CN114145722B - Pulse pathological feature mining method for pancreatitis patients - Google Patents
Pulse pathological feature mining method for pancreatitis patients Download PDFInfo
- Publication number
- CN114145722B CN114145722B CN202111484127.1A CN202111484127A CN114145722B CN 114145722 B CN114145722 B CN 114145722B CN 202111484127 A CN202111484127 A CN 202111484127A CN 114145722 B CN114145722 B CN 114145722B
- Authority
- CN
- China
- Prior art keywords
- curve
- pancreatitis
- period
- pulse signal
- pulse
- 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.)
- Active
Links
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/42—Detecting, measuring or recording for evaluating the gastrointestinal, the endocrine or the exocrine systems
- A61B5/4222—Evaluating particular parts, e.g. particular organs
- A61B5/425—Evaluating particular parts, e.g. particular organs pancreas
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/48—Other medical applications
- A61B5/4857—Indicating the phase of biorhythm
-
- 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
- A61B5/7225—Details of analog processing, e.g. isolation amplifier, gain or sensitivity adjustment, filtering, baseline or drift compensation
-
- 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
- A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
- A61B5/7267—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems involving training the classification device
-
- 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
- A61B5/7271—Specific aspects of physiological measurement analysis
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Molecular Biology (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Public Health (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Medical Informatics (AREA)
- Pathology (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- Veterinary Medicine (AREA)
- Physiology (AREA)
- Artificial Intelligence (AREA)
- Signal Processing (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Psychiatry (AREA)
- Cardiology (AREA)
- Evolutionary Computation (AREA)
- Fuzzy Systems (AREA)
- Mathematical Physics (AREA)
- Power Engineering (AREA)
- Endocrinology (AREA)
- Gastroenterology & Hepatology (AREA)
- Measuring Pulse, Heart Rate, Blood Pressure Or Blood Flow (AREA)
Abstract
The invention provides a method for mining pulse pathological characteristics of pancreatitis patients, which comprises the following steps: obtaining pulse signal curve samples marked with labels, wherein the labels are used for distinguishing the pulse signal curve of the sample as the pulse signal curve of a normal person or a pancreatitis patient, then carrying out periodic division on the signal curve, extracting basic time domain characteristics, stability indexes and double peak index characteristics to form input characteristic vectors, inputting the input characteristic vectors into a classification model, supervising the input characteristic vectors by the labels corresponding to all the samples, carrying out training, obtaining a trained classification model, and then diagnosing that the person to be diagnosed is ill. The extracted dual-peak index represents two peaks and an obvious concave arc structure in a single-period curve, and the stability index is used for representing the change rate of the latter half of the single-period curve, so that pancreatitis patients can be well pointed, case feature vectors can be provided for subsequent training classification models, and the obtained trained classification models have positive significance on classification patients.
Description
Technical Field
The invention belongs to the technical field of traditional Chinese medicine pulse condition acquisition systems, and particularly relates to a pulse pathological feature mining method for pancreatitis patients.
Background
The Chinese medicine occupies an important position in the world medical system, the Chinese pulse feeling science has two thousands of years of experience summary in the medical field, and the rich experience is accumulated, so that the Chinese medicine is also one of the most representative diagnosis modes in the Chinese medicine 'inspection, smelling, asking and cutting'. At present, pulse feeling diagnosis is mainly carried out by touching the veins of a patient by a doctor of traditional Chinese medicine, but the diagnosis mode has great influence on the diagnosis result due to different hosts and objects, and the diagnosis of pulse conditions has no quantitative data and lacks objectivity. With the development of sensors and artificial intelligence technology, computer-aided medical treatment has been widely used.
The pulse condition is accurately analyzed and diagnosed through more human body information, and the multi-mode information acquisition has important significance for pulse condition diagnosis of traditional Chinese medicine.
Blood flows through the whole body of the human body, the physiological changes of the human body can influence the changes of the blood vessel pressure in the artery of the arm, and in the traditional Chinese medicine theory, the pulse contains various information of the human body. The pulse is a quasi-periodic signal, and after the pulse signal is preprocessed and periodically divided, the single-period pulse can reflect various physiological and pathological information of human body, such as age, gender, physical condition, health condition and the like. However, at present, the traditional Chinese medical practitioner uses manual measurement to make diagnosis only by experience and pulse diagnosis, which causes uncertainty of diagnosis results.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides a pulse pathology feature mining algorithm for pancreatitis patients, which provides a feature design formula of pancreatitis patients by comparing the pulse of pancreatitis patients with the pulse of non-pancreatitis patients, wherein the feature design formula comprises a stability index and a double-peak index, and compared with the pulse of non-pancreatitis patients, the pulse of the pulse-condition pancreatitis patients has two peaks and an obvious concave arc structure, the two features can well describe the stability of pulse monocycle signals, the two pathology features are fused with basic time domain features and are used for training a classification model, so that the preliminary classification of suspected pancreatitis patients is realized, and the classification effect is good.
In order to solve the technical problems, the invention adopts the following technical scheme: a method of mining pathological features for pancreatitis patients, the method comprising:
acquiring a labeled sample pulse signal curve, wherein the label is used for distinguishing individuals corresponding to the sample pulse signal curve from non-pancreatitis patients or pancreatitis patients;
dividing the sample pulse signal curve according to the period to obtain each single-period curve;
extracting the basic time domain characteristics of each single-period curve;
extracting the stability index and the double peak index of each single-period curve as case characteristics of pancreatitis corresponding to the single-period curve, wherein the double peak index represents two peaks and an obvious concave arc structure in the single-period curve; the stability index is used for representing the change rate of the second half of the single-period curve;
the basic time domain features of pulse signal curves of all samples and the case features of pancreatitis form input feature vectors, the input feature vectors are input into a classification model, and the classification model is supervised by labels corresponding to all samples and trained, so that the trained classification model is obtained.
Preferably, the labels are divided into patient labels and non-patient labels;
the trained classification model further comprises:
acquiring a pulse signal curve of a person to be diagnosed;
in pulse signal curves of a person to be diagnosed, inputting feature vectors composed of basic time domain features of all single-period curves and case features of pancreatitis into a trained classification model to obtain a diagnosis tag set; each label in the diagnosis label set corresponds to a single-period curve one by one;
if the occurrence frequency of the patient labels in the diagnosis label set is larger than a preset threshold, determining that the person to be diagnosed is a suspected pancreatitis patient.
Preferably, the extraction of the stability index of each monocycle curve is specifically: the variance of each sampling point is obtained by selecting a plurality of sampling points in a single-period curve and calculating the variance of each sampling point by a first formula, wherein the variances of all the sampling points are used for representing the stability index of the pulse single-period signal of each single-period curve, and the first formula is as follows:
in the first formula, N represents the number of sampling points in a single-period curve, N represents the number of sampling points in the whole period of a sample pulse signal curve, S F Represents the sampling frequency, t represents the starting position of the sampling point, D i Representing the variance of the sampling point at i, where i is a positive integer from 0 to N, parameter X i Representing the absolute coordinate value, x, of the sampling point at the abscissa i i Represents the abscissa of the sampling point at i, y i Representing the ordinate of the sampling point at i.
Preferably, the sampling frequency is a value 2 times the beating frequency of the human pulse.
Preferably, extracting the peak index of each monocycle curve is specifically: selecting a plurality of sampling points in the single-period curve, calculating to obtain the average value of the ordinate values of each sampling point of the non-pancreatitis patient, and calculating the sum of differences between the ordinate values of the sampling points between two wave peaks in the single-period curve and the average value of the ordinate values of the sampling points of the same non-pancreatitis patient through a second formula, wherein the sum is used as a dual-wave peak index; the second formula is as follows:
in the second formula, Y represents the data set of the ordinate of the sampling point, N represents the number of the sampling points in the whole period of the sample pulse signal curve, N represents the number of the sampling points in the single period curve, T represents the point with the highest ordinate value in the single period pulse signal curve, T represents the point with the next highest ordinate value in the single period curve, S F Represents the sampling frequency, A represents the average value of the ordinate numerical value of each sampling point of a normal person, B i The sum of the difference between the ordinate value of the sampling point at i and A of pancreatitis patient is shown.
Preferably, the basic time domain features of each monocycle pulse signal curve include a pulse first peak height, a second peak height, a third peak height, a first peak to second peak time interval, a second peak to third peak time interval, a pulse start to first trough time interval, a first trough to second trough time interval, and a second trough to pulse start time interval.
Preferably, the pulse signal curve of the person to be diagnosed is at least one pulse signal curve of a single period.
Preferably, the classification model is trained by using an SVM kernel method.
Compared with the prior art, the invention has the following advantages:
1. based on the traditional Chinese medicine principle, the invention discovers the unique pathological characteristics of the pulse image of the pancreatitis patient compared with normal people, and in order to make the characteristics and differences be represented, two algorithms are designed for representing the two pathological characteristics of pancreatitis through data, the two pathological characteristics are a stability index and a dual peak index, and a classification model is trained by combining the basic time domain characteristics of a single period pulse signal curve, and the trained classification model is used for primarily diagnosing whether the person to be diagnosed is ill or not.
2. The invention uses a computer to detect pancreatitis based on the principle of traditional Chinese medicine for the first time, inputs the pancreatitis into a trained classification model to judge whether the person to be diagnosed is a suspected pancreatitis patient or not through the pulse signal curve of the person to be diagnosed, and the accuracy of classification is up to more than 95%, wherein the accuracy of young patients is even close to 100%. The elderly subjects will have a reduced vascular elasticity due to their older age, and the peak structure of typical pulse image features will become less pronounced, thus reducing the diagnostic accuracy compared to young patients, and the experimental model has a very good experimental effect on pancreatitis and other chronic disease detection, and has a very high accuracy.
The technical scheme of the invention is further described in detail through the drawings and the embodiments.
Drawings
FIG. 1 is a block flow diagram of training a classification model according to an embodiment of the invention.
Fig. 2 is a single cycle pulse signal plot of a normal person.
Fig. 3 is a pulse signal plot for a pancreatitis patient.
Fig. 4 is a basic time domain feature of the single cycle pulse signal curve of the present invention.
Detailed Description
The embodiment discloses a pulse pathological feature mining method for pancreatitis patients, which comprises the following steps:
s101, acquiring a labeled sample pulse signal curve, wherein the label is used for distinguishing the pulse signal curve from a non-pancreatitis patient and a pancreatitis patient, and can be defined as: the non-pancreatitis patient label is "1", and the pancreatitis patient label is "0";
the specific steps of obtaining the labeled sample pulse signal curve are as follows: collecting pulse data of a plurality of non-pancreatitis patients and a plurality of pancreatitis patients, respectively preprocessing the collected pulse data, drawing a pulse signal curve by taking the collected time as an abscissa and the pulse amplitude of the pulse as an ordinate, so as to obtain a labeled sample pulse signal curve; preprocessing pulse data is based on the prior art, such as signal filtering by using a Butterworth algorithm, removing baseline drift from the filtered signal, and processing the signal again by using a smoothing function to make the drawn pulse signal curve smooth;
s102, dividing pulse data period: the human body pulse is a quasi-periodic signal, each pulse period has similar period, amplitude and waveform, but the characteristics are different from person to person, so that pulse signal curves drawn by non-pancreatitis patients and pancreatitis patients are divided periodically to obtain each monocycle curve, and pulse curve characteristics representing a curve collector can be obtained by researching each monocycle pulse signal curve;
s103, extracting basic time domain characteristics of each monocycle curve; the basic time domain features comprise a first peak height, a second peak height, a third peak height, a first peak to second peak time interval, a second peak to third peak time interval, a pulse start point to first trough time interval, a first trough to second trough time interval and a second trough to pulse start point time interval of a single period curve;
s104, extracting a stability index of each monocycle curve, wherein the stability index is used for representing the change rate of the second half part of the monocycle curve, and has classification significance for distinguishing pancreatitis patients from non-pancreatitis patients;
s105, extracting a double wave peak index of each single period curve, wherein the double wave peak index represents two peak values and an obvious concave arc structure in the single period curve; since the normal person's bimodal index is greater than that of pancreatitis patients,
extracting the stability index and the dual peak index of each monocycle curve as case features of pancreatitis corresponding to the monocycle curve;
s106, the basic time domain features of the pulse signal curves of all samples and the case features of pancreatitis form input feature vectors, the input feature vectors are input into a classification model, the classification model is supervised by labels corresponding to all samples, and a SVM kernel method is used for training the classification model, so that a trained classification model is obtained.
In this embodiment, the non-pancreatitis patient is a healthy normal person without any disease; the pulse data is acquired by measuring the pulse of a non-pancreatitis patient or pancreatitis patient through a pulse sensor and the like, collecting and acquiring the pulse data, and uploading the acquired pulse data to a computer for disease detection research. 57 non-pancreatitis patients and 7 pancreatitis patients are selected during collection, are respectively measured in a plurality of time periods each day and are measured for a plurality of continuous days to obtain pulse data, and the pulse data is provided by the 211 rd hospital of the Chinese people's liberation army.
In this embodiment, the stability index of each monocycle curve is extracted specifically as follows: the variance of each sampling point is obtained by selecting a plurality of sampling points in a single-period curve and calculating the variance of each sampling point by a first formula, wherein the variances of all the sampling points are used for representing the stability index of the pulse single-period signal of each single-period curve, and the first formula is as follows:
in the first formula, N represents the number of sampling points in a single-period curve, N represents the number of sampling points in the whole period of a sample pulse signal curve, S F Represents the sampling frequency, t represents the starting position of the sampling point, D i Representing the variance of the sampling point at i, where i is a positive integer from 0 to N, parameter X i Representing the absolute coordinate value, x, of the sampling point at the abscissa i i Represents the abscissa of the sampling point at i, y i Representing the ordinate of the sampling point at i.
As shown in fig. 2 and 3, compared with the pulse of a normal person, the pulse condition (within the range of the considered region) of the pancreatitis patient has two peaks and an obvious concave arc structure, and the feature can well describe the stability of the pulse monocycle signal and can describe the peak number and the concave arc degree in the pulse. Therefore, in this embodiment, the extraction of the bi-peak index of each single-period curve is specifically: selecting a plurality of sampling points in the single-period curve, calculating to obtain the average value of the ordinate values of each sampling point of the non-pancreatitis patient, and calculating the sum of differences between the ordinate values of the sampling points between two wave peaks in the single-period curve and the average value of the ordinate values of the sampling points of the same non-pancreatitis patient through a second formula, wherein the sum is used as a dual-wave peak index; the second formula is as follows:
in the second formula, Y represents the data set of the ordinate of the sampling point, N represents the number of the sampling points in the whole period of the sample pulse signal curve, N represents the number of the sampling points in the single period curve, T represents the point with the highest ordinate value in the single period pulse signal curve, and T represents the single periodPoints with the next highest ordinate values in the curve S F Represents the sampling frequency, A represents the average value of the ordinate numerical value of each sampling point of a normal person, B i The sum of the difference between the ordinate value of the sampling point at i and A of pancreatitis patient is shown.
As shown in fig. 4, the basic time domain features of the monocycle pulse signal include: first peak height h of pulse from pulse start point 1 Second peak height h 2 Third peak height h 3 Time interval t from first peak to second peak a Time interval t from second peak to third peak b Time interval t from pulse start point to first trough 2 -t 1 Time interval t from first trough to second trough 3 -t 2 The time interval t from the second trough to the pulse start point 3 -t 1 。
In this embodiment, the trained classification model further includes:
s201, acquiring a pulse signal curve of a person to be diagnosed;
s202, inputting a feature vector composed of basic time domain features of all monocycle curves and case features of pancreatitis into a trained classification model in a pulse signal curve of a person to be diagnosed to obtain a diagnosis tag set; each label in the diagnosis label set corresponds to a single-period curve one by one;
and S203, if the occurrence frequency of the patient labels in the diagnosis label set is greater than a preset threshold, determining that the person to be diagnosed is a suspected pancreatitis patient.
In this embodiment, the pulse signal curve of the person to be diagnosed is a pulse signal curve of the person to be diagnosed, which is at least one pulse signal curve of a single cycle.
The trained classification model is verified by selecting the patients with confirmed diagnosis and the non-patients, and the verification result shows that the accuracy of the final diagnosis of the patients with pancreatitis by the personnel to be diagnosed which are preliminarily classified by adopting the method is more than 95 percent, especially for young patients, the accuracy is even close to 100 percent. The elderly subjects will have reduced vascular elasticity due to their older age, and the peak structure of typical pulse image features will become less pronounced, thus reducing the diagnostic accuracy compared to young patients, and the experimental model has very good experimental effect on pancreatitis and extremely high accuracy.
The above description is only of the preferred embodiments of the present invention, and is not intended to limit the present invention. Any simple modification, variation and equivalent variation of the above embodiments according to the technical substance of the invention still fall within the scope of the technical solution of the invention.
Claims (6)
1. A method of mining pathological features for pancreatitis patients, the method comprising:
acquiring a labeled sample pulse signal curve, wherein the label is used for distinguishing individuals corresponding to the sample pulse signal curve from non-pancreatitis patients or pancreatitis patients;
dividing the sample pulse signal curve according to the period to obtain each single-period curve;
extracting the basic time domain characteristics of each single-period curve;
extracting the stability index and the double peak index of each single-period curve as case characteristics of pancreatitis corresponding to the single-period curve, wherein the double peak index represents two peaks and an obvious concave arc structure in the single-period curve; the stability index is used for representing the change rate of the second half of the single-period curve;
the basic time domain features of pulse signal curves of all samples and the case features of pancreatitis form input feature vectors, the input feature vectors are input into a classification model, and the classification model is supervised by labels corresponding to all samples and trained to obtain a trained classification model;
the stability index of each monocycle curve is extracted specifically as follows: the variance of each sampling point is obtained by selecting a plurality of sampling points in a single-period curve and calculating the variance of each sampling point by a first formula, wherein the variances of all the sampling points are used for representing the stability index of the pulse single-period signal of each single-period curve, and the first formula is as follows:
；
in the first formula, N represents the number of sampling points in a single-period curve, N represents the number of sampling points in the whole period of a sample pulse signal curve, S F Representing the sampling frequency, t representing the starting position of the sampling point,D i representing the variance of the sampling point at i, where i is a positive integer from 0 to N, the parameterX i Representing the absolute coordinate value, x, of the sampling point at the abscissa i i Represents the abscissa of the sampling point at i, y i Representing the ordinate of the sampling point at i;
the extraction of the double peak index of each single period curve is specifically as follows: selecting a plurality of sampling points in the single-period curve, calculating to obtain the average value of the ordinate values of each sampling point of the non-pancreatitis patient, and calculating the sum of differences between the ordinate values of the sampling points between two wave peaks in the single-period curve and the average value of the ordinate values of the sampling points of the same non-pancreatitis patient through a second formula, wherein the sum is used as a dual-wave peak index; the second formula is as follows:
；
in the second formula, Y represents the data set of the ordinate of the sampling point, N represents the number of the sampling points in the whole period of the sample pulse signal curve, N represents the number of the sampling points in the single period curve, T represents the point with the highest ordinate value in the single period pulse signal curve, T represents the point with the next highest ordinate value in the single period curve, S F Representing the sampling frequency of the sample,Arepresents the average value of the ordinate values of each sampling point of a normal person,B i the sum of the difference between the ordinate value of the sampling point at i and A of pancreatitis patient is shown.
2. A method of mining pathology towards pancreatitis patients according to claim 1, wherein the labels are divided into patient labels and non-patient labels;
the trained classification model further comprises:
acquiring a pulse signal curve of a person to be diagnosed;
in pulse signal curves of a person to be diagnosed, inputting feature vectors composed of basic time domain features of all single-period curves and case features of pancreatitis into a trained classification model to obtain a diagnosis tag set; each label in the diagnosis label set corresponds to a single-period curve one by one;
if the occurrence frequency of the patient labels in the diagnosis label set is larger than a preset threshold, determining that the person to be diagnosed is a suspected pancreatitis patient.
3. A method of mining pathological features in patients with pancreatitis according to claim 1 or 2, characterized in that the sampling frequency is 2 times the value of the beating frequency of human pulse.
4. The method of claim 1 or 2, wherein the basic time domain features of each monocycle pulse signal curve include a first peak height, a second peak height, a third peak height, a first peak to second peak time interval, a second peak to third peak time interval, a pulse start to first trough time interval, a first trough to second trough time interval, and a second trough to pulse start time interval of the pulse.
5. The method for mining pathological features of pancreatitis-oriented patients according to claim 2, wherein the pulse signal profile of the person to be diagnosed is at least one single cycle pulse signal profile.
6. The method for mining pathological features of pancreatitis-oriented patients according to claim 1, wherein the classification model is trained by SVM kernel method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111484127.1A CN114145722B (en) | 2021-12-07 | 2021-12-07 | Pulse pathological feature mining method for pancreatitis patients |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111484127.1A CN114145722B (en) | 2021-12-07 | 2021-12-07 | Pulse pathological feature mining method for pancreatitis patients |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114145722A CN114145722A (en) | 2022-03-08 |
| CN114145722B true CN114145722B (en) | 2023-08-01 |
Family
ID=80452959
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111484127.1A Active CN114145722B (en) | 2021-12-07 | 2021-12-07 | Pulse pathological feature mining method for pancreatitis patients |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114145722B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| RU2482793C1 (en) * | 2012-01-11 | 2013-05-27 | Учреждение Российской академии наук Институт химической биологии и фундаментальной медицины Сибирского отделения РАН (ИХБФМ СО РАН) | Method of differential diagnostics of acute pancreatitis forms in experiment |
| CN103932686A (en) * | 2014-04-22 | 2014-07-23 | 北京印刷学院 | Method and device for extracting pulse condition signal |
| CN112842288A (en) * | 2021-01-29 | 2021-05-28 | 清华大学深圳国际研究生院 | Pulse data classification model establishing device, classification recognition device and measurement system |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140073966A1 (en) * | 2012-09-11 | 2014-03-13 | Nellcor Puritan Bennett Llc | Methods and systems for determining algorithm settings for use in determining physiological information |
| US20140081152A1 (en) * | 2012-09-14 | 2014-03-20 | Nellcor Puritan Bennett Llc | System and method for determining stability of cardiac output |
| CN111374639A (en) * | 2018-12-28 | 2020-07-07 | 陈柏龄 | Septicemia strain prediction system and method |
-
2021
- 2021-12-07 CN CN202111484127.1A patent/CN114145722B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| RU2482793C1 (en) * | 2012-01-11 | 2013-05-27 | Учреждение Российской академии наук Институт химической биологии и фундаментальной медицины Сибирского отделения РАН (ИХБФМ СО РАН) | Method of differential diagnostics of acute pancreatitis forms in experiment |
| CN103932686A (en) * | 2014-04-22 | 2014-07-23 | 北京印刷学院 | Method and device for extracting pulse condition signal |
| CN112842288A (en) * | 2021-01-29 | 2021-05-28 | 清华大学深圳国际研究生院 | Pulse data classification model establishing device, classification recognition device and measurement system |
Non-Patent Citations (1)
| Title |
|---|
| Wrist Pulse Signal Classification for Inflammation of Appendix, Pancreas, and Duodenum;Chow WH 等;ECON 2014 - 40TH ANNUAL CONFERENCE OF THE IEEE INDUSTRIAL ELECTRONICS SOCIETY;2479-2483 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114145722A (en) | 2022-03-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20240221948A1 (en) | Systems and methods of analyte measurement analysis | |
| Altan et al. | Deep learning with 3D-second order difference plot on respiratory sounds | |
| Zhang et al. | A sensor-based wrist pulse signal processing and lung cancer recognition | |
| Chen et al. | Computerized wrist pulse signal diagnosis using modified auto-regressive models | |
| CN111067503A (en) | Sleep staging method based on heart rate variability | |
| JP2012517849A (en) | System and method for automatically capturing and archiving clinically meaningful vital signs | |
| CN109645979A (en) | Ambulatory ecg signal artifact identification method and device | |
| CN112493995B (en) | Anesthesia state evaluation system and method suitable for patients of different ages | |
| Alivar et al. | Motion detection in bed-based ballistocardiogram to quantify sleep quality | |
| CN114711725B (en) | Sleep apnea detection method and device based on dual-attention mechanism | |
| Liu et al. | Motion artifact detection in ppg signals based on gramian angular field and 2-d-cnn | |
| CN115089145A (en) | Intelligent blood pressure prediction method based on multi-scale residual error network and PPG signal | |
| Ma et al. | PPG-based continuous BP waveform estimation using polarized attention-guided conditional adversarial learning model | |
| Kumar et al. | CACHET-CADB: A contextualized ambulatory electrocardiography arrhythmia dataset | |
| CN114145722B (en) | Pulse pathological feature mining method for pancreatitis patients | |
| CN115836847A (en) | Blood pressure prediction device and equipment | |
| Saadi et al. | Low-cost ecg monitoring system with classification using deep learning | |
| Kandukuri et al. | Constant Q-transform-based deep learning architecture for detection of obstructive sleep apnea | |
| Susana et al. | Review of Non-Invasive Blood Glucose Level Estimation based on Photoplethysmography and Artificial Intelligent Technology | |
| Zhang et al. | Computerized radial artery pulse signal classification for lung cancer detection | |
| Lin et al. | Research on feature mining algorithm and disease diagnosis of pulse signal based on piezoelectric sensor | |
| Chen et al. | A novel real-time depth of anaesthesia monitoring method using detrended fluctuation analysis and ann | |
| Vyas et al. | Sleep Stage Classification Using Non-Invasive Bed Sensing and Deep Learning | |
| Savostin et al. | DEVISING A METHOD FOR PREDICTING A BLOOD PRESSURE LEVEL BASED ON ELECTROCARDIOGRAM AND PHOTOPLETHYSMOGRAM SIGNALS. | |
| WO2023221068A1 (en) | Method and system for intelligent pulse wave analysis based on human body ordinal parameter model |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |