means 22 can be partitioned separate or distant from the processing unit 28.
  • functional entities of the processing unit 28 can be implemented in distributed processing devices which can be connected via cable or wireless connections or networks.
  • Fig. 2a illustrates an (image) frame 54 comprising a representation of the subject 12.
  • a skin portion of the subject 12 in particular a face portion of the subject 12, is considered to be highly indicative of the desired vital signals.
  • an indicative frame section 56 is chosen in the frame 54 .
  • pattern detection or recognition can be applied.
  • the indicative frame section 56 can be determined upon face detection.
  • the indicative frame section 56 comprises or corresponds to a region of interest 58.
  • the indicative frame section 56 and the region of interest 58 do not match.
  • the term "region of interest” typically refers to a set or an array of pixels. Since motion-related disturbances have to be expected for remote
  • Fig. 2b is an enlarged view of the region of interest 58 in the subject 12 chosen in Fig. 2a.
  • the region of interest 58 can be locally corrupted by distortions.
  • distorted regions 60a, 60b can be present in the region of interest 58.
  • the forehead region can be susceptible to specular reflections. This applies in particular when the forehead region is sweaty or greasy. Therefore, the distorted region 60a can adversely influence the results of data processing focussing on the region of interest 58 as a whole.
  • the same may apply to the distorted region 60b.
  • the distorted region 60b can be formed by a portion of the subject's 12 skin which is barely illuminated.
  • the distorted regions 60a, 60b are merely exemplary representatives of various types of distorted regions that may occur in the region of interest 58. So, presumably, distortions may locally occur everywhere in the region of interest 58. It is therefore noted that a mere reduction of the size of the region of interest 58 does not provide a sufficient solution, since the smaller the size, the greater the influence of any distortion on the overall processing results. It would therefore be beneficial to maintain the size of the region of interest 58 while providing a data processing approach, allowing to somehow selectively addressing sub-regions of the region of interest 58.
  • a preferred approach to these demands may comprise partitioning the region of interest 58 into at least two defined signal subsets.
  • partitioning the region of interest 58 comprises a multiple partition.
  • Multiply partitioning may comprise deriving at least two groups of at least two defined signal subsets from the original region of interest 58.
  • Figs. 3a and 3b are referred to.
  • Fig. 3a shows an exemplary set of groups 64, 64', 64", 64"'
  • Fig. 3b shows an alternative exemplary set of groups 64a, 64a', 64a", 64a'.
  • Each of the groups 64, 64a may correspond to a defined region of interest 58. This may include a representation of the region of interest 58, 58a in each of the groups 64, 64a.
  • the region of interest 58 is multiply partitioned.
  • the first group 64 basically may comprise a single signal subset which may correspond to an original set 73 covering the whole area of the region of interest 58.
  • the second group 64' may comprise two defined signal subsets which are denoted for illustrative purposes by reference numerals 66a, 66b.
  • the third group 64" may comprise four defined subsets.
  • the fourth group 64"' may comprise sixteen signal subsets.
  • the first group 64a also comprises a single signal subset which can correspond to an unpartitioned original signal subset 73 basically corresponding to the whole area of the region of interest 58.
  • the second group 64a' may comprise two defined signal subsets.
  • the third group 64a" may comprise four defined signal subsets.
  • 64a'" may comprise nine defined signal subsets. It goes without saying that a great variety of sets of groups 64, 64a comprising a great variety of signal subsets can be envisaged. The size, orientation and order of the subsets within the groups 64, 64a may vary as well.
  • Fig. 4 illustrates an exemplary approach to data processing which is based on a partitioned region of interest.
  • a group 64" is formed comprising four signal subsets 66a, 66b, 66c, 66d.
  • the signal subsets 66a, 66b, 66c, 66d may represent the region of interest 58.
  • the defined signal subsets 66a, 66b, 66c, 66d may have the same shape and size, but may also comprise different shapes and sizes.
  • the defined signal subsets 66a, 66b, 66c, 66d may represent disjoined, adjoined or even overlapping portions of the region of interest.
  • a characteristic sub-signal 74a, 74b, 74c, 74d can be derived through sub-processing (indicated by the arrows 68a, 68b, 68c, 68d).
  • the characteristic sub-signals 74 may be more or less affected by local disturbances occurring in some of the defined signal subsets 66a, 66b, 66c, 66d.
  • the characteristic sub-signals 74a, 74b, 74c, 74d eventually can be combined, so as to generate an enhanced characteristic signal 72.
  • Signal combining 70 may involve applying statistical algorithms, so as to attenuate or disregard characteristic sub-signals 74a, 74b, 74c, 74d which are considered to be massively corrupted.
  • Figs. 5 and 6 illustrate alternative exemplary partitioning, processing and combining approaches.
  • Fig. 5 exemplarily three groups 64, 64', 64" of signal subsets are shown.
  • the respective characteristic sub-signal 74 can be processed.
  • Signal combining 70 may involve combining some or each of the characteristic sub-signals 74.
  • each of the characteristic sub-signals 74 is handled on the same level when combining. In other words, group-wise signal "pre-combining" is not applied.
  • Fig. 6 also shows three groups 64, 64', 64" of signal subsets. For at least some or all of the signal subsets, characteristic sub-signals 74 can be processed. An intermediate group-wise combining step 76, 76', 76" may follow. As indicated by a dashed curly bracket, signal combining 76 for the first group 64 is basically not necessary, since the characteristic sub-signal 74 is the single signal representative of group 64.
  • the respective characteristic sub-signal 74 can be "looped through" in the signal combining step 76 since the characteristic sub-signal 74 basically corresponds to an intermediate characteristic signal 78.
  • intermediate group-wise signal combination 76', 76" involves a combination of the respective characteristic sub- signals 74a', 74b', and 74a", 74b", 74c", 74d", respectively. Consequently, intermediate characteristic signals 78', 78" can be generated.
  • the intermediate characteristic signals 78, 78', 78" each of which is attributable to a group 64, 64', 64" can be combined (reference numeral 70), so as to generate a resulting enhanced characteristic signal 72.
  • Both intermediate group-wise combining 76 and signal combining 70 may involve statistical data shape processing.
  • Figs. 7a and 7b describe an alternative approach to partitioning measures applicable to the region of interest 58. It should be understood that this approach can be combined with the signal combining approaches described in connection with Figs. 4, 5 and 6.
  • an initial region of interest 58 is partitioned or split into an initial partition grid 82 which is composed of a plurality of initial sub-samples 80a ... 80n.
  • at least some of the initial sub-samples 80a ... 80n can be recombined, so as to define at least two signal subsets 84a, 84b. Consequently, each of the at least two defined signal subsets 84a, 84b can comprise a variety of different shapes.
  • Signal subset determination is far more flexible and can address distortions and disturbances occurring in the region of interest 58 even more precisely.
  • Fig. 7b may basically correspond to Fig. 4.
  • Fig. 7b illustrates that for each of the defined subsets 84a, 84b, a respective characteristic sub-signal 74a, 74b can be processed.
  • an enhanced characteristic signal 72 can be generated upon combining the respective characteristic sub-signals 74a, 74b.
  • Figs. 8a, 8b and 8c show exemplary spectrograms illustrating results of remote photoplethysmography analyses utilizing several approaches.
  • f denotes frequency
  • t denotes time.
  • the frequency axes can represent Hz (Hertz) values, while the time axes may also stand for the number of processed image frames.
  • each of the respective time axes in Figs. 8a, 8b and 8c may cover a span of about 3200 frames which may basically correspond to a recording duration of about 160 s (seconds) at a sample rate (or: frame rate) of about 20 frames per second.
  • the spectrograms shown in Fig. 8a, Fig. 8b and Fig. 8c, respectively, exemplify characteristic signals extracted at the same situation, namely signal detection results obtained from a subject initially stationary (standing still) and, later on, performing some workout on a fitness device. Under these circumstances, given that considerable periodic subject motion is involved, motion-related disturbances may render the detection and processing challenging.
  • the spectrogram of Fig. 8a illustrates a conventional signal extraction approach wherein the desired characteristic signal is derived (directly) from the whole portion of interest. Consequently, the signal-to-noise ratio is poor. The desired signal is almost completely hidden in noise.
  • the spectrogram of Fig. 8b represents a data processing approach wherein the region of interest is partitioned into basically equally sized subsets from which respective characteristic sub-signals were extracted. Then, the characteristic sub-signals are combined, so as to arrive at an enhanced characteristic signal. In Fig. 8b, signal combination is performed under consideration of a weighted average of the characteristic sub-signals.
  • the extracted signal is processed under consideration of a plurality of groups each of which having a different number of signal subsets.
  • four groups each of which representing the region of interest can be utilized, wherein the first group may comprise a (single) signal subset, wherein a second group may comprise two signal subsets, wherein a third group may comprise four signal subsets, and wherein a fourth group may comprise eight signal subsets.
  • the first group may comprise a (single) signal subset
  • a second group may comprise two signal subsets
  • a third group may comprise four signal subsets
  • a fourth group may comprise eight signal subsets.
  • the characteristic sub-signals are combined, so as to arrive at an improved enhanced characteristic signal.
  • Signal combination can be performed under consideration of weighted averaging algorithms which may include variance calculation for each of the characteristic sub-signals. It can be clearly seen in Fig. 8c that the vital signal 94 obviously dominates the spectrogram.
  • FIG. 9 is referred to, schematically illustrating a method for extracting
  • an input data stream or a sequence 102 comprising several frames 104a, 104b, 104c is received.
  • a time axis is indicated by an arrow t.
  • the data stream can be delivered from the sensor means 22 or a data buffer or storage means.
  • the data stream can be embodied, by way of example, by a sequence of image frames varying over time.
  • the image frames can comprise RGB-based pixel data.
  • the data stream typically comprises a representation of a subject of interest.
  • a segmentation or pattern detection can be applied to at least one of the frames 104a, 104b, 104c.
  • a region of interest 108 can be determined.
  • face recognition and/or skin segmentation can be utilized.
  • Another subsequent step 110 may follow, which may be directed to motion compensation, so as to allow to track the region of interest.
  • a positional shift 112 of the region of interest 108 in consecutive frames 104a, 104b, 104c can be calculated, so as to determine a presumed position of the region of interest 108a, 108b over time.
  • a further step 114 may follow which may comprise partitioning the region of interest 108 into at least two defined signal subsets 66a, 66b, 66c, 66d. Furthermore, a sequence comprising the region of interest 108 can be multiply partitioned, so as to derive two or more groups of signal subsets 66a, 66b, 66c, 66d therefrom (not shown in Fig. 9). Over time, for each of the defined signal subsets 66a, 66b, 66c, 66d, a respective subsequence 116a, 116b, 116c, 116d may be formed based on a sequence 102' representing the region of interest 108. As indicated by a plurality of parallel connecting arrows, at least some of the following steps 118, 124, 134 may be applied in parallel to each of sub-sequences 116a, 116b, 116c, 116d.
  • a characteristic sub-signal 122 per group can be detected.
  • the characteristic sub-signal 122 may comprise a characteristic index element 128 per frame.
  • the characteristic index element 128 can be attributed to a signal space 126.
  • the signal space 126 may comprise at least two, preferably three, complementary channels 130a, 130b, 130c, each of which is indicative of a defined spectral portion. For instance, the at least two
  • complementary channels 130a, 130b, 130c can be embodied by R-, G-, and B-channels. So, basically the characteristic index element 128 can comprise three respective components. Based on a linear combination of the at least two components of the characteristic index element 128, several signal enhancement measures can be achieved. For instance, compensation for motion and/or changing illumination conditions can be envisaged.
  • a processed characteristic sub-signal 132 can be obtained for each of the at least two sub-sequences 116a, 116b, 116c, 116d.
  • a further step 134 the plurality of processed characteristic sub-signals 132 can be combined, so as to generate an enhanced characteristic signal 135.
  • statistical signal improvement algorithms can be applied.
  • further analyzing measures can be applied to the enhanced characteristic signal 135.
  • filtering, weighting and further signal enhancement operations ca be envisaged. For instance, a temporal representation 144 and/or a frequency-based representation 140 of desired vital signals of interest 138, 142 can be achieved.
  • the present invention can be applied in the field of health care, e.g., unobtrusive remote patient monitoring, general surveillances, security monitoring and so-called life style environments, such as fitness equipment, or the like.
  • Applications may include monitoring of oxygen saturation (pulse oximetry), heart rate, blood pressure, cardiac output, changes of blood perfusion, assessment of autonomic functions, and detection of peripheral vascular diseases.
  • the word “comprising” does not exclude other elements or steps, and the indefinite article “a” or “an” does not exclude a plurality.
  • a single element or other unit may fulfill the functions of several items recited in the claims.
  • the mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage.
  • a computer program may be stored/distributed on a suitable medium, such as an optical storage medium or a solid-state medium supplied together with or as part of other hardware, but may also be distributed in other forms, such as via the Internet or other wired or wireless telecommunication systems.

Abstract

La présente invention concerne un dispositif et un procédé d'extraction d'informations physiologiques d'un rayonnement (15) électromagnétique détecté de manière à distance émis ou réfléchi par un sujet. Un courant de données (24) déduit d'un rayonnement (14) électromagnétique détecté est reçu, le courant de données (24) comprenant une séquence (102) d'échantillons de signal (104) représentant une région d'intérêt (58 ; 108) présentant un signal caractéristique continu ou discret (128) comprenant des informations physiologiques indicatives d'au moins un signal vital au moins partiellement périodique (20 ; 138, 142). La séquence (102) d'échantillons de signal (104) est partitionnée de manière sélective en au moins deux sous-séquences distinctes (116a, 116b, 116c, 116d) de sous-ensembles de signal définis (66, 84) représentant des sous-régions spatiales de la région d'intérêt (58 ; 108). Chacune d'une pluralité des au moins deux sous-séquences distinctes (116a, 116b, 116c, 116d) est traitée et ainsi une pluralité de sous-signaux caractéristiques indicatifs distincts (74a, 74b, 74c, 74d ; 122) est extraite, chacun des au moins deux sous-signaux caractéristiques indicatifs distincts (74a, 74b, 74c, 74d ; 122) étant extrait d'une sous-séquence distincte respective (116a, 116b, 116c, 116d). Par conséquent, un signal caractéristique amélioré (72 ; 135) est généré par combinaison des sous-signaux caractéristiques extraits (74a, 74b, 74c, 74d ; 122).
PCT/IB2013/056324 2012-08-06 2013-08-01 Dispositif et procédé d'extraction d'informations physiologiques WO2014024104A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US201261679913P 2012-08-06 2012-08-06
US61/679,913 2012-08-06

Publications (1)

Publication Number Publication Date
WO2014024104A1 true WO2014024104A1 (fr) 2014-02-13

Family

ID=49223830

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/IB2013/056324 WO2014024104A1 (fr) 2012-08-06 2013-08-01 Dispositif et procédé d'extraction d'informations physiologiques

Country Status (1)

Country Link
WO (1) WO2014024104A1 (fr)

Cited By (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140334697A1 (en) * 2013-05-08 2014-11-13 Koninklijke Philips N.V. Device for obtaining a vital sign of a subject
US9385768B2 (en) 2012-11-02 2016-07-05 Koninklijke Philips N.V. Device and method for extracting physiological information
CN106343984A (zh) * 2015-07-17 2017-01-25 松下知识产权经营株式会社 注意信息提示装置和注意信息提示方法
JP2018114015A (ja) * 2017-01-16 2018-07-26 パナソニックＩｐマネジメント株式会社 生体情報検出装置、生体情報検出方法及び生体情報検出システム
JPWO2017085894A1 (ja) * 2015-11-20 2018-08-02 富士通株式会社 脈波分析装置、脈波分析方法、および脈波分析プログラム
EP3378384A4 (fr) * 2015-11-20 2018-10-24 Fujitsu Limited Dispositif, procédé et programme de traitement de l'information
EP3449820A1 (fr) 2017-08-30 2019-03-06 Qompium Procédé et système mis en uvre par ordinateur de photopléthysmographie directe (ppg)
US10238292B2 (en) 2013-03-15 2019-03-26 Hill-Rom Services, Inc. Measuring multiple physiological parameters through blind signal processing of video parameters
EP3473173A1 (fr) 2017-10-19 2019-04-24 Qompium Procédé et système mis en uvre par ordinateur de photopléthysmographie directe (ppg) comportant plusieurs capteurs
US10349900B2 (en) 2014-05-07 2019-07-16 Koninklijke Philips N.V. Device, system and method for extracting physiological information
EP4050885A1 (fr) * 2021-02-25 2022-08-31 Samsung Electronics Co., Ltd. Appareil et procédé de mesure de signaux biologiques multiples à l'aide d'un capteur d'image et dispositif électronique

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2011042858A1 (fr) * 2009-10-06 2011-04-14 Koninklijke Philips Electronics N.V. Procédé et système pour traiter un signal comportant au moins une composante caractéristique d'un phénomène périodique chez un être vivant
WO2011042839A1 (fr) * 2009-10-06 2011-04-14 Koninklijke Philips Electronics N.V. Procédé et système destinés à obtenir un premier signal pour analyse en vue de caractériser au moins une composante périodique de celui-ci
US20110251493A1 (en) 2010-03-22 2011-10-13 Massachusetts Institute Of Technology Method and system for measurement of physiological parameters

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2011042858A1 (fr) * 2009-10-06 2011-04-14 Koninklijke Philips Electronics N.V. Procédé et système pour traiter un signal comportant au moins une composante caractéristique d'un phénomène périodique chez un être vivant
WO2011042839A1 (fr) * 2009-10-06 2011-04-14 Koninklijke Philips Electronics N.V. Procédé et système destinés à obtenir un premier signal pour analyse en vue de caractériser au moins une composante périodique de celui-ci
US20110251493A1 (en) 2010-03-22 2011-10-13 Massachusetts Institute Of Technology Method and system for measurement of physiological parameters

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
VERKRUYSSE, W.: "Remote Plethysmographic Imaging Using Ambient Light", OPCTICS EXPRESS, OPTICAL SOCIETY OF AMERICA, vol. 16, no. 26, 2008, pages 21434 - 21445
WIM VERKRUYSSE1 ET AL: "Remote plethysmographic imaging using ambient light", OPTICS EXPRESS, OSA (OPTICAL SOCIETY OF AMERICA), WASHINGTON DC, (US), vol. 16, no. 26, 22 December 2008 (2008-12-22), pages 21434 - 21445, XP007913060, ISSN: 1094-4087 *

Cited By (19)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9385768B2 (en) 2012-11-02 2016-07-05 Koninklijke Philips N.V. Device and method for extracting physiological information
US10238292B2 (en) 2013-03-15 2019-03-26 Hill-Rom Services, Inc. Measuring multiple physiological parameters through blind signal processing of video parameters
US9339210B2 (en) * 2013-05-08 2016-05-17 Koninklijke Philips N.V. Device for obtaining a vital sign of a subject
US20140334697A1 (en) * 2013-05-08 2014-11-13 Koninklijke Philips N.V. Device for obtaining a vital sign of a subject
US10349900B2 (en) 2014-05-07 2019-07-16 Koninklijke Philips N.V. Device, system and method for extracting physiological information
CN106343984A (zh) * 2015-07-17 2017-01-25 松下知识产权经营株式会社 注意信息提示装置和注意信息提示方法
JPWO2017085894A1 (ja) * 2015-11-20 2018-08-02 富士通株式会社 脈波分析装置、脈波分析方法、および脈波分析プログラム
EP3378383A4 (fr) * 2015-11-20 2018-10-24 Fujitsu Limited Dispositif d'analyse d'onde pulsée, procédé d'analyse d'onde pulsée et programme d'analyse d'onde pulsée
EP3378384A4 (fr) * 2015-11-20 2018-10-24 Fujitsu Limited Dispositif, procédé et programme de traitement de l'information
US10743783B2 (en) 2015-11-20 2020-08-18 Fujitsu Limited Pulse wave analysis apparatus, pulse wave analysis method, and non-transitory computer-readable storage medium
JP2018114015A (ja) * 2017-01-16 2018-07-26 パナソニックＩｐマネジメント株式会社 生体情報検出装置、生体情報検出方法及び生体情報検出システム
EP3449820A1 (fr) 2017-08-30 2019-03-06 Qompium Procédé et système mis en uvre par ordinateur de photopléthysmographie directe (ppg)
CN111050638A (zh) * 2017-08-30 2020-04-21 康比姆公司 用于接触式光学体积描记术(ppg)的计算机实现的方法和系统
WO2019042739A1 (fr) 2017-08-30 2019-03-07 Qompium Procédé mis en œuvre par ordinateur et système pour photopléthysmographie (ppg) par contact
US11583198B2 (en) 2017-08-30 2023-02-21 Qompium Computer-implemented method and system for contact photoplethysmography (PPG)
WO2019076510A1 (fr) 2017-10-19 2019-04-25 Qompium Procédé mis en œuvre par ordinateur et système pour photopléthysmographie (ppg) directe au moyen d'une pluralité de capteurs
EP3473173A1 (fr) 2017-10-19 2019-04-24 Qompium Procédé et système mis en uvre par ordinateur de photopléthysmographie directe (ppg) comportant plusieurs capteurs
US11701015B2 (en) 2017-10-19 2023-07-18 Qompium Computer-implemented method and system for direct photoplethysmography (PPG) with multiple sensors
EP4050885A1 (fr) * 2021-02-25 2022-08-31 Samsung Electronics Co., Ltd. Appareil et procédé de mesure de signaux biologiques multiples à l'aide d'un capteur d'image et dispositif électronique

Similar Documents

Publication Publication Date Title
US20220054089A1 (en) Device, system and method for generating a photoplethysmographic image carrying vital sign information of a subject
WO2014024104A1 (fr) Dispositif et procédé d'extraction d'informations physiologiques
US9385768B2 (en) Device and method for extracting physiological information
EP2748762B1 (fr) Détection de signal à distorsion réduite
US9928607B2 (en) Device and method for obtaining a vital signal of a subject
CN112074226B (zh) 用于生命体征的远程测量的系统和方法
EP2967378B1 (fr) Dispositif et procédé pour déterminer des signes vitaux chez un sujet
EP2967376B1 (fr) Dispositif et procédé pour déterminer des signes vitaux chez un sujet
CN106999116B (zh) 用于皮肤检测的设备和方法
Blackford et al. Effects of frame rate and image resolution on pulse rate measured using multiple camera imaging photoplethysmography
US20160220128A1 (en) Improved signal selection for obtaining a remote photoplethysmographic waveform
Fan et al. Non-contact remote estimation of cardiovascular parameters
JP2018510669A (ja) 皮膚検出に関するデバイス、システム、及び方法
CN105813564A (zh) 用于确定对象的生命体征的设备和方法
Feng et al. Motion artifacts suppression for remote imaging photoplethysmography
WO2013038326A1 (fr) Détection d'un signal à distorsion réduite
Dosso et al. Eulerian magnification of multi-modal RGB-D video for heart rate estimation
WO2019145142A1 (fr) Dispositif, système et procédé de détermination d'au moins un signe vital d'un sujet

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 13765514

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 13765514

Country of ref document: EP

Kind code of ref document: A1