WO2023179218A1

WO2023179218A1 - Quantitative evaluation method and system for static tremor of dyskinesia diseases

Info

Publication number: WO2023179218A1
Application number: PCT/CN2023/074823
Authority: WO
Inventors: 付思超; 董博雅; 何麒; 肖郧峰; 易典; 程耿
Original assignee: 凝动万生医疗科技(武汉)有限公司
Priority date: 2022-03-24
Filing date: 2023-02-07
Publication date: 2023-09-28
Also published as: CN116831519A

Abstract

A quantitative evaluation method and system for static tremor of dyskinesia diseases. The evaluation method comprises the following steps: S1, acquiring, by means of an imaging device, a video image of static tremor at a tremor-prone part of a patient; S2, extracting coordinates of key points of the 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 a tremor frequency of static tremor of the patient; S4, performing band-pass filtering on the coordinate position curve which is already subjected to frequency domain transformation, and extracting a maximum-pixel tremor amplitude of static tremor of the patient; and S5, obtaining an actual human body feature length of the patient and a pixel length corresponding to the actual human body feature length, and converting, according to the proportion, the actual human body feature length and the pixel length corresponding to the same into a maximum actual tremor amplitude of the patient. The method has the advantages of convenience, accuracy, and non-contact property, and therefore is particularly convenient to use in remote inquiry.

Description

Method and system for quantitative assessment of resting tremor in movement disorders

This application claims priority to Chinese Patent Application No. 202220667737.9 submitted on March 24, 2022. The disclosure of the above-mentioned Chinese Patent Application is hereby cited in its entirety as part of this application.

Technical field

This application relates to the field of quantitative assessment of movement disorders, and in particular to methods and systems for quantitative assessment of resting tremor in movement disorders.

Background technique

Resting tremor is one of the main clinical symptoms of movement disorders such as Parkinson's disease and essential tremor. Quantitative indicators such as whether tremor occurs, the amplitude and frequency of tremor are important references for the clinical diagnosis, evaluation and grading of such diseases. significance.

At present, the diagnostic evaluation of resting tremor is mainly carried out through doctors' visual scale scoring, electromyography examination, wearable sensors and other methods. The visual scale scoring mechanism relies heavily on doctors' clinical experience and is highly subjective and differentiated. Therefore, in clinical work, it is often seen that different doctors have large differences in their diagnosis, treatment, and prognosis judgments for the same patient. Electromyography examinations often use needle electrodes and apply electrical stimulation technology. The subjects will suffer certain pain and injury during the examination. Wearable sensors require a series of operating procedures such as putting on, taking off, charging, and disinfecting during use, and the operation is not convenient enough. Chinese patent CN105701806B provides a method for detecting Parkinson's tremor motion characteristics based on depth images, but the use requires the patient to wear solid-colored gloves and use a specific device (depth camera) to take pictures. The operation is not convenient enough, and it can only capture the stillness of the hand. It cannot be used to evaluate sexual tremors and cannot be applied to other parts of the body such as the face, legs, etc.

In short, it is difficult to objectively and quantitatively assess resting tremor symptoms in a convenient, non-contact manner with existing technology.

The information included in this Background section of the specification, including any references cited herein and any descriptions or discussions thereof, is included for technical reference purposes only and is not to be considered limiting. subject matter within the scope of the invention.

Contents of the invention

The present invention has been proposed in view of the above and other concepts.

In view of the shortcomings of the existing technology, according to one aspect of the present invention, it is intended to provide a quantitative assessment method for resting tremor symptoms that is easy to operate, non-contact, and has objective results.

According to one aspect of the present application, a method for quantitative assessment of resting tremor in movement disorders is provided, which can objectively and quantitatively assess resting tremor in a convenient and non-wearable manner. The method includes the following steps:

S1: Collect video images of resting tremor in the patient's tremor-prone parts through imaging equipment;

S2: Extract the coordinates of key points of the human body in each frame of the video image to form a coordinate position curve of the key points of the human body;

S3: Perform frequency domain transformation on the coordinate position curve to extract the tremor frequency of the patient's resting tremor;

S4: Perform band-pass filtering on the coordinate position curve after frequency domain transformation, and extract the maximum pixel tremor amplitude of the patient's resting tremor;

S5: Obtain the patient's actual human body feature length and the pixel length corresponding to the actual human body feature length, and convert the patient's maximum actual tremor amplitude in proportion.

Preferably, the imaging device needs to be kept stationary during the acquisition process in step S1, and the distance between the imaging device and the patient remains fixed.

Preferably, the coordinate position curve in step S2 is a curve formed by the coordinates of the same key point of the human body in all the frames of the video image, wherein the coordinate position curve includes the following directional dimension curves: At least one of: x-direction curve, y-direction curve, and z-direction curve.

More specifically, the step S3 includes the following steps:

S3-1: Perform 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: Convert the frequency domain sequence in each direction dimension into a frequency domain amplitude curve;

S3-3: Determine the frequency point with the largest absolute value of the frequency domain amplitude curve in each direction dimension in a frequency band of not less than 4 Hz, as the directional tremor frequency of that direction dimension;

S3-4: Maximum value of tremor frequency in the stated direction As the frequency domain tremor amplitude in this direction dimension, the directional tremor frequency corresponding to the direction dimension with the largest frequency domain tremor amplitude is used as The patient's tremor frequency.

Preferably, the step S4 specifically includes the following steps:

S4-1: Select a key frequency segment, select a filtering method and a band-pass filter, filter the frequency domain sequences in each direction dimension of the coordinate position curve, and obtain filtered curves in each direction dimension;

S4-2: Calculate comprehensive tremor amplitude curves in multiple direction dimensions based on the filtered curves in each direction dimension;

S4-3: Determine the maximum value in the comprehensive tremor amplitude curve, where the maximum value corresponds to the maximum pixel tremor amplitude of the patient.

Preferably, the selection of key frequency segments in step S4-1 includes: a) selecting a frequency segment with the patient's tremor frequency as the center and a reasonable threshold as the radius; b) selecting a frequency segment related to the patient's diagnosed disease frequency range.

More specifically, the step S5 includes the following steps:

S5-1: According to the tremor-prone part captured in the video image, measure the actual length between two key points on the human body whose distances basically do not change with the tremor movement on this part;

S5-2: Obtain the pixel distance between the two key points of the human body in the video image;

S5-3: Convert the maximum actual tremor amplitude of the patient in proportion 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, a resting tremor quantitative assessment system for movement disorders is provided, and the system includes:

A video image collection module configured to collect video images of resting tremor in tremor-prone parts of the patient through an imaging device;

A coordinate position curve forming module configured to extract the coordinates of key points of the human body in each frame of the video image and form a coordinate position curve of the key points of the human body;

A tremor frequency extraction module configured to perform frequency domain transformation on the coordinate position curve and extract the tremor frequency of the patient's resting tremor;

a pixel tremor amplitude extraction module configured to perform band-pass filtering on the coordinate position curve after frequency domain transformation, and extract the maximum pixel tremor amplitude of the patient's resting tremor; and

The actual tremor amplitude conversion module is configured to obtain the patient's actual human body feature length and the pixel length in the video image corresponding to the actual human body feature length, and convert the patient's maximum real body feature length in proportion. tremor amplitude.

Preferably, during the video image collection process, the imaging device needs to be kept stationary, and the distance between the imaging device and the patient remains fixed.

Preferably, the coordinate position curve is a curve formed by the coordinates of the same key point of the human body in all frames of the video image, wherein the coordinate position curve includes at least one of the following directional and dimension curves: x direction curve, y-direction curve, z-direction curve.

Preferably, the tremor frequency extraction module is configured to also be used for:

Perform discrete Fourier transform or fast Fourier transform on each direction dimension of the coordinate position curve, respectively, to obtain a frequency domain sequence of each direction dimension;

Convert the frequency domain sequence in each direction dimension into a frequency domain amplitude curve;

Determine the frequency point with the largest absolute value of the frequency domain amplitude curve in each direction dimension in a frequency band of not less than 4 Hz, respectively, as the directional tremor frequency of that direction dimension;

The maximum value of the tremor frequency in the stated direction As the frequency domain tremor amplitude in this direction dimension, the directional tremor frequency corresponding to the direction dimension with the largest frequency domain tremor amplitude is used as the patient's tremor frequency.

Preferably, the pixel tremor amplitude extraction module is configured to also be used for:

Select a key frequency segment, select a filtering method and a band-pass filter, filter the frequency domain sequences of each direction dimension of the coordinate position curve respectively, and obtain filtered curves of each direction dimension;

Calculate comprehensive tremor amplitude curves in multiple direction dimensions according to the filtered curves in each direction dimension;

The maximum value in the comprehensive tremor amplitude curve is determined, and the maximum value corresponds to the maximum pixel tremor amplitude of the patient.

Preferably, the selection of the key frequency segment includes: a) selecting a frequency segment with the patient's tremor frequency as the center and a reasonable threshold as the radius; b) selecting a frequency segment related to the patient's diagnosed disease.

Preferably, the actual tremor amplitude conversion module is configured to also be used for:

According to the tremor-prone part captured in the video image, measure the actual length between two key points on the human body whose distance basically does not change with the tremor movement on this part;

Obtain the pixel distance between two key points of the human body in the video image;

According to the maximum pixel tremor amplitude, the actual human body feature length and the pixel distance, Scale the patient's stated maximum actual tremor amplitude.

Compared with the prior art, the embodiments of the present application have the following beneficial effects:

The embodiments of the present application have the advantages of convenience, accuracy, and non-contact, provide a new and convenient solution for the quantitative assessment of resting tremor symptoms in current movement disorders, and are easy to use in remote consultation.

More embodiments of the present invention can also achieve other advantageous technical effects not listed one by one. These other technical effects may be partially described below, and for those skilled in the art after reading the present invention, they can expected and understood.

Description of the drawings

The above features and advantages and other features and advantages of these embodiments, as well as the manner of achieving them, will be more apparent, and embodiments of the invention may be better understood, by referring to the following description taken in conjunction with the accompanying drawings, in which:

Figure 1 is a schematic diagram of the flow of a quantitative assessment method for resting tremor in movement disorders according to an embodiment of the present application;

Figure 2 is a schematic diagram of the coordinate position curve of the mandibular tip on the face of a typical Parkinson's disease patient according to an embodiment of the present application;

Figure 3 is a schematic diagram of a frequency domain amplitude curve in Hz according to an embodiment of the present application;

Figure 4 is a schematic diagram of a coordinate position curve after band-pass filtering according to an embodiment of the present application;

Figure 5 is a schematic diagram of a comprehensive tremor amplitude curve according to an embodiment of the present application; and

Figure 6 is a schematic structural diagram of a resting tremor quantitative assessment system for movement disorders according to an embodiment of the present application.

Detailed ways

The details of one or more embodiments of the invention are set forth in the following description of the drawings and the detailed description. Other features, objects and advantages of the invention will be apparent from the description, drawings and claims.

It is to be understood that the illustrated and described embodiments are not limited in application to the details of construction and arrangement of components set forth in the following description or illustrated in the drawings. The illustrated embodiments may be other implementations Embodiments and can be implemented or performed in various ways. Each example is provided by way of explanation of the disclosed embodiments, not limitation. In fact, it will be apparent to those skilled in the art that various modifications and variations can be made in the various embodiments of the present invention without departing from the scope or spirit of the invention as disclosed. For example, features illustrated or described as part of one embodiment can be used with another embodiment to still produce further embodiments. Thus, it is intended that the present disclosure cover such modifications and variations as come within the scope of the appended claims and their equivalents.

Likewise, it is to be understood that the phrases and terminology used herein are for the purpose of description and should not be regarded as limiting. The use of "including," "including," or "having" and variations thereof herein is intended to include the items listed thereafter and their equivalents as well as additional items in an open-ended manner.

The present invention will be described in more detail below with reference to specific embodiments of the invention.

Figure 1 shows a schematic flowchart of a quantitative assessment method for resting tremor in movement disorders according to an embodiment of the present application. The quantitative assessment method for resting tremor according to this embodiment includes steps S1 to S5, specifically as follows.

Step S1: Use imaging equipment to collect video images of resting tremor in the patient's tremor-prone parts.

Imaging equipment that can be used includes, but is not limited to, ordinary monocular cameras, binocular and multi-eye cameras, depth cameras, infrared cameras and other imaging equipment commonly used in this field. Specific parts that can be quantitatively assessed for resting tremor, that is, parts that are the subject of photography or imaging, are generally tremor-prone parts, including but not limited to the patient's lips, mandible, entire face, palms, soles, entire upper limbs, and entire lower limbs and other parts of the body. The types of video images collected include but are not limited to ordinary RGB color videos, grayscale videos, color or grayscale videos with depth information, binocular and multi-eye color or grayscale videos, etc.

For example, in professional diagnosis and treatment institutions such as hospitals, more complex and accurate imaging equipment such as binocular cameras and depth cameras can be used to collect video images of multiple tremor-prone parts of patients while they are sitting still, and obtain videos with depth information. images to facilitate more precise processing and analysis. In the process of remote consultation, an ordinary mobile phone can be simply fixed and then the mobile phone's own camera can be used to collect whole-body video images of the patient while he is sitting quietly. This is more convenient for the accompanying staff to use ordinary imaging equipment to collect. video images.

In addition, during the acquisition process of video images, in order to conduct accurate quantitative evaluation later, it is generally necessary to keep the imaging equipment stationary, and the distance between the imaging equipment and the patient is required to remain fixed.

Step S2: Extract the coordinates of key points of the human body in each frame of the collected video image to form a coordinate position curve.

Among them, specific methods for extracting coordinates include but are not limited to template matching, target tracking, target recognition and other methods, as well as the comprehensive application of these methods. Key points on the human body are generally specific points selected on parts prone to tremor, including but not limited to the midpoint of the upper lip, midpoint of the lower lip, tip of the mandible, wrist joints, fingertips, ankle joints, toe tips, etc. where tremor is prone to occur .

The coordinate position curve is a curve formed by tracking the coordinates of the same human body key points in all video frames. The curve usually has multiple direction dimensions, including at least one of an x-direction curve, a y-direction curve, and a z-direction curve.

For example, a template matching method can be used to use preset models of key parts of the human body, such as facial contour models, mouth contour models, hand contour models, etc., to match the video images of the corresponding parts, and extract the corresponding parts of the video images. The coordinates of the point with the highest degree of matching of the corresponding human body key points in the human body key part model are used as the coordinates of the human body key points.

The target tracking method can also be used to manually mark the coordinate positions of key points of the human body in the first frame of the video image, extract the image features of the key points of the human body and their neighborhoods in the first frame of the image, and subsequently The corresponding image features are extracted from each frame of image, and the target tracking algorithm is applied to calculate the coordinates of the human body key points marked in each subsequent frame of image. The image features include but are not limited to image templates, frequency domain information, Hessian matrices, edge contour information, grayscale histograms, integral histograms, Harris corner points, SIFT feature points, SURF feature points, etc. The target tracking algorithm includes but is not limited to sliding window template matching, mean shift, particle filter, SIFT algorithm, SURF algorithm, etc.

The target recognition method can also be used to train the artificial neural network through sample images pre-labeled with the coordinates of certain human key points, so as to obtain the trained artificial neural network model to process the video images frame by frame and identify each The coordinates of the human body key points in the frame.

Generally, the coordinate positions of human body key points extracted based on template matching methods or target recognition methods may fluctuate slightly between different image frames, while the coordinate positions of human body key points extracted based on target tracking methods are more accurate between consecutive frames. and stable, but tracking drift or loss may occur due to excessive local deformation. Therefore, in practice, the above extraction methods often need to be applied comprehensively.

For example, in one embodiment of the present application, the key point of the patient's mandibular tip is identified through the target recognition method in the first frame of the video image, and then the key point of the patient's mandibular tip is identified through the target tracking method in subsequent frames. Key points are tracked. When the confidence of the following algorithm is lower than the set threshold during the tracking process, the target recognition method is applied to re-identify the key point, and then the tracking process is re-entered, and so on. In this way, a more accurate and stable coordinate position curve can be obtained without manual annotation.

Figure 2 shows a schematic diagram of the coordinate position curve of the mandibular tip on the face of a typical Parkinson's disease patient according to an embodiment of the present application. As shown in Figure 2, this embodiment extracts the coordinate position curve of the mandibular tip of a typical Parkinson's disease patient's face from ordinary RGB video images. Since this embodiment does not introduce depth information, there are only two dimensions of x and y. If an imaging device with depth information is used for video image collection in another embodiment of the present application, an additional z direction can be provided. Dimension information. As can be seen from Figure 2, due to the slight macroscopic movement of the patient, the patient's tremor amplitude cannot be obtained simply and intuitively through the difference between the maximum and minimum values of the coordinate position curve; in addition, the tremor frequency is also difficult to calculate simply and intuitively. Therefore, subsequent processing steps provided by the embodiments of the present application are needed to perform more accurate calculations.

Step S3: Perform frequency domain transformation on the coordinate position curve obtained in step S2 to extract the tremor frequency of the patient's resting tremor. Step S3 specifically includes the following steps:

Step S3-1: Perform Discrete Fourier Transform (DFT) or Fast Fourier Transform (FFT) on each direction dimension of the coordinate position curve to obtain frequency domain sequences of each direction dimension. 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)

Among them, x(t), y(t) and z(t) respectively represent the three directional dimensions of the coordinate position curve, t=1,...,N represents the time sampling point, n=1,...,N represents the frequency domain sampling point, N represents the total number of frames of the video image, DFT(□) represents discrete Fourier transform or fast Fourier transform, F _x (n), F _y (n) and F _z (n) represent three directions respectively. Dimensional frequency domain sequence.

Step S3-2: Convert the frequency domain sequence in each direction dimension into a frequency domain amplitude curve in Hz. The conversion formula is as follows (2):

where f represents the frequency, and satisfies f=n·f _ps /N, n≤N/2, f _ps represents the frame rate of the video image, abs(□) represents the absolute value operation, and F represents any value in formula (1) A one-dimensional frequency domain sequence, Represents the frequency domain amplitude curve.

Step S3-3: Determine the frequency domain amplitude curve of each direction dimension in a frequency band of no less than 4Hz. The frequency point with the largest absolute value within is used as the directional tremor frequency in that direction dimension. The calculation method is as follows: formula (3):

where f _max represents the directional tremor frequency.

Step S3-4, use the maximum value of the tremor frequency in the direction As the frequency domain tremor amplitude in this direction dimension, the directional tremor frequency corresponding to the direction dimension with the largest frequency domain tremor amplitude is used as the tremor frequency of the patient's resting tremor.

Figure 3 shows a schematic diagram of a frequency domain amplitude curve in Hz according to an embodiment of the present application. The frequency domain amplitude curve shown in Figure 3 is obtained from the coordinate position curve in Figure 2 through steps S3-1 and S3-2. As can be seen from Figure 3, there is an obvious peak at 5Hz after removing the low-frequency component below 4Hz (the low-frequency component below 4Hz is removed because the typical tremor frequencies of movement disorders such as Parkinson's disease and essential tremor are uniform. above 4Hz), so the directional tremor frequencies in the x and y directions extracted through step S3-3 are both 5Hz. At the same time, the frequency domain tremor amplitude in the y direction is greater than the x direction, so after steps S3-3 and S3- 4The patient's tremor frequency obtained is 5Hz.

Step S4: Band-pass filter the coordinate position curve after frequency domain transformation to extract the maximum pixel tremor amplitude of the patient's resting tremor. Step S4 specifically includes the following steps:

Step S4-1, select the key frequency range, select the filtering method and bandpass filter, filter the frequency domain sequence of each direction dimension of the coordinate position curve obtained in step S3-1, and obtain the filtered curve of each direction dimension. .

Among them, the selection of the key frequency segment includes but is not limited to: a) selecting a frequency segment with the patient's tremor frequency as the center and a reasonable threshold (such as 1 Hz) as the radius; b) selecting a frequency segment related to the patient's diagnosed disease . For example, according to the "Guidelines for Primary Diagnosis and Treatment of Parkinson's Disease (2019)", the typical tremor frequency range for patients with Parkinson's disease is 4 to 6 Hz. In one embodiment of the present application, for patients with confirmed Parkinson's disease, the key frequency range is selected to be 4 to 6 Hz.

The filtering method may include frequency domain filtering and time domain filtering methods. The bandpass filter includes but is not limited to: a frequency domain bandpass filter and a time domain finite impulse response (Finite Impulse Response, FIR) filter.

Step S4-2: Based on the filtered curves of each direction dimension, calculate the comprehensive value of multiple direction dimensions. Tremor amplitude curve. More specifically, the filtered curves in each direction dimension can be squared point by point, summed, then squared, and then multiplied by 2 to obtain a comprehensive tremor amplitude curve in multiple direction dimensions. The calculation formula is as follows (4):

in, and They are the filtered curves of the three directional dimensions respectively. If only one or two of the directional dimensions are calculated in step S2, the filtered curves of the uncalculated directional dimensions will be filled with all 0s.

Step S4-3: Determine the maximum value in the comprehensive tremor amplitude curve, where the maximum value corresponds to the maximum pixel tremor amplitude of the patient.

Figure 4 shows a schematic diagram of a coordinate position curve after band-pass filtering according to 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 frequency domain transformation in Figure 2 is passed through a frequency band-pass filter with a passband of 4 to 6 Hz, and then subjected to inverse discrete Fourier transform (IDFT) or inverse fast Fourier transform (IFFT), and then perform the absolute value operation to obtain the filtered curve in the time domain as shown in Figure 4. It can be seen that after filtering, the low-frequency component corresponding to the patient's slow macroscopic motion in the coordinate position curve is filtered out, and a curve with more obvious tremor characteristics in Figure 4 is obtained. At the same time, since the patient may have different moments of maximum amplitude in each direction dimension, we cannot simply calculate the maximum amplitude in each direction dimension and then superimpose it. Instead, we need to perform frame-by-frame calculations through steps S4-2 and S4-3.

Figure 5 shows a schematic diagram of a comprehensive tremor amplitude curve according to an embodiment of the present application.

In one embodiment of the present application, the comprehensive tremor amplitude curve obtained after processing the filtered curve shown in Figure 4 in step S4-2 is shown in Figure 5. The maximum value point marked with a circle in Figure 5 corresponds to The maximum pixel tremor amplitude determined through step S4-3.

Step S5: Obtain the patient's actual human body feature length and the pixel length corresponding to the actual human body feature length, and calculate the patient's maximum actual tremor amplitude in proportion. Step S5 specifically includes the following steps:

Step S5-1: According to the tremor-prone part captured by the video image, measure the actual length between two key points on the human body that is approximately the distance from the part to the imaging device, where the distance between the two key points on the human body is measured. Basically does not change with tremor movement. For example, in one embodiment of the present application, the tremor-prone part captured by the video image is the face, and the tremor on the face is mainly concentrated in the lower jaw or For the lower lip, when the patient's face is facing the imaging device, the distance from the patient's pupil to the imaging device is basically the same as the distance from the mandible or lower lip to the imaging device, and the interpupillary distance basically does not change with tremor movement, so the patient can be measured The interpupillary distance is used as the characteristic length of the actual human body.

Step S5-2: Obtain the pixel distance between the two corresponding human body key points in the video image. For example, in one embodiment of the present application, if the actual human body characteristic length measured in step S5-1 is the interpupillary distance, the pixel distance between the two pupil centers needs to be obtained in the video image. The acquisition method includes manual calibration in the picture or automatic detection of the pixel coordinate position of the pupil through a deep learning algorithm, and then calculating the pixel distance between the pixel coordinate positions;

Step S5-3: Convert the patient's maximum actual tremor amplitude proportionally. The conversion formula is as follows (5):

in, is the maximum pixel tremor amplitude obtained in step S4, L _act is the actual human body feature length, L _pix is the pixel distance in the video image, is the maximum actual tremor amplitude.

For example, in one embodiment of the present application, the patient's interpupillary distance was measured to be 60 mm, the pixel distance of the pupils in the video image was 50 pixels, and the patient's maximum pixel tremor amplitude was 8.2 pixels as shown in Figure 5. Then the maximum actual tremor amplitude calculated by formula (5) It is 8.2×60/50≈9.8 mm.

Through the above steps S1 to S5, quantitative indicators such as the amplitude and frequency of tremor can be obtained. These indicators can be used to quantitatively evaluate resting tremor in movement disorders. The tremor frequency of movement disorders such as Parkinson's disease is usually 4-8 times/second, which is generally slightly slower and slightly larger than simple tremor, but faster and slightly smaller than action tremor. This feature can also help us distinguish other diseases, such as those caused by chorea, cerebellar disorders, and hyperthyroidism.

The above describes in detail the quantitative assessment method of resting tremor in movement disorders according to the embodiments of the present application. In order to facilitate better implementation of the above solution of the embodiments of the present application, accordingly, the system device of the embodiments of the present application is provided below.

Referring to Figure 6, Figure 6 is a schematic structural diagram of a resting tremor quantitative assessment system for movement disorders according to an embodiment of the present application. The system may include: a video image collection module configured to collect the patient's tremor-prone parts through an imaging device. A video image of static tremor; a coordinate position curve forming module configured to extract the coordinates of key points of the human body in each frame of the video image to form all The 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 to extract the tremor frequency of the patient's resting tremor; the pixel tremor amplitude extraction module is configured to perform frequency domain transformation in pairs The coordinate position curve is band-pass filtered to extract the maximum pixel tremor amplitude of the patient's resting tremor; and an actual tremor amplitude conversion module is configured to obtain the patient's actual human body feature length and the video image corresponding to the actual human body feature length. Pixel length, proportionally converted to the patient's maximum actual tremor amplitude.

The maximum actual tremor amplitude of the patient is converted in proportion according to the maximum pixel tremor amplitude, the actual human body characteristic length and the pixel distance.

Compared with the existing technology, the embodiments of the present application have the following beneficial effects: the technical solution of the present application has the advantages of convenience, accuracy, and non-contact, and provides a quantitative assessment of resting tremor symptoms in current movement disorders. A new, convenient solution that is easy to use in remote consultations.

The foregoing description of embodiments of the invention has been presented for purposes of illustration. The foregoing description is not intended to be exhaustive or to limit the invention to the precise steps and/or forms disclosed. It will be apparent that, in light of the above teachings, one skilled in the art will be able to make many modifications within the scope of the claims. and variants. The scope of the invention and all equivalents thereto are intended to be defined by the appended claims.

Claims

A quantitative assessment method for resting tremor in movement disorders, characterized in that the method includes the following steps:

S1: Collect video images of resting tremor in the patient's tremor-prone parts through imaging equipment;

S2: Extract the coordinates of key points of the human body in each frame of the video image to form a coordinate position curve of the key points of the human body;

S3: Perform frequency domain transformation on the coordinate position curve, and extract the tremor frequency of the patient's resting tremor;

S4: Perform band-pass filtering on the coordinate position curve after frequency domain transformation, and extract the maximum pixel tremor amplitude of the patient's resting tremor; and

S5: Obtain the patient's actual human body feature length and the pixel length corresponding to the actual human body feature length, and convert the patient's maximum actual tremor amplitude in proportion.
The quantitative assessment method for resting tremor in movement disorders according to claim 1, wherein the imaging device needs to be kept stationary during the acquisition process in step S1, and the distance between the imaging device and the patient should be kept fixed.
The quantitative assessment method for resting tremor in movement disorders according to claim 1, wherein the coordinate position curve in step S2 is the same key point of the human body in all the frames of the video image. A curve formed by coordinates, wherein the coordinate position curve includes at least one of the following directional dimension curves: x-direction curve, y-direction curve, and z-direction curve.
The quantitative assessment method for resting tremor in movement disorders according to claim 1, wherein step S3 specifically includes the following steps:

S3-1: Perform 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: Convert the frequency domain sequence in each direction dimension into a frequency domain amplitude curve;

S3-3: Determine the frequency point with the largest absolute value of the frequency domain amplitude curve in each direction dimension in a frequency band of not less than 4 Hz, as the directional tremor frequency of that direction dimension;

S3-4: Maximum value of tremor frequency in the stated direction As the frequency domain tremor amplitude in this direction dimension, the directional tremor frequency corresponding to the direction dimension with the largest frequency domain tremor amplitude is used as the patient's tremor frequency.
The quantitative assessment method for resting tremor in movement disorders according to claim 1, wherein step S4 specifically includes the following steps:

S4-1: Select a key frequency segment, select a filtering method and a band-pass filter, filter the frequency domain sequences in each direction dimension of the coordinate position curve, and obtain filtered curves in each direction dimension;

S4-2: Calculate comprehensive tremor amplitude curves in multiple direction dimensions based on the filtered curves in each direction dimension;

S4-3: Determine the maximum value in the comprehensive tremor amplitude curve, where the maximum value corresponds to the maximum pixel tremor amplitude of the patient.
The quantitative assessment method for resting tremor in movement disorders according to claim 5, wherein the selecting the key frequency segment in step S4-1 includes: a) selecting the patient's tremor frequency as the center , the reasonable threshold is the frequency segment of the radius; b) Select the frequency segment related to the patient's diagnosed disease.
The quantitative assessment method for resting tremor in movement disorders according to claim 1, wherein step S5 specifically includes the following steps:

S5-1: According to the tremor-prone part captured in the video image, measure the actual length between two key points on the human body whose distances basically do not change with the tremor movement on this part;

S5-2: Obtain the pixel distance between the two key points of the human body in the video image; and

S5-3: Convert the maximum actual tremor amplitude of the patient in proportion according to the maximum pixel tremor amplitude, the actual human body characteristic length and the pixel distance.
A resting tremor quantitative assessment system for movement disorders, characterized in that the system includes:

A video image collection module configured to collect video images of resting tremor in tremor-prone parts of the patient through an imaging device;

A coordinate position curve forming module configured to extract the coordinates of key points of the human body in each frame of the video image and form a coordinate position curve of the key points of the human body;

A tremor frequency extraction module configured to perform frequency domain transformation on the coordinate position curve and extract the tremor frequency of the patient's resting tremor;

a pixel tremor amplitude extraction module configured to perform band-pass filtering on the coordinate position curve after frequency domain transformation, and extract the maximum pixel tremor amplitude of the patient's resting tremor; and

The actual tremor amplitude conversion module is configured to obtain the patient's actual human body characteristic length and the pixel length in the video image corresponding to the actual human body characteristic length, and convert the patient's maximum actual tremor amplitude in proportion.
The resting tremor quantitative assessment system for movement disorders according to claim 8, characterized in that during the video image collection process, the imaging device needs to be kept stationary, and the imaging device The distance between the device and the patient remains fixed.
The resting tremor quantitative assessment system for movement disorders according to claim 8, wherein the coordinate position curve is a curve formed by the coordinates of the same key point of the human body in all the frames of the video image, wherein , the coordinate position curve includes at least one of the following directional dimension curves: x-direction curve, y-direction curve, and z-direction curve.
The quantitative assessment system for resting tremor in movement disorders according to claim 8, wherein the tremor frequency extraction module is configured to:

Perform discrete Fourier transform or fast Fourier transform on each direction dimension of the coordinate position curve, respectively, to obtain a frequency domain sequence of each direction dimension;

Convert the frequency domain sequence in each direction dimension into a frequency domain amplitude curve;

Determine the frequency point with the largest absolute value of the frequency domain amplitude curve in each direction dimension in a frequency band of not less than 4 Hz, respectively, as the directional tremor frequency of that direction dimension;

The maximum value of the tremor frequency in the stated direction As the frequency domain tremor amplitude in this direction dimension, the directional tremor frequency corresponding to the direction dimension with the largest frequency domain tremor amplitude is used as the patient's tremor frequency.
The quantitative assessment system for resting tremor in movement disorders according to claim 8, wherein the pixel tremor amplitude extraction module is configured to also be used for:

Select a key frequency segment, select a filtering method and a band-pass filter, filter the frequency domain sequences of each direction dimension of the coordinate position curve respectively, and obtain filtered curves of each direction dimension;

Calculate comprehensive tremor amplitude curves in multiple direction dimensions according to the filtered curves in each direction dimension;

The maximum value in the comprehensive tremor amplitude curve is determined, and the maximum value corresponds to the maximum pixel tremor amplitude of the patient.
The resting tremor quantitative assessment system for movement disorders according to claim 12, wherein the selecting a key frequency segment includes: a) selecting a frequency segment with the patient's tremor frequency as the center and a reasonable threshold as the radius. ; b) Select frequency segments related to the patient's diagnosed disease.
The quantitative assessment system for resting tremor in movement disorders according to claim 8, wherein the actual tremor amplitude conversion module is configured to also be used for:

According to the tremor-prone part captured in the video image, measure the actual length between two key points on the human body whose distance basically does not change with the tremor movement on this part;

Obtain the pixel distance between two key points of the human body in the video image;

According to the maximum pixel tremor amplitude, the actual human body feature length and the pixel distance, Scale the patient's stated maximum actual tremor amplitude.