CN116831519A - Quantitative assessment method and system for stationary tremor of dyskinesia - Google Patents
Quantitative assessment method and system for stationary tremor of dyskinesia Download PDFInfo
- Publication number
- CN116831519A CN116831519A CN202210313352.7A CN202210313352A CN116831519A CN 116831519 A CN116831519 A CN 116831519A CN 202210313352 A CN202210313352 A CN 202210313352A CN 116831519 A CN116831519 A CN 116831519A
- Authority
- CN
- China
- Prior art keywords
- tremor
- patient
- frequency
- curve
- amplitude
- Prior art date
- Legal status (The legal status 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 status listed.)
- Pending
Links
- 206010044565 Tremor Diseases 0.000 title claims abstract description 218
- 238000000034 method Methods 0.000 title claims abstract description 50
- 208000012661 Dyskinesia Diseases 0.000 title claims abstract description 25
- 238000003384 imaging method Methods 0.000 claims abstract description 28
- 238000001914 filtration Methods 0.000 claims abstract description 24
- 230000009466 transformation Effects 0.000 claims abstract description 16
- 206010071390 Resting tremor Diseases 0.000 claims description 20
- 238000000605 extraction Methods 0.000 claims description 13
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 claims description 7
- 239000000284 extract Substances 0.000 claims description 7
- 208000016285 Movement disease Diseases 0.000 claims description 4
- 238000006243 chemical reaction Methods 0.000 claims description 4
- 208000035475 disorder Diseases 0.000 claims 1
- 238000011158 quantitative evaluation Methods 0.000 abstract description 8
- 238000011156 evaluation Methods 0.000 abstract description 4
- 208000018737 Parkinson disease Diseases 0.000 description 6
- 210000001747 pupil Anatomy 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 208000024891 symptom Diseases 0.000 description 5
- 238000004364 calculation method Methods 0.000 description 4
- 238000003745 diagnosis Methods 0.000 description 4
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 208000027089 Parkinsonian disease Diseases 0.000 description 2
- 206010034010 Parkinsonism Diseases 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 201000010099 disease Diseases 0.000 description 2
- 238000002567 electromyography Methods 0.000 description 2
- 201000006517 essential tremor Diseases 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 208000015879 Cerebellar disease Diseases 0.000 description 1
- 206010008748 Chorea Diseases 0.000 description 1
- 206010020850 Hyperthyroidism Diseases 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000003466 anti-cipated effect Effects 0.000 description 1
- 210000000544 articulatio talocruralis Anatomy 0.000 description 1
- 208000012601 choreatic disease Diseases 0.000 description 1
- 238000003759 clinical diagnosis Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000035876 healing Effects 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 210000003141 lower extremity Anatomy 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 238000011002 quantification Methods 0.000 description 1
- 230000001954 sterilising effect Effects 0.000 description 1
- 230000000638 stimulation Effects 0.000 description 1
- 210000001364 upper extremity Anatomy 0.000 description 1
- 210000003857 wrist joint Anatomy 0.000 description 1
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/40—Detecting, measuring or recording for evaluating the nervous system
- A61B5/4076—Diagnosing or monitoring particular conditions of the nervous system
- A61B5/4082—Diagnosing or monitoring movement diseases, e.g. Parkinson, Huntington or Tourette
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/103—Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
- A61B5/11—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/103—Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
- A61B5/11—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb
- A61B5/1101—Detecting tremor
-
- 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
- A61B5/7235—Details of waveform analysis
- A61B5/725—Details of waveform analysis using specific filters therefor, e.g. Kalman or adaptive filters
-
- 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
- A61B5/7235—Details of waveform analysis
- A61B5/7253—Details of waveform analysis characterised by using transforms
- A61B5/7257—Details of waveform analysis characterised by using transforms using Fourier transforms
-
- 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
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Public Health (AREA)
- Medical Informatics (AREA)
- Biomedical Technology (AREA)
- General Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- Pathology (AREA)
- Animal Behavior & Ethology (AREA)
- Veterinary Medicine (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Surgery (AREA)
- Heart & Thoracic Surgery (AREA)
- Physiology (AREA)
- Neurology (AREA)
- Neurosurgery (AREA)
- Dentistry (AREA)
- Signal Processing (AREA)
- Psychiatry (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Epidemiology (AREA)
- Mathematical Physics (AREA)
- Developmental Disabilities (AREA)
- Primary Health Care (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The embodiment of the application discloses a quantitative evaluation method and a quantitative evaluation system for stationary tremor of dyskinesia, wherein the evaluation method comprises the following steps: s1, acquiring a video image of stationary tremor at a tremor prone part of a patient through imaging equipment; s2, extracting coordinates of key points of a human body from each frame of the video image to form a coordinate position curve of the key points of the human body; s3, carrying out frequency domain transformation on the coordinate position curve, and extracting tremor frequency of stationary tremor of the patient; s4, carrying out band-pass filtering on the coordinate position curve after the frequency domain transformation, and extracting the maximum pixel tremor amplitude of the static tremor of the patient; s5, acquiring the actual human body characteristic length of the patient and the pixel length corresponding to the actual human body characteristic length, and scaling the maximum actual tremor amplitude of the patient. The technical scheme provided by the embodiment of the application has the advantages of convenience, accuracy and non-contact, and is particularly convenient to use in remote inquiry.
Description
Technical Field
The application relates to the field of quantitative evaluation of dyskinesia, in particular to a quantitative evaluation method and an evaluation system for stationary tremors of dyskinesia.
Background
The stationary tremor is one of the main clinical symptoms of dyskinesia diseases such as parkinsonism, essential tremor and the like, and quantitative indexes such as whether tremor occurs, the amplitude, the frequency and the like of tremor have important reference significance for clinical diagnosis, evaluation and grading of the diseases.
At present, diagnosis and evaluation of resting tremors are mainly carried out by means of scoring a visual scale of a doctor, electromyography examination, a wearable sensor and the like. The visual scale scoring mechanism is seriously dependent on the clinical experience of doctors, and has strong subjectivity and variability, so that the phenomenon that different doctors have great differences in diagnosis, treatment and judgment after healing of the same patient can be frequently seen in clinical work. The electromyography examination uses needle electrodes and applies an electric stimulation technology, so that the examined person has certain pain and damage in the examination process. The wearable sensor needs a series of operation flows of wearing, detaching, charging, sterilizing and the like in use, and the operation is not convenient enough. Chinese patent CN105701806B provides a method for detecting parkinsonian tremor motion characteristics based on depth images, but in use, a patient is required to wear a pure-color glove and shoot with a specific device (depth camera), so that the operation is not convenient enough, and only static tremor of the hand can be evaluated, but the method cannot be applied to other parts of the body such as the face, the legs and the like.
In summary, it is difficult in the prior art to objectively and quantitatively evaluate symptoms of resting tremor in a convenient, non-contact manner.
The information included in this background section of the specification of the present application, including any references cited herein and any descriptions or discussions thereof, is included solely for the purpose of technical reference and is not to be construed as a subject matter that would limit the scope of the present application.
Disclosure of Invention
The present application has been made in view of the above and other further ideas.
In view of the shortcomings of the prior art, according to an aspect of the present application, it is an object of the present application to provide a method for quantitatively evaluating symptoms of resting tremor, which is convenient to operate, non-contact, and objective in result.
According to one aspect of the present application, there is provided a quantitative assessment method for stationary tremor of dyskinesia, capable of objectively and quantitatively assessing stationary tremor in a convenient and non-wearable manner, the method comprising the steps of:
s1: collecting a video image of stationary tremor at a tremor-prone part of a patient through imaging equipment;
s2: extracting coordinates of key points of a human body from each frame of the video image to form a coordinate position curve of the key points of the human body;
s3: carrying out frequency domain transformation on the coordinate position curve, and extracting tremor frequency of stationary tremor of a patient;
s4: band-pass filtering the coordinate position curve after the frequency domain transformation, and extracting the maximum pixel tremor amplitude of the stationary tremor of the patient;
s5: the method comprises the steps of obtaining the actual human body characteristic length of a patient and the pixel length corresponding to the actual human body characteristic length, and scaling the maximum actual tremor amplitude of the patient.
Preferably, the imaging device is kept stationary during the acquisition in step S1, and the distance of the imaging device from the patient is kept fixed.
Preferably, the coordinate position curve in the step S2 is a curve formed by coordinates of the same human body key point in all frames of the video image, wherein the coordinate position curve includes at least one of the following directional dimension curves: an x-direction curve, a y-direction curve, a z-direction curve.
More specifically, the step S3 includes the steps of:
s3-1: performing discrete Fourier transform or fast Fourier transform on each direction dimension of the coordinate position curve to obtain a frequency domain sequence of each direction dimension;
s3-2: converting the frequency domain sequence of each direction dimension into a frequency domain amplitude curve;
s3-3: respectively determining frequency points with maximum absolute values of the frequency domain amplitude curves of all direction dimensions in a frequency band not less than 4Hz as direction tremble frequencies of the direction dimensions;
s3-4: maximum of tremor frequency in the directionAnd taking the direction tremor frequency corresponding to the direction dimension with the largest frequency domain tremor amplitude as the tremor frequency of the patient as the frequency domain tremor amplitude of the direction dimension.
Preferably, the step S4 specifically includes the following steps:
s4-1: selecting a key frequency segment, selecting a filtering method and a band-pass filter, and respectively filtering the frequency domain sequences of each direction dimension of the coordinate position curve to obtain a filtered curve of each direction dimension;
s4-2: according to the filtered curves of all the direction dimensions, calculating comprehensive tremor amplitude curves of a plurality of the direction dimensions;
s4-3, determining a maximum value in the comprehensive tremor amplitude curve, wherein the maximum value corresponds to the maximum pixel tremor amplitude of the patient.
Preferably, the selecting the critical frequency band in step S4-1 includes: a) Selecting a frequency band with the tremor frequency of the patient as a center and a reasonable threshold value as a radius; b) Frequency bins are selected that are associated with the patient diagnosed with the condition.
More specifically, the step S5 includes the steps of:
s5-1: measuring the actual length between two human body key points, the distance of which is not basically changed along with tremor movement, on the part according to the tremor-prone part shot in the video image;
s5-2: acquiring the pixel distance between two human body key points in the video image;
s5-3: and scaling the maximum actual tremor amplitude of the patient according to the maximum pixel tremor amplitude, the actual human body characteristic length and the pixel distance.
According to another aspect of the present application, there is provided a resting tremor quantitative assessment system for dyskinesia, the system comprising:
the video image acquisition module is configured to acquire a video image of stationary tremor of a tremor prone part of a patient through the imaging equipment;
the coordinate position curve forming module is configured to extract coordinates of key points of a human body in each frame of the video image and form a coordinate position curve of the key points of the human body;
the tremor frequency extraction module is configured to perform frequency domain transformation on the coordinate position curve and extract tremor frequency of stationary tremor of a patient;
the pixel tremor amplitude extraction module is configured to carry out band-pass filtering on the coordinate position curve after the frequency domain transformation, and extract the maximum pixel tremor amplitude of the stationary tremor of the patient; and
the actual tremor amplitude conversion module is configured to acquire the actual human body characteristic length of the patient and the pixel length in the video image corresponding to the actual human body characteristic length, and convert the maximum actual tremor amplitude of the patient according to the proportion.
Preferably, the imaging device is kept stationary during video image acquisition and the distance from the imaging device to the patient is kept fixed.
Preferably, the coordinate position curve is a curve formed by coordinates of the same human body key point in all frames of the video image, wherein the coordinate position curve comprises at least one of the following directional dimension curves: an x-direction curve, a y-direction curve, a z-direction curve.
Preferably, the tremor frequency extraction module is configured to further:
performing discrete Fourier transform or fast Fourier transform on each direction dimension of the coordinate position curve to obtain a frequency domain sequence of each direction dimension;
converting the frequency domain sequence of each direction dimension into a frequency domain amplitude curve;
respectively determining frequency points with maximum absolute values of the frequency domain amplitude curves of all direction dimensions in a frequency band not less than 4Hz as direction tremble frequencies of the direction dimensions;
maximum of tremor frequency in the directionAnd taking the direction tremor frequency corresponding to the direction dimension with the largest frequency domain tremor amplitude as the tremor frequency of the patient as the frequency domain tremor amplitude of the direction dimension.
Preferably, the pixel tremor amplitude extraction module is configured to further:
selecting a key frequency segment, selecting a filtering method and a band-pass filter, and respectively filtering the frequency domain sequences of each direction dimension of the coordinate position curve to obtain a filtered curve of each direction dimension;
according to the filtered curves of all the direction dimensions, calculating comprehensive tremor amplitude curves of a plurality of the direction dimensions;
determining a maximum in the integrated tremor magnitude curve, the maximum corresponding to the maximum pixel tremor magnitude of the patient.
Preferably, the selecting the key frequency band includes: a) Selecting a frequency band with the tremor frequency of the patient as a center and a reasonable threshold value as a radius; b) Frequency bins are selected that are associated with the patient diagnosed with the condition.
Preferably, the actual tremor magnitude scaling module is configured to further:
measuring the actual length between two human body key points, the distance of which is not basically changed along with tremor movement, on the part according to the tremor-prone part shot in the video image;
acquiring the pixel distance between two human body key points in the video image;
and scaling the maximum actual tremor amplitude of the patient according to the maximum pixel tremor amplitude, the actual human body characteristic length and the pixel distance.
Compared with the prior art, the embodiment of the application has the following beneficial effects:
the embodiment of the application has the advantages of convenience, accuracy and non-contact, provides a brand new and convenient solution for quantitative evaluation of the current stationary tremor symptoms of the movement disorder disease, and is convenient to use in remote inquiry.
Further embodiments of the application also enable other advantageous technical effects not listed one after another, which may be partly described below and which are anticipated and understood by a person skilled in the art after reading the present application.
Drawings
The above-mentioned and other features and advantages of these embodiments, and the manner of attaining them, will become more apparent and the embodiments of the application will be better understood by reference to the following description taken in conjunction with the accompanying drawings, wherein:
FIG. 1 is a schematic illustration of a flow of a method for quantitative assessment of resting tremor of dyskinesia according to an embodiment of the application;
FIG. 2 is a schematic representation of a coordinate position curve of the mandibular tip of a typical parkinsonian patient face in accordance with an embodiment of the application;
FIG. 3 is a schematic diagram of a frequency domain amplitude plot in Hz according to an embodiment of the present application;
FIG. 4 is a schematic illustration of a band-pass filtered coordinate position curve in accordance with an embodiment of the present application; and
fig. 5 is a schematic representation of an integrated tremor magnitude curve in accordance with an embodiment of the present application.
Fig. 6 is a schematic diagram of a system for quantitative assessment of resting tremor of dyskinesia according to an embodiment of the present application.
Detailed Description
The details of one or more embodiments of the application are set forth in the accompanying drawings and the description below. Other features, objects, and advantages of the application will be apparent from the description and drawings, and from the claims.
It is to be understood that the illustrated and described embodiments are not limited in application to the details of construction and the arrangement of components set forth in the following description or illustrated in the drawings. The illustrated embodiments may be other embodiments and can be implemented or performed in various ways. Examples are provided by way of explanation, not limitation, of the disclosed embodiments. Indeed, it will be apparent to those skilled in the art that various modifications and variations can be made to the various embodiments of the application without departing from the scope or spirit of the disclosure. For example, features illustrated or described as part of one embodiment can be used with another embodiment to yield still a further embodiment. Accordingly, the present disclosure is intended to cover such modifications and variations as fall within the scope of the appended claims and their equivalents.
Also, it is to be understood that the phraseology and terminology used herein is for the purpose of description and should not be regarded as limiting. The use of "including," "comprising," or "having" and variations thereof herein is meant to encompass the items listed thereafter and equivalents thereof as well as additional items.
The present application will be described in more detail below with reference to specific embodiments thereof.
Fig. 1 shows a flow chart of a method for quantitative assessment of resting tremor of dyskinesia according to an embodiment of the present application. The resting tremor quantitative evaluation method according to the present embodiment includes steps S1 to S5, specifically as follows.
Step S1, acquiring a video image of stationary tremor of a tremor prone part of a patient through imaging equipment.
Imaging devices that may be used include, but are not limited to, conventional monocular, binocular and multi-view cameras, depth cameras, infrared cameras, and the like, as are commonly used in the art. The specific sites capable of quantitative assessment of resting tremor, i.e., sites that are subject to photography or imaging, are typically sites susceptible to tremor, including but not limited to, the lips, lower jaw, whole face, palm, sole, whole upper limb, whole lower limb, whole body, etc. of the patient. Types of video images captured include, but are not limited to, normal RGB color video, gray scale video, color or gray scale video with depth information, binocular and multi-view color or gray scale video, and the like.
For example, in a diagnosis and treatment institution of a hospital or the like, more complex and accurate imaging equipment such as a binocular camera and a depth camera can be adopted to respectively acquire video images of a plurality of tremor prone positions of a patient in a sitting state, acquire video images with depth information, and facilitate more accurate processing analysis. In the remote consultation process, after the common mobile phone is simply fixed, the camera of the mobile phone is used for collecting the whole body video image of the patient in the sitting state, so that the operation of the accompanying person is more convenient, and the common imaging equipment is used for collecting the video image.
In addition, in the process of capturing video images, in order to perform accurate quantitative evaluation later, the imaging device is generally required to be kept still, and the distance between the imaging device and the patient is required to be kept fixed.
And S2, extracting coordinates of key points of the human body from each frame of the acquired video image to form a coordinate position curve.
Specific methods for extracting coordinates include, but are not limited to, template matching, target tracking, target recognition and the like, and comprehensive application of the methods. The key points of the human body are specific points selected from the tremor-prone parts, including but not limited to the middle point of the upper lip, the middle point of the lower lip, the mandibular tip, the wrist joint, the finger tip, the ankle joint, the toe tip and the like which are easy to tremor.
The coordinate position curve is a curve formed by tracking the coordinates of the same human body key point in all video frames. The curve generally has a plurality of directional dimensions including at least one of an x-direction curve, a y-direction curve, a z-direction curve.
For example, a template matching method may be adopted, a preset human body key part model such as a face contour model, a mouth contour model, a hand contour model and the like is used for matching with a video image of a corresponding part, and coordinates of a point with the highest matching degree with a corresponding human body key point in the human body key part model in the video image are extracted as coordinates of the human body key point.
The method can also adopt a target tracking method, the coordinates of the human body key points marked in each subsequent frame of image are calculated by manually marking the coordinate positions of the human body key points in the first frame of image of the video image, extracting the image characteristics of the human body key points and the neighborhood thereof in the first frame of image, extracting the corresponding image characteristics in each subsequent frame of image and applying a target tracking algorithm. The image features include, but are not limited to, image templates, frequency domain information, hessian matrices, edge contour information, gray histograms, integral histograms, harris corner points, SIFT feature points, SURF feature points, and the like. The target tracking algorithms include, but are not limited to, sliding window template matching, mean shift, particle filtering, SIFT algorithm, SURF algorithm, and the like.
The target recognition method can also be adopted, the artificial neural network is trained through a sample image which is marked with some human body key point coordinates in advance, so that a trained artificial neural network model is obtained, the video image is processed frame by frame, and the human body key point coordinates in each frame are recognized.
In general, the coordinate positions of the human body key points extracted based on the template matching method or the target recognition method may slightly fluctuate between different image frames, while the coordinate positions of the human body key points extracted based on the target tracking method are more accurate and stable between consecutive frames, but tracking drift or loss may occur due to overlarge local deformation, so that in practice, the above extraction methods often need to be comprehensively applied.
For example, in one embodiment of the application, a patient's mandibular tip keypoint is identified in a first frame of an image of a video image by a target recognition method, and then tracked in a subsequent frame by a target tracking method. And when the confidence coefficient of the following algorithm in the tracking process is lower than a set threshold value, re-identifying the key point by using a target identification method, and then re-turning to the tracking process, and circulating in this way. Thus, a more accurate and stable coordinate position curve can be obtained without manual labeling.
Fig. 2 shows a schematic diagram of a coordinate position of the mandibular tip of a typical parkinson's disease patient face in accordance with an embodiment of the present application. As shown in fig. 2, the present embodiment extracts a coordinate position curve of the mandibular tip of a typical parkinson's disease patient face from a common RGB video image. Since no depth information is introduced in this embodiment, only two dimensions of x and y exist, and if an imaging device with depth information is used for video image acquisition in another embodiment of the present application, additional dimension information in the z direction can be provided. As can be seen from fig. 2, the tremor amplitude of the patient cannot be obtained simply and intuitively from the difference between the maximum and minimum of the coordinate position curve due to the slight macroscopic motion of the patient; in addition, tremor frequencies are difficult to calculate simply and intuitively, and thus require the subsequent processing steps provided by embodiments of the present application to perform more accurate calculations.
And step S3, carrying out frequency domain transformation on the coordinate position curve obtained in the step S2 so as to extract the tremor frequency of the stationary tremor of the patient. The step S3 specifically comprises the following steps:
step S3-1, performing Discrete Fourier transform (Discrete FourierTransform, DFT) or fast Fourier transform (FastFourierTransform, FFT) on each direction dimension of the coordinate position curve to obtain a frequency domain sequence of each direction dimension, wherein the transformation formula is as follows (1):
F x (n)＝DFT(x(t)),F y (n)＝DFT(y(t)),F z (n)＝DFT(z(t)) (1)
wherein x (t), y (t) and z (t) respectively represent three directional dimensions of a coordinate position curve, t=1, …, N represents a time sampling point, n=1, …, N represents a frequency domain sampling point, N represents a total frame number of a video image, DFT () represents a discrete fourier transform or a fast fourier transform, and F () x (n)、F y (n) and F z (n) represents the frequency domain sequences of three directional dimensions, respectively.
And S3-2, converting the frequency domain sequence of each direction dimension into a frequency domain amplitude curve in Hz, wherein the conversion formula is as follows in the formula (2):
wherein f represents a frequency and satisfies f=n·f ps /N,n≤N/2，f ps Representing the frame rate of the video image, abs () represents the absolute value taking operation, F represents the frequency domain sequence of either direction dimension in equation (1),representing a frequency domain amplitude curve.
Step S3-3, respectively determining frequency points with maximum absolute values of frequency domain amplitude curves of all direction dimensions in a frequency band not smaller than 4Hz, and taking the frequency points as direction tremor frequencies of the direction dimensions, wherein the calculation method is as follows formula (3):
wherein f max Representing the directional tremor frequency.
Step S3-4, maximum value of the tremor frequency in directionAnd taking the direction tremor frequency corresponding to the direction dimension with the largest frequency domain tremor amplitude as the tremor frequency of the stationary tremor of the patient as the frequency domain tremor amplitude of the frequency domain tremor in the direction dimension.
Fig. 3 shows a frequency domain amplitude plot in Hz according to an embodiment of the application. The frequency domain amplitude curve as shown in fig. 3 is obtained from the coordinate position curve in fig. 2 through steps S3-1 and S3-2. As can be seen from fig. 3, there is a significant peak at 5Hz after removing the low frequency part below 4Hz (removing the low frequency component below 4Hz is because typical tremor frequencies of dyskinesias such as parkinson' S disease and essential tremor disease are all above 4 Hz), so that tremor frequencies in both x and y directions extracted through step S3-3 are 5Hz, and at the same time, tremor amplitude in the frequency domain in the y direction is greater than that in the x direction, so that tremor frequency of the patient obtained through steps S3-3 and S3-4 is 5Hz.
And S4, carrying out band-pass filtering on the coordinate position curve after the frequency domain transformation, and extracting the maximum pixel tremor amplitude of the static tremor of the patient. The step S4 specifically comprises the following steps:
and S4-1, selecting a key frequency segment, selecting a filtering method and a band-pass filter, and respectively filtering the frequency domain sequences of each direction dimension of the coordinate position curve obtained in the step S3-1 to obtain a filtered curve of each direction dimension.
Wherein, the selection of the key frequency segment includes but is not limited to: a) Selecting a frequency band with the tremor frequency of the patient as a center and a reasonable threshold (such as 1 Hz) as a radius; b) Frequency bins associated with the diagnosed condition of the patient are selected. For example, according to the "basic diagnosis and treatment guidelines for parkinson's disease (2019), the typical tremor frequency band of parkinson's disease patients is 4 to 6Hz. In one embodiment of the application, the critical frequency band is selected to be 4 to 6Hz for patients with established parkinson's disease.
The filtering method may include a frequency domain filtering method and a time domain filtering method. The band pass filter includes, but is not limited to: a frequency domain bandpass filter and a time domain finite impulse response (Finite Impulse Response, FIR) filter.
And S4-2, calculating comprehensive tremor amplitude curves of multiple direction dimensions according to the filtered curves of the direction dimensions. More specifically, the filtered curves of all direction dimensions can be summed after point-by-point squaring, then squared and multiplied by 2 to obtain a comprehensive tremor amplitude curve of multiple direction dimensions. The calculation formula is as follows formula (4):
wherein,,and->The filtered curves of the three direction dimensions are respectively, and if only one or two direction dimensions are calculated in the step S2, the filtered curves of the direction dimensions which are not calculated are filled with all 0S.
Step S4-3, determining a maximum value in the integrated tremor magnitude curve, the maximum value corresponding to the maximum pixel tremor magnitude of the patient.
FIG. 4 shows a schematic representation of a band-pass filtered coordinate position graph in accordance with an embodiment of the present application.
For example, in one embodiment of the present application, the frequency domain sequence of each direction dimension obtained after the frequency domain transformation in fig. 2 is passed through a frequency band-pass filter with a passband of 4 to 6Hz, and then is subjected to Inverse Discrete Fourier Transform (IDFT) or Inverse Fast Fourier Transform (IFFT), and then is subjected to an absolute value taking operation, so as to obtain a filtered curve in the time domain as shown in fig. 4. It can be seen that, after the filtering treatment, the low-frequency component corresponding to the slow macroscopic motion of the patient in the coordinate position curve is filtered, so that the curve with more obvious tremor characteristics in fig. 4 is obtained. Meanwhile, since the time points at which the maximum amplitude occurs in each direction dimension may be different, the maximum amplitude cannot be simply calculated in each direction dimension and then superimposed, and frame-by-frame calculation is required in steps S4-2 and S4-3.
Fig. 5 shows a schematic diagram of an integrated tremor magnitude curve in accordance with an embodiment of the present application.
In one embodiment of the present application, the integrated tremor magnitude curve obtained after the filtered curve shown in fig. 4 is processed in step S4-2 is shown in fig. 5, and the maximum point marked with a circle in fig. 5 corresponds to the maximum pixel tremor magnitude determined in step S4-3.
And S5, acquiring the actual human body characteristic length of the patient and the pixel length corresponding to the actual human body characteristic length, and scaling the maximum actual tremor amplitude of the patient. The step S5 specifically comprises the following steps:
and S5-1, measuring the actual length between two human body key points which are approximate to the distance between the part and the imaging equipment according to the tremor prone part shot by the video image, wherein the distance between the two human body key points is basically unchanged with tremor movement. For example, in one embodiment of the present application, the tremor prone part photographed by the video image is a face, tremor of the face is mainly concentrated on the lower jaw or lower lip, in the case that the face of the patient is photographed against the imaging device, the distance from the pupil of the patient to the imaging device is substantially the same as the distance from the lower jaw or lower lip to the imaging device, and the interpupillary distance is not substantially changed with tremor movement, so that the interpupillary distance of the patient can be measured as the actual human body characteristic length.
And S5-2, obtaining the pixel distance between two corresponding human body key points in the video image. For example, in one embodiment of the present application, if the actual human feature length measured in step S5-1 is the pupil distance, the pixel distance between the two pupil centers is acquired in the video image. The acquisition method comprises the steps of manually calibrating in a picture or automatically detecting pixel coordinate positions of pupils through a deep learning algorithm, and then calculating pixel distances between the pixel coordinate positions;
step S5-3, scaling the maximum actual tremor amplitude of the patient according to the following formula (5):
wherein,,for the maximum pixel tremor amplitude obtained in step S4, L act For the actual human body characteristic length L pix For the pixel distance in the video image, +.>Is the maximum actual tremor amplitude.
For example, in one embodiment of the present application, the pupil distance of the patient is measured to be 60 mm, the pixel distance of the pupil in the video image is 50 pixels, the maximum pixel tremor amplitude of the patient is 8.2 pixels as shown in fig. 5, and the maximum actual tremor amplitude is calculated by the formula (5)Is 8.2X10/50.about.9.8 mm.
Through the steps S1 to S5 described above, quantitative indexes such as the amplitude and frequency of tremors can be obtained. These indicators can be used for quantitative assessment of resting tremor of dyskinesia. The tremors of dyskinesia, such as parkinsonism, are usually at a frequency of 4-8 times/second, typically slightly slower than simple tremors, and at a slightly greater magnitude than actionable tremors. This feature can also help us distinguish between other diseases such as those caused by chorea, cerebellar disorders, and also hyperthyroidism.
The above-described method for quantitatively evaluating stationary tremor of movement disorder according to the embodiment of the present application is described in detail, and accordingly, in order to facilitate better implementation of the above-described aspects of the embodiment of the present application, a system device of the embodiment of the present application is provided below.
Referring to fig. 6, fig. 6 is a schematic structural diagram of a resting tremor quantification assessment system of dyskinesia according to an embodiment of the present application, which may include: the video image acquisition module is configured to acquire a video image of stationary tremor of a tremor prone part of a patient through the imaging equipment; the coordinate position curve forming module is configured to extract coordinates of key points of a human body in each frame of the video image and form a coordinate position curve of the key points of the human body; the tremor frequency extraction module is configured to perform frequency domain transformation on the coordinate position curve and extract tremor frequency of stationary tremor of a patient; the pixel tremor amplitude extraction module is configured to carry out band-pass filtering on the coordinate position curve after the frequency domain transformation, and extract the maximum pixel tremor amplitude of the stationary tremor of the patient; and the actual tremor amplitude conversion module is configured to acquire the actual human body characteristic length of the patient and the pixel length in the video image corresponding to the actual human body characteristic length, and convert the maximum actual tremor amplitude of the patient according to the proportion.
Preferably, the imaging device is kept stationary during video image acquisition and the distance from the imaging device to the patient is kept fixed.
Preferably, the coordinate position curve is a curve formed by coordinates of the same human body key point in all frames of the video image, wherein the coordinate position curve comprises at least one of the following directional dimension curves: an x-direction curve, a y-direction curve, a z-direction curve.
Preferably, the tremor frequency extraction module is configured to further:
performing discrete Fourier transform or fast Fourier transform on each direction dimension of the coordinate position curve to obtain a frequency domain sequence of each direction dimension;
converting the frequency domain sequence of each direction dimension into a frequency domain amplitude curve;
respectively determining frequency points with maximum absolute values of the frequency domain amplitude curves of all direction dimensions in a frequency band not less than 4Hz as direction tremble frequencies of the direction dimensions;
maximum of tremor frequency in the directionAnd taking the direction tremor frequency corresponding to the direction dimension with the largest frequency domain tremor amplitude as the tremor frequency of the patient as the frequency domain tremor amplitude of the direction dimension.
Preferably, the pixel tremor amplitude extraction module is configured to further:
selecting a key frequency segment, selecting a filtering method and a band-pass filter, and respectively filtering the frequency domain sequences of each direction dimension of the coordinate position curve to obtain a filtered curve of each direction dimension;
according to the filtered curves of all the direction dimensions, calculating comprehensive tremor amplitude curves of a plurality of the direction dimensions;
determining a maximum in the integrated tremor magnitude curve, the maximum corresponding to the maximum pixel tremor magnitude of the patient.
Preferably, the selecting the key frequency band includes: a) Selecting a frequency band with the tremor frequency of the patient as a center and a reasonable threshold value as a radius; b) Frequency bins are selected that are associated with the patient diagnosed with the condition.
Preferably, the actual tremor magnitude scaling module is configured to further:
measuring the actual length between two human body key points, the distance of which is not basically changed along with tremor movement, on the part according to the tremor-prone part shot in the video image;
acquiring the pixel distance between two human body key points in the video image;
and scaling the maximum actual tremor amplitude of the patient according to the maximum pixel tremor amplitude, the actual human body characteristic length and the pixel distance.
Compared with the prior art, the embodiment of the application has the following beneficial effects: the technical scheme of the application has the advantages of convenience, accuracy and non-contact, provides a brand new and convenient solution for quantitative evaluation of the current stationary tremor symptoms of the movement disorder disease, and is convenient to use in remote inquiry.
The foregoing description of the embodiments of the application has been presented for the purposes of illustration and description. The foregoing description is not intended to be exhaustive or to limit the application to the precise steps and/or forms disclosed, and obviously many modifications and variations are possible in light of the above teachings, within the scope of the claims by those skilled in the art. The scope of the application and all equivalents are intended to be defined by the appended claims.
Claims (14)
1. A method for quantitative assessment of resting tremor of dyskinesia, the method comprising the steps of:
s1: collecting a video image of stationary tremor at a tremor-prone part of a patient through imaging equipment;
s2: extracting coordinates of key points of a human body from each frame of the video image to form a coordinate position curve of the key points of the human body;
s3: performing frequency domain transformation on the coordinate position curve, and extracting tremor frequency of stationary tremor of the patient;
s4: band-pass filtering the coordinate position curve after the frequency domain transformation, and extracting the maximum pixel tremor amplitude of the static tremor of the patient; and
s5: the method comprises the steps of obtaining the actual human body characteristic length of a patient and the pixel length corresponding to the actual human body characteristic length, and scaling the maximum actual tremor amplitude of the patient.
2. The method for quantitative assessment of stationary tremor of dyskinesia according to claim 1, wherein the imaging device is kept stationary during the acquisition in step S1 and the distance of the imaging device from the patient is kept fixed.
3. The method according to claim 1, wherein the coordinate position curve in step S2 is a curve formed by coordinates of the same human keypoint in all frames of the video image, wherein the coordinate position curve includes at least one of the following directional dimensions: an x-direction curve, a y-direction curve, a z-direction curve.
4. The method for quantitative assessment of resting tremor of dyskinesia according to claim 1, wherein step S3 comprises the steps of:
s3-1: performing discrete Fourier transform or fast Fourier transform on each direction dimension of the coordinate position curve to obtain a frequency domain sequence of each direction dimension;
s3-2: converting the frequency domain sequence of each direction dimension into a frequency domain amplitude curve;
s3-3: respectively determining frequency points with maximum absolute values of the frequency domain amplitude curves of all direction dimensions in a frequency band not less than 4Hz as direction tremble frequencies of the direction dimensions;
s3-4: maximum of tremor frequency in the directionAnd taking the direction tremor frequency corresponding to the direction dimension with the largest frequency domain tremor amplitude as the tremor frequency of the patient as the frequency domain tremor amplitude of the direction dimension.
5. The method for quantitative assessment of resting tremor of dyskinesia according to claim 1, wherein step S4 comprises the steps of:
s4-1: selecting a key frequency segment, selecting a filtering method and a band-pass filter, and respectively filtering the frequency domain sequences of each direction dimension of the coordinate position curve to obtain a filtered curve of each direction dimension;
s4-2: according to the filtered curves of all the direction dimensions, calculating comprehensive tremor amplitude curves of a plurality of the direction dimensions;
s4-3, determining a maximum value in the comprehensive tremor amplitude curve, wherein the maximum value corresponds to the maximum pixel tremor amplitude of the patient.
6. The method for quantitative assessment of resting tremor of dyskinesia according to claim 5, wherein the selected critical frequency band in step S4-1 comprises: a) Selecting a frequency band with the tremor frequency of the patient as a center and a reasonable threshold value as a radius; b) Frequency bins are selected that are associated with the patient diagnosed with the condition.
7. The method for quantitative assessment of resting tremor of dyskinesia according to claim 1, wherein step S5 comprises the steps of:
s5-1: measuring the actual length between two human body key points, the distance of which is not basically changed along with tremor movement, on the part according to the tremor-prone part shot in the video image;
s5-2: acquiring the pixel distance between two human body key points in the video image; and
s5-3: and scaling the maximum actual tremor amplitude of the patient according to the maximum pixel tremor amplitude, the actual human body characteristic length and the pixel distance.
8. A system for quantitative assessment of resting tremor of a movement disorder, the system comprising:
the video image acquisition module is configured to acquire a video image of stationary tremor of a tremor prone part of a patient through the imaging equipment;
the coordinate position curve forming module is configured to extract coordinates of key points of a human body in each frame of the video image and form a coordinate position curve of the key points of the human body;
the tremor frequency extraction module is configured to perform frequency domain transformation on the coordinate position curve and extract tremor frequency of stationary tremor of a patient;
the pixel tremor amplitude extraction module is configured to carry out band-pass filtering on the coordinate position curve after the frequency domain transformation, and extract the maximum pixel tremor amplitude of the stationary tremor of the patient; and
the actual tremor amplitude conversion module is configured to acquire the actual human body characteristic length of the patient and the pixel length in the video image corresponding to the actual human body characteristic length, and convert the maximum actual tremor amplitude of the patient according to the proportion.
9. The quantitative assessment system for resting tremor of dyskinesia according to claim 8, wherein the imaging device is kept stationary and the distance from the imaging device to the patient is kept fixed during video image acquisition.
10. The system of claim 8, wherein the coordinate position curve is a curve formed by coordinates of the same human keypoint in frames of all the video images, wherein the coordinate position curve comprises at least one of the following directional dimensions: an x-direction curve, a y-direction curve, a z-direction curve.
11. The kinetic disorder stationary tremor quantitative assessment system of claim 8, wherein the tremor frequency extraction module is configured to further:
performing discrete Fourier transform or fast Fourier transform on each direction dimension of the coordinate position curve to obtain a frequency domain sequence of each direction dimension;
converting the frequency domain sequence of each direction dimension into a frequency domain amplitude curve;
respectively determining frequency points with maximum absolute values of the frequency domain amplitude curves of all direction dimensions in a frequency band not less than 4Hz as direction tremble frequencies of the direction dimensions;
maximum of tremor frequency in the directionAnd taking the direction tremor frequency corresponding to the direction dimension with the largest frequency domain tremor amplitude as the tremor frequency of the patient as the frequency domain tremor amplitude of the direction dimension.
12. The system for quantitative assessment of resting tremor of dyskinesia according to claim 8, wherein the pixel tremor magnitude extraction module is configured to further:
selecting a key frequency segment, selecting a filtering method and a band-pass filter, and respectively filtering the frequency domain sequences of each direction dimension of the coordinate position curve to obtain a filtered curve of each direction dimension;
according to the filtered curves of all the direction dimensions, calculating comprehensive tremor amplitude curves of a plurality of the direction dimensions;
determining a maximum in the integrated tremor magnitude curve, the maximum corresponding to the maximum pixel tremor magnitude of the patient.
13. The system for quantitative assessment of resting tremor of dyskinesia according to claim 12, wherein the selecting key frequency bins comprises: a) Selecting a frequency band with the tremor frequency of the patient as a center and a reasonable threshold value as a radius; b) Frequency bins are selected that are associated with the patient diagnosed with the condition.
14. The system for quantitative assessment of resting tremor of dyskinesia according to claim 8, wherein the actual tremor magnitude scaling module is configured to further:
measuring the actual length between two human body key points, the distance of which is not basically changed along with tremor movement, on the part according to the tremor-prone part shot in the video image;
acquiring the pixel distance between two human body key points in the video image;
and scaling the maximum actual tremor amplitude of the patient according to the maximum pixel tremor amplitude, the actual human body characteristic length and the pixel distance.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202220667737 | 2022-03-24 | ||
| CN2022206677379 | 2022-03-24 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116831519A true CN116831519A (en) | 2023-10-03 |
Family
ID=88099802
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210313352.7A Pending CN116831519A (en) | 2022-03-24 | 2022-03-28 | Quantitative assessment method and system for stationary tremor of dyskinesia |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN116831519A (en) |
| WO (1) | WO2023179218A1 (en) |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5293879A (en) * | 1991-09-23 | 1994-03-15 | Vitatron Medical, B.V. | System an method for detecting tremors such as those which result from parkinson's disease |
| US20130338539A1 (en) * | 2012-06-14 | 2013-12-19 | International Business Machines Corporation | Software program for monitoring a hand tremor of an end-user via a computer mouse input device |
| US20210244316A1 (en) * | 2020-02-11 | 2021-08-12 | Prince Mohammad Bin Fahd University | Smart watch for tremor monitoring and control |
| CN116234493A (en) * | 2020-08-03 | 2023-06-06 | 陀螺仪装置有限公司 | Systems and methods for tremor management |
| CN112674762A (en) * | 2020-12-28 | 2021-04-20 | 江苏省省级机关医院 | Parkinson tremble evaluation device based on wearable inertial sensor |
| RU2764568C1 (en) * | 2021-04-05 | 2022-01-18 | Автономная некоммерческая образовательная организация высшего образования «Сколковский институт науки и технологий» | Method for diagnosing parkinson's disease based on video data analysis using machine learning |
-
2022
- 2022-03-28 CN CN202210313352.7A patent/CN116831519A/en active Pending
-
2023
- 2023-02-07 WO PCT/CN2023/074823 patent/WO2023179218A1/en unknown
Also Published As
| Publication number | Publication date |
|---|---|
| WO2023179218A1 (en) | 2023-09-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9996739B2 (en) | System and method for automatic gait cycle segmentation | |
| Tasli et al. | Remote PPG based vital sign measurement using adaptive facial regions | |
| EP2960862B1 (en) | A method for stabilizing vital sign measurements using parametric facial appearance models via remote sensors | |
| CN107616798B (en) | Gait asymmetry detection method based on gravity acceleration | |
| CN109993068B (en) | Non-contact human emotion recognition method based on heart rate and facial features | |
| US11663845B2 (en) | Method and apparatus for privacy protected assessment of movement disorder video recordings | |
| CN111933275A (en) | Depression evaluation system based on eye movement and facial expression | |
| CN112200162B (en) | Non-contact heart rate measuring method, system and device based on end-to-end network | |
| CN110338759B (en) | Facial pain expression data acquisition method | |
| Zhu et al. | A computer vision-based system for stride length estimation using a mobile phone camera | |
| Li et al. | The sensitive and efficient detection of quadriceps muscle thickness changes in cross-sectional plane using ultrasonography: a feasibility investigation | |
| CN110313900A (en) | It is a kind of based on deep learning to the classification method of knee joint swinging signal | |
| KR101996027B1 (en) | Method and system for extracting Heart Information of Frequency domain by using pupil size variation | |
| Banzi et al. | An automated tool for non-contact, real time early detection of diabetes by computer vision | |
| CN115147769A (en) | Physiological parameter robustness detection method based on infrared video | |
| Lee et al. | Video analysis of human gait and posture to determine neurological disorders | |
| CN115054250A (en) | Image overflow motion detection analysis method and system | |
| Zhang et al. | Computer identification and quantification of fissured tongue diagnosis | |
| CN110801227B (en) | Method and system for testing three-dimensional color block obstacle based on wearable equipment | |
| Lahmiri et al. | Combined partial differential equation filtering and particle swarm optimization for noisy biomedical image segmentation | |
| CN116831519A (en) | Quantitative assessment method and system for stationary tremor of dyskinesia | |
| Kumar et al. | Performance improvement using an automation system for segmentation of multiple parametric features based on human footprint | |
| Tan et al. | A real-time image analysis system for computer-assisted diagnosis of neurological disorders | |
| CN113807323A (en) | Accurate hand function evaluation system and method based on image recognition | |
| CN109446871B (en) | Based on it is many fitting of a polynomial model walk-show action evaluation method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |