CN111931866B - Medical data processing method, device, equipment and storage medium - Google Patents
Medical data processing method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN111931866B CN111931866B CN202010991493.5A CN202010991493A CN111931866B CN 111931866 B CN111931866 B CN 111931866B CN 202010991493 A CN202010991493 A CN 202010991493A CN 111931866 B CN111931866 B CN 111931866B
- Authority
- CN
- China
- Prior art keywords
- data
- audio
- target
- image
- fuzzy
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L21/00—Processing of the speech or voice signal to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility
- G10L21/02—Speech enhancement, e.g. noise reduction or echo cancellation
- G10L21/0208—Noise filtering
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/03—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the type of extracted parameters
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H30/00—ICT specially adapted for the handling or processing of medical images
-
- 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
- G16H50/70—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients
Abstract
The invention relates to the technical field of internet medical treatment, and discloses a medical data processing method, a device, equipment and a storage medium, wherein the method comprises the steps of carrying out image characteristic optimization on image data in medical data to be processed to obtain target image data, carrying out audio denoising on audio data to obtain target audio data; and respectively extracting the characteristics of the target image data and the target audio data, and selecting a target fuzzy clustering algorithm according to the extracted data characteristic information to extract and classify the characteristics of the medical data to be processed to obtain a classification result. The image data and the audio data are respectively subjected to targeted processing, then the data characteristic information is obtained by carrying out characteristic extraction according to the obtained data, and then the target fuzzy clustering algorithm is selected according to the data characteristic information to carry out the characteristic extraction and classification of the medical data.
Description
Technical Field
The invention relates to the technical field of internet medical treatment, in particular to a medical data processing method, a medical data processing device, medical data processing equipment and a storage medium.
Background
In the medical field, the extraction and classification of data features are important for subsequent medical research work, so that the feature extraction and classification of medical data are very meaningful matters. For the extraction and classification of medical data features, the conventional method is to strictly divide each data to be extracted into a certain class by a clustering algorithm, but a large number of clustering problems with unclear boundaries exist in the medical field, the categories and the behaviors of a large number of data have mediality, the feature extraction and classification of the medical data by the clustering algorithm in a general way are not reasonable, and the accuracy is low.
The above is only for the purpose of assisting understanding of the technical aspects of the present invention, and does not represent an admission that the above is prior art.
Disclosure of Invention
The invention mainly aims to provide a medical data processing method, a medical data processing device, medical data processing equipment and a medical data processing storage medium, and aims to solve the technical problem that medical data cannot be reasonably and accurately classified in the prior art.
In order to achieve the above object, the present invention provides a medical data processing method, including the steps of:
acquiring image data and audio data contained in medical data to be processed;
performing image characteristic optimization on the image data to obtain target image data, and performing audio denoising on the audio data to obtain target audio data;
respectively extracting the characteristics of the target image data and the target audio data, and determining a target fuzzy clustering algorithm according to the extracted data characteristic information;
performing feature extraction on the medical data to be processed based on the target fuzzy clustering algorithm to obtain feature data to be classified;
and classifying the characteristic data to be classified to obtain a classification result.
Preferably, the step of performing image feature optimization on the image data to obtain target image data includes:
carrying out transform domain processing on the image data to obtain image data to be compressed;
compressing the image data to be compressed by adopting an image compression technology to obtain compressed image data;
and detecting whether a target characteristic object exists in the compressed image data, and if so, performing characteristic optimization on the target characteristic object to obtain target image data.
Preferably, the step of detecting whether a target feature object exists in the compressed image data, and if so, performing feature optimization on the target feature object to obtain target image data includes:
detecting whether a target characteristic object exists in the compressed image data, if so, performing image enhancement and restoration on the compressed image data to obtain an image to be cut;
cutting the image to be cut to obtain a partial image containing the target feature object;
acquiring edge feature information of the partial image and region feature information of the target feature object in the partial image;
and taking the edge characteristic information and the area characteristic information as target image data.
Preferably, the step of performing audio denoising on the audio data to obtain target audio data includes:
pre-emphasis is carried out on the audio data to obtain audio data to be processed;
framing the audio data to be processed to obtain framed audio data, and windowing the framed audio data to obtain audio data to be denoised;
and carrying out end point detection on the audio data to be denoised, and acquiring target audio data according to a detection result.
Preferably, when the medical data to be processed includes image data and audio data, the target fuzzy clustering algorithm includes: a fuzzy C-means clustering algorithm and a fuzzy C-means clustering algorithm based on an ant colony algorithm;
the step of extracting the characteristics of the medical data to be processed based on the target fuzzy clustering algorithm to obtain the characteristic data to be classified comprises the following steps:
performing feature extraction on image data in the medical data to be processed according to the fuzzy C-means clustering algorithm to obtain image feature data;
performing feature extraction on the audio data in the medical data to be processed according to the ant colony algorithm-based fuzzy C-means clustering algorithm to obtain audio feature data;
and taking the image characteristic data and the audio characteristic data as characteristic data to be classified.
Preferably, the step of performing feature extraction on the image data in the medical data to be processed according to the fuzzy C-means clustering algorithm to obtain image feature data includes:
setting the clustering number corresponding to the image data and the clustering center of fuzzy classification according to the fuzzy C-means clustering algorithm;
acquiring membership functions corresponding to different data samples in the image data;
calculating fuzzy classification corresponding to the image data according to the clustering number, the clustering center of the fuzzy classification and the membership function through the following formula;
wherein j is the jth fuzzy classification, C is the number of clusters, and n is the data sample xiNumber of (d), muj（xi) Is a membership function of the ith data sample to the jth fuzzy classification, and
，mjthe cluster center of the jth fuzzy classification is defined, and p is a fuzzy degree control coefficient;
and performing feature extraction on the image data according to the fuzzy classification to obtain image feature data.
Preferably, the step of performing feature extraction on the audio data in the medical data to be processed according to the ant colony algorithm-based fuzzy C-means clustering algorithm to obtain audio feature data includes:
acquiring a plurality of audio characteristic vectors corresponding to audio data in the medical data to be processed;
taking a plurality of audio feature vectors as ant colonies, and determining clustering centers and clustering numbers corresponding to the audio feature vectors according to an ant colony algorithm;
calculating Euclidean distances between the audio feature vectors and the clustering center, and determining a membership function according to the Euclidean distances;
determining fuzzy classification corresponding to the audio data according to the cluster number, the cluster center and the membership function;
and performing feature extraction on the audio data according to the fuzzy classification to obtain audio feature data.
In addition, to achieve the above object, the present invention also proposes a medical data processing apparatus including:
the data acquisition module is used for acquiring image data and audio data contained in the medical data to be processed;
the data processing module is used for optimizing image characteristics of the image data to obtain target image data and carrying out audio denoising on the audio data to obtain target audio data;
the algorithm determining module is used for respectively extracting the characteristics of the target image data and the target audio data and determining a target fuzzy clustering algorithm according to the extracted data characteristic information;
the characteristic extraction module is used for extracting the characteristics of the medical data to be processed based on the target fuzzy clustering algorithm to obtain the characteristic data to be classified;
and the data classification module is used for classifying the characteristic data to be classified to obtain a classification result.
Further, to achieve the above object, the present invention also proposes a medical data processing apparatus comprising: a memory, a processor and a medical data processing program stored on the memory and executable on the processor, the medical data processing program being configured to implement the steps of the medical data processing method as described above.
Furthermore, to achieve the above object, the present invention also proposes a storage medium having stored thereon a medical data processing program which, when executed by a processor, implements the steps of the medical data processing method as described above.
The method comprises the steps of acquiring image data and audio data contained in medical data to be processed; performing image characteristic optimization on the image data to obtain target image data, and performing audio denoising on the audio data to obtain target audio data; respectively extracting the characteristics of the target image data and the target audio data, and determining a target fuzzy clustering algorithm according to the extracted data characteristic information; performing feature extraction on medical data to be processed based on a target fuzzy clustering algorithm to obtain feature data to be classified; and classifying the characteristic data to be classified to obtain a classification result. The image data and the audio data are respectively processed in a targeted manner, then the feature extraction is carried out according to the processed data to obtain the data feature information, and then the target fuzzy clustering algorithm is selected according to the data feature information to carry out the feature extraction and classification of the medical data.
Drawings
Fig. 1 is a schematic structural diagram of a medical data processing device in a hardware operating environment according to an embodiment of the present invention;
FIG. 2 is a schematic flow chart diagram illustrating a first embodiment of a medical data processing method according to the present invention;
FIG. 3 is a flow chart illustrating a second embodiment of a medical data processing method according to the present invention;
FIG. 4 is a schematic flow chart diagram illustrating a third embodiment of a medical data processing method according to the present invention;
fig. 5 is a block diagram showing the configuration of the first embodiment of the medical data processing apparatus according to the present invention.
The implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.
Detailed Description
It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Referring to fig. 1, fig. 1 is a schematic structural diagram of a medical data processing device in a hardware operating environment according to an embodiment of the present invention.
As shown in fig. 1, the medical data processing apparatus may include: a processor 1001, such as a Central Processing Unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. Wherein a communication bus 1002 is used to enable connective communication between these components. The user interface 1003 may include a Display screen (Display), an input unit such as a Keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface, a wireless interface. The network interface 1004 may optionally include a standard wired interface, a WIreless interface (e.g., a WIreless-FIdelity (WI-FI) interface). The Memory 1005 may be a Random Access Memory (RAM) Memory, or may be a Non-Volatile Memory (NVM), such as a disk Memory. The memory 1005 may alternatively be a storage device separate from the processor 1001.
Those skilled in the art will appreciate that the configuration shown in fig. 1 does not constitute a limitation of the medical data processing apparatus and may include more or fewer components than those shown, or some components may be combined, or a different arrangement of components.
As shown in fig. 1, the memory 1005, which is one type of storage medium, may include therein an operating system, a data storage module, a network communication module, a user interface module, and a medical data processing program.
In the medical data processing apparatus shown in fig. 1, the network interface 1004 is mainly used for data communication with a network server; the user interface 1003 is mainly used for data interaction with a user; the processor 1001 and the memory 1005 in the medical data processing apparatus according to the present invention may be provided in the medical data processing apparatus which calls the medical data processing program stored in the memory 1005 by the processor 1001 and executes the medical data processing method provided by the embodiment of the present invention.
An embodiment of the present invention provides a medical data processing method, and referring to fig. 2, fig. 2 is a schematic flow chart of a first embodiment of the medical data processing method according to the present invention.
In this embodiment, the medical data processing method includes the steps of:
step S10: acquiring image data and audio data contained in medical data to be processed;
it should be noted that the execution main body of the present solution may be a computing service device (hereinafter, referred to as a terminal device) with data processing, image recognition, network communication and program running functions, such as a smart phone, a tablet computer, a personal computer, and the like. The medical data to be processed may be disease diagnosis data collected by the medical equipment, and the data may include image/video data, audio data, text data, and the like.
In practical applications, the processing performed by different terminal devices of data types is also different. For example, for text data, medical staff can read intuitively, and terminal equipment can also recognize directly, so that the data cannot have an unclear boundary in clustering, and the terminal equipment can not process the data; however, the image/video data and the audio data are easy to have an unclear boundary during clustering, so that the image/video data and the audio data need to be respectively processed correspondingly to ensure the accuracy of subsequent medical data classification.
Step S20: performing image characteristic optimization on the image data to obtain target image data, and performing audio denoising on the audio data to obtain target audio data;
in this step, the terminal device may perform image feature optimization on the image data in the medical data to be processed, and may perform image transformation, image compression, image enhancement and restoration, and/or image segmentation on the image data. The audio denoising of the audio data in the medical data to be processed may be performed by performing pre-emphasis, framing, windowing, and/or endpoint detection on the audio data.
In a specific implementation, after acquiring the medical data to be processed, the terminal device may first determine whether image data and audio data exist in the medical data to be processed, and if so, respectively process the image data and the audio data in the medical data to be processed according to the preprocessing manner (i.e., image feature optimization and audio denoising), so as to obtain target image data and target audio data.
Step S30: respectively extracting the characteristics of the target image data and the target audio data, and determining a target fuzzy clustering algorithm according to the extracted data characteristic information;
it should be noted that, in order to ensure the classification accuracy of the medical data, the configuration of the clustering algorithm may be performed in advance for the medical data with different types and different data characteristics, so that the configured clustering algorithm can accurately perform clustering and feature extraction on the data. For example, a fuzzy C-means clustering algorithm can be configured for the image data to perform feature extraction; further, considering that voiceprint recognition of audio data is a large sample data classification problem and the dimension is high, if the feature extraction is still performed by adopting the fuzzy C-means clustering algorithm, reasonable parameters can be set in the algorithm very difficultly, the reasonable parameters can be selected only through multiple experiments, time is consumed for calculating the clustering center and the membership degree matrix every time, the global optimum cannot be guaranteed, and therefore the improved fuzzy C-means clustering method must be used for feature extraction of the audio data.
In specific implementation, a mapping relation between data characteristic information and different fuzzy clustering algorithms can be established in advance to improve the determination speed of the target fuzzy clustering algorithm.
It should be understood that the feature extraction in this step is different for different types of data, for example, the feature extraction may be performed by using mel-frequency cepstrum coefficient algorithm for audio data, and the feature extraction may be performed by using histogram method for image data.
Step S40: performing feature extraction on the medical data to be processed based on the target fuzzy clustering algorithm to obtain feature data to be classified;
it should be understood that after the target fuzzy clustering algorithm for performing feature extraction on different types of data is determined according to the feature information of the different types of data in the medical data, the terminal device may perform feature extraction on the medical data to be processed based on the target fuzzy clustering algorithm to obtain feature data to be classified.
Specifically, the terminal device may perform feature extraction on image data in the medical data to be processed according to the fuzzy C-means clustering algorithm, and may perform feature extraction on audio data in the medical data to be processed according to the improved fuzzy C-means clustering algorithm, thereby obtaining feature data to be classified. For example, the terminal device may extract the features of the region of interest of the lung CT image by using a fuzzy C-means clustering algorithm; for the audio data, the fuzzy C-means clustering algorithm based on the ant colony algorithm (i.e. the improved fuzzy C-means clustering algorithm) can be used for carrying out feature extraction on the audio data generated by the patient.
Step S50: and classifying the characteristic data to be classified to obtain a classification result.
It should be understood that the classification is to aggregate data with similar features into one class, and to obtain a classification result for subsequent use. For example, the classification result can be input into a disease diagnosis system for lesion diagnosis, and the disease diagnosis system can determine the proportion of various lesions, such as the probability of lung inflammation of 90%, the probability of lung tumor of 10%, etc., so as to provide doctors with disease diagnosis and improve the efficiency and accuracy of diagnosis.
Further, in order to ensure the identification accuracy of the target fuzzy clustering algorithm, the terminal device further aggregates the data of the same category in the classification result, labels the aggregated data, and inputs the data to the corresponding fuzzy clustering algorithm for model training.
In order to further ensure the privacy and security of the medical data to be processed, the medical data to be processed may also be stored in a node of a block chain.
The embodiment acquires image data and audio data contained in medical data to be processed; performing image characteristic optimization on the image data to obtain target image data, and performing audio denoising on the audio data to obtain target audio data; respectively extracting the characteristics of the target image data and the target audio data, and determining a target fuzzy clustering algorithm according to the extracted data characteristic information; performing feature extraction on medical data to be processed based on a target fuzzy clustering algorithm to obtain feature data to be classified; and classifying the characteristic data to be classified to obtain a classification result. In the embodiment, the image data and the audio data are respectively processed in a targeted manner, then the feature extraction is carried out according to the processed data to obtain the data feature information, and then the target fuzzy clustering algorithm is selected according to the data feature information to carry out the feature extraction and classification of the medical data.
Referring to fig. 3, fig. 3 is a flowchart illustrating a medical data processing method according to a second embodiment of the present invention.
Based on the first embodiment, in this embodiment, the performing image feature optimization on the image data in step S20 to obtain target image data includes:
step S201: carrying out transform domain processing on the image data to obtain image data to be compressed;
step S202: compressing the image data to be compressed by adopting an image compression technology to obtain compressed image data;
step S203: and detecting whether a target characteristic object exists in the compressed image data, and if so, performing characteristic optimization on the target characteristic object to obtain target image data.
It should be noted that, in consideration of the fact that the image array corresponding to the image data is large, if the image array is directly processed in the spatial domain, the amount of calculation involved is large, and therefore in this embodiment, the terminal device may convert the image processing in the spatial domain into the image processing in the transform domain by using an indirect processing technique such as fourier transform or discrete cosine transform, which reduces the amount of calculation on the one hand and enables the image data to be processed more efficiently on the other hand.
The target feature object may be a diagnostic object or a collection object corresponding to medical data collected by the medical device, for example, a diagnostic object corresponding to a lung CT image, that is, a lung of a human body, and a diagnostic object corresponding to a gastroscope image, that is, a stomach of the human body. The diagnosis object may be determined based on the category of the clinic included in the medical data to be processed or based on the medical equipment that directly provides the medical data to be processed. Specifically, the terminal device may perform transform domain processing on the image data to obtain image data to be compressed; then, compressing the image data to be compressed by adopting an image compression technology to obtain compressed image data; and detecting whether a target characteristic object exists in the compressed image data, and if so, performing characteristic optimization on the target characteristic object to obtain target image data.
Further, in order to ensure that the obtained target image data is more practical and convenient for feature extraction, in this embodiment, the terminal device further detects whether a target feature object exists in the compressed image data, and if so, performs image enhancement and restoration on the compressed image data to obtain an image to be cut; cutting the image to be cut to obtain a partial image containing the target feature object; acquiring edge feature information of the partial image and region feature information of the target feature object in the partial image; using the edge feature information and the region feature information as target image data
The image compression is to reduce the data quantity of the description image by the image compression technology, thereby saving the image transmission and processing time and reducing the occupied memory capacity; enhancing and restoring the image, namely removing noise and improving the definition of the image, and highlighting an interested part in the image by utilizing an image enhancement technology so that the outline of an object in the image is clear and the details are obvious; and (4) image cutting, namely extracting meaningful characteristic parts in the image, and extracting edge information of the image and area information of a meaningful object. The edge feature information and the region feature information of the image data can be obtained by image segmentation.
It should be understood that common methods for image enhancement include histogram equalization, image smoothing, image sharpening, and pseudo-color. The image restoration principle is implemented by image filtering, such as wiener filtering, optimal filtering in various optimization meanings, and various types of median filtering are methods for image restoration.
Accordingly, the audio denoising of the audio data in the step S20 to obtain the target audio data includes:
step S204: pre-emphasis is carried out on the audio data to obtain audio data to be processed;
step S205: framing the audio data to be processed to obtain framed audio data, and windowing the framed audio data to obtain audio data to be denoised;
step S206: and carrying out end point detection on the audio data to be denoised, and acquiring target audio data according to a detection result.
It should be noted that, in order to ensure that the acquired target audio data is valid audio data without a mute portion and a noise portion. In this embodiment, the terminal device may pre-emphasize the audio data when the audio data exists in the medical data to be processed, so as to obtain the audio data to be processed; then framing the audio data to be processed to obtain framed audio data, and windowing the framed audio data to obtain audio data to be denoised; and carrying out end point detection on the audio data to be denoised, and acquiring target audio data according to a detection result.
Wherein the audio pre-emphasis aims to emphasize the high-frequency part of the audio, remove the influence of lip radiation and increase the high-frequency resolution of the audioAnd (4) rate. Generally by the transfer function H (z) =1-az-1In which Z is audio frequency, a is pre-emphasis coefficient, 0.9<a<1.0. And windowing and framing, because the audio voice signal is a non-stationary signal, in order to process the audio, the audio voice signal between 10 ms and 30ms can be assumed to be stationary, and the voice spectrum characteristic and the voice characteristic parameter are constant. At this time, the audio speech signal needs to be divided into short time segments, each of which is a frame, and in order to obtain a sample signal from the audio speech signal, the original speech signal needs to be multiplied by a time window function, which is windowing. Endpoint Detection, also called Voice Activity Detection (VAD), aims to distinguish between speech and non-speech areas. It is commonly understood that endpoint detection is to accurately locate the beginning and ending points of a speech from noisy speech, remove the silent part and the noise part, and find a piece of content where the speech is really valid.
According to the method and the device, the image data and the audio data in the medical data are processed respectively, so that the subsequent target fuzzy clustering algorithm can be selected more accurately, and the classification accuracy of the medical data is improved.
Referring to fig. 4, fig. 4 is a flowchart illustrating a third embodiment of the medical data processing method according to the present invention.
Based on the foregoing embodiments, in this embodiment, when the medical data to be processed includes image data and audio data, the target fuzzy clustering algorithm includes: a fuzzy C-means clustering algorithm and a fuzzy C-means clustering algorithm based on an ant colony algorithm;
correspondingly, the step S40 may specifically include:
step S401: performing feature extraction on image data in the medical data to be processed according to the fuzzy C-means clustering algorithm to obtain image feature data;
it should be understood that, in a Fuzzy C-Means clustering Algorithm (Fuzzy C-Means Algorithm, FCMA or FCM), the membership of each sample point to all the clustering centers is obtained by optimizing an objective function, so as to determine the category of the sample points, thereby achieving the purpose of automatically classifying the sample data.
In this embodiment, the terminal device may perform optimization of the objective function by using an iterative method through a fuzzy C-means clustering algorithm to obtain fuzzy classification of the image data, and then perform feature extraction on the image data by the fuzzy classification to obtain image feature data.
Specifically, when the terminal device classifies the image data by using the fuzzy C-means clustering algorithm, the clustering number and the fuzzy degree control coefficient (the numerical value is more than 1 and is used for controlling the fuzzy degree of the clustering result) need to be set firstly, then each clustering center is initialized, and the membership function is calculated; secondly, recalculating the clustering center by using the calculated membership function; the repeating technique of the clustering centers is repeatedly performed until a stable clustering center is obtained.
In practical application, the terminal equipment can set the clustering number corresponding to the image data and the clustering center of fuzzy classification according to the fuzzy C-means clustering algorithm; then acquiring membership functions corresponding to different data samples in the image data; calculating fuzzy classification corresponding to the image data according to the clustering number, the clustering center of the fuzzy classification and the membership function through the following formula; and finally, carrying out feature extraction on the image data according to the fuzzy classification to obtain image feature data.
Wherein j is the jth fuzzy classification, C is the number of clusters, and n is the data sample xiNumber of (d), muj（xi) Is a membership function of the ith data sample to the jth fuzzy classification, and
，mjthe cluster center of the jth fuzzy classification is defined, and p is a fuzzy degree control coefficient;
it should be noted that, in the minimum value solving process of the objective function j, different types of clustering centers and membership degrees can be obtained, so that the fuzzy clustering and partitioning process of the image data is completed. In addition, in this embodiment, the terminal device performs the stability test on the cluster center obtained by each calculation, and if the stability test result is unstable, the calculation of the membership function is repeatedly performed, and the operation of recalculating the cluster center by using the calculated membership function is performed until the finally obtained cluster center is sufficiently stable.
Step S402: performing feature extraction on the audio data in the medical data to be processed according to the ant colony algorithm-based fuzzy C-means clustering algorithm to obtain audio feature data;
it should be noted that the voiceprint recognition of audio data is a large sample data classification problem, the dimension is high, if the feature extraction is still performed by using the fuzzy C-means clustering algorithm, it is very difficult to set reasonable parameters in the algorithm, the reasonable parameters can only be selected through multiple experiments, the calculation of the clustering center and the membership matrix each time is time-consuming, and the global optimum cannot be guaranteed, so that the improved fuzzy C-means clustering method must be used for the feature extraction of the audio data.
Considering that the ant colony algorithm is a bionic evolutionary algorithm, namely, an optimal path of a food source can be always found in the process of simulating ants to find food, and when the path is blocked, the ants can quickly bypass barriers and find the optimal path again. Each ant releases an information hormone in the random walking process, the hormone is volatilized continuously along with the time, and when more ants select the path, the hormone on the path is enhanced, so that more ants select the path. The method is a heuristic algorithm, can merge surrounding data together according to the information quantity of a clustering center, meanwhile, the information quantity on each path changes in the moving process of ants, and the clustering center is continuously adjusted according to a membership function, so that the classification of audio data is finally obtained through iteration.
In this embodiment, the fuzzy C-means clustering algorithm is improved according to the ant colony algorithm, so as to obtain the fuzzy C-means clustering algorithm based on the ant colony algorithm. When the terminal equipment processes the audio data by using the fuzzy C-mean clustering algorithm based on the ant colony algorithm, the voice feature vectors are used as different ants, the clustering center is used as a food source, and the process of determining the clustering center is the process of searching the food source by the ants with different features. Specifically, a plurality of audio feature vectors corresponding to audio data in the medical data to be processed may be obtained first; then, taking a plurality of audio feature vectors as ant colonies, and determining clustering centers and clustering numbers corresponding to the audio feature vectors according to an ant colony algorithm; then, calculating Euclidean distances between the audio feature vectors and the clustering center, and determining a membership function according to the Euclidean distances; determining fuzzy classification corresponding to the audio data according to the cluster number, the cluster center and the membership function; and performing feature extraction on the audio data according to the fuzzy classification to obtain audio feature data. The method for calculating the fuzzy classification in this step may refer to the step S302, which is not described herein again.
Step S403: and taking the image characteristic data and the audio characteristic data as characteristic data to be classified.
In a specific implementation, after the image feature data and the audio feature data are obtained by the terminal device in the above manner, the image feature data and the audio feature data can be used as feature data to be classified for subsequent classification.
In the embodiment, the relevance and the behavior of a lot of medical data are considered to be mediated, and the fuzzy set theory provides a powerful analysis tool for the classification of the data, so that the fuzzy set theory and the clustering algorithm are combined into the fuzzy clustering algorithm to extract and classify the characteristics of the medical data, the classification is more reasonable, and the accuracy is higher. Meanwhile, in the embodiment, different fuzzy clustering algorithms are adopted for the audio data and the image data to perform feature extraction, so that the accuracy of data classification according to the features of the data is further improved.
Furthermore, an embodiment of the present invention further provides a storage medium, in which a medical data processing program is stored, and the medical data processing program, when executed by a processor, implements the steps of the medical data processing method as described above.
Referring to fig. 5, fig. 5 is a block diagram showing the structure of a first embodiment of the medical data processing apparatus according to the present invention.
As shown in fig. 5, a medical data processing apparatus according to an embodiment of the present invention includes:
a data obtaining module 501, configured to obtain image data and audio data included in medical data to be processed;
a data processing module 502, configured to perform image feature optimization on the image data to obtain target image data, and perform audio denoising on the audio data to obtain target audio data;
an algorithm determining module 503, configured to perform feature extraction on the target image data and the target audio data, respectively, and determine a target fuzzy clustering algorithm according to extracted data feature information;
a feature extraction module 504, configured to perform feature extraction on the medical data to be processed based on the target fuzzy clustering algorithm, so as to obtain feature data to be classified;
and a data classification module 505, configured to classify the feature data to be classified, so as to obtain a classification result.
The embodiment acquires image data and audio data contained in medical data to be processed; performing image characteristic optimization on the image data to obtain target image data, and performing audio denoising on the audio data to obtain target audio data; respectively extracting the characteristics of the target image data and the target audio data, and determining a target fuzzy clustering algorithm according to the extracted data characteristic information; performing feature extraction on medical data to be processed based on a target fuzzy clustering algorithm to obtain feature data to be classified; and classifying the characteristic data to be classified to obtain a classification result. In the embodiment, the image data and the audio data are respectively processed in a targeted manner, then the feature extraction is carried out according to the processed data to obtain the data feature information, and then the target fuzzy clustering algorithm is selected according to the data feature information to carry out the feature extraction and classification of the medical data.
A second embodiment of the medical data processing apparatus of the present invention is proposed based on the above-described first embodiment of the medical data processing apparatus of the present invention.
In this embodiment, the data processing module 502 is further configured to perform transform domain processing on the image data to obtain image data to be compressed; compressing the image data to be compressed by adopting an image compression technology to obtain compressed image data; and detecting whether a target characteristic object exists in the compressed image data, and if so, performing characteristic optimization on the target characteristic object to obtain target image data.
The data processing module 502 is further configured to detect whether a target feature object exists in the compressed image data, and if so, perform image enhancement and restoration on the compressed image data to obtain an image to be cut; cutting the image to be cut to obtain a partial image containing the target feature object; acquiring edge feature information of the partial image and region feature information of the target feature object in the partial image; and taking the edge characteristic information and the area characteristic information as target image data.
The data processing module 502 is further configured to pre-emphasize the audio data to obtain audio data to be processed; framing the audio data to be processed to obtain framed audio data, and windowing the framed audio data to obtain audio data to be denoised; and carrying out end point detection on the audio data to be denoised, and acquiring target audio according to a detection result.
The feature extraction module 504 is further configured to perform feature extraction on image data in the medical data to be processed according to the fuzzy C-means clustering algorithm to obtain image feature data; performing feature extraction on the audio data in the medical data to be processed according to the ant colony algorithm-based fuzzy C-means clustering algorithm to obtain audio feature data; and taking the image characteristic data and the audio characteristic data as characteristic data to be classified.
The feature extraction module 504 is further configured to set a cluster number corresponding to the image data and a cluster center of the fuzzy classification according to the fuzzy C-means clustering algorithm; acquiring membership functions corresponding to different data samples in the image data; calculating fuzzy classification corresponding to the image data according to the clustering number, the clustering center of the fuzzy classification and the membership function through the following formula;
wherein j is the jth fuzzy classification, C is the number of clusters, and n is the data sample xiNumber of (d), muj（xi) Is a membership function of the ith data sample to the jth fuzzy classification, and
，mjthe cluster center of the jth fuzzy classification is defined, and p is a fuzzy degree control coefficient;
and performing feature extraction on the image data according to the fuzzy classification to obtain image feature data.
The feature extraction module 504 is further configured to obtain a plurality of audio feature vectors corresponding to audio data in the medical data to be processed; taking a plurality of audio feature vectors as ant colonies, and determining clustering centers and clustering numbers corresponding to the audio feature vectors according to an ant colony algorithm; calculating Euclidean distances between the audio feature vectors and the clustering center, and determining a membership function according to the Euclidean distances; determining fuzzy classification corresponding to the audio data according to the cluster number, the cluster center and the membership function; and performing feature extraction on the audio data according to the fuzzy classification to obtain audio feature data.
Other embodiments or specific implementations of the medical data processing apparatus according to the present invention may refer to the above method embodiments, and are not described herein again.
The block chain is a novel application mode of computer technologies such as distributed data storage, point-to-point transmission, a consensus mechanism, an encryption algorithm and the like. A block chain (Blockchain), which is essentially a decentralized database, is a series of data blocks associated by using a cryptographic method, and each data block contains information of a batch of network transactions, so as to verify the validity (anti-counterfeiting) of the information and generate a next block. The blockchain may include a blockchain underlying platform, a platform product service layer, an application service layer, and the like.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or system. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or system that comprises the element.
The above-mentioned serial numbers of the embodiments of the present invention are merely for description and do not represent the merits of the embodiments.
Through the above description of the embodiments, those skilled in the art will clearly understand that the method of the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but in many cases, the former is a better implementation manner. Based on such understanding, the technical solutions of the present invention may be embodied in the form of a software product, which is stored in a storage medium (e.g., a rom/ram, a magnetic disk, an optical disk) and includes instructions for enabling a terminal device (e.g., a mobile phone, a computer, a server, an air conditioner, or a network device) to execute the method according to the embodiments of the present invention.
The above description is only a preferred embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes, which are made by using the contents of the present specification and the accompanying drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.
Claims (9)
1. A medical data processing method, characterized in that the medical data processing method comprises:
acquiring image data and audio data contained in medical data to be processed;
performing image characteristic optimization on the image data to obtain target image data, and performing audio denoising on the audio data to obtain target audio data;
respectively extracting the characteristics of the target image data and the target audio data, and determining a target fuzzy clustering algorithm according to the extracted data characteristic information;
performing feature extraction on the medical data to be processed based on the target fuzzy clustering algorithm to obtain feature data to be classified;
classifying the characteristic data to be classified to obtain a classification result;
the step of respectively extracting the characteristics of the target image data and the target audio data and determining a target fuzzy clustering algorithm according to the extracted data characteristic information comprises the following steps:
performing feature extraction on the target audio data by adopting a Mel frequency cepstrum coefficient algorithm, performing feature extraction on the target image data by adopting a histogram method, and determining a target fuzzy clustering algorithm according to extracted data feature information;
the step of extracting the features of the medical data to be processed based on the target fuzzy clustering algorithm to obtain the feature data to be classified comprises the following steps:
performing feature extraction on image data in the medical data to be processed according to a fuzzy C-means clustering algorithm to obtain image feature data;
performing feature extraction on the audio data in the medical data to be processed according to a fuzzy C-means clustering algorithm based on an ant colony algorithm to obtain audio feature data;
and taking the image characteristic data and the audio characteristic data as characteristic data to be classified.
2. The medical data processing method according to claim 1, wherein the step of performing image feature optimization on the image data to obtain target image data comprises:
carrying out transform domain processing on the image data to obtain image data to be compressed;
compressing the image data to be compressed by adopting an image compression technology to obtain compressed image data;
and detecting whether a target characteristic object exists in the compressed image data, and if so, performing characteristic optimization on the target characteristic object to obtain target image data.
3. The medical data processing method according to claim 2, wherein the step of detecting whether a target feature object exists in the compressed image data, and if so, performing feature optimization on the target feature object to obtain target image data comprises:
detecting whether a target characteristic object exists in the compressed image data, if so, performing image enhancement and restoration on the compressed image data to obtain an image to be cut;
cutting the image to be cut to obtain a partial image containing the target feature object;
acquiring edge feature information of the partial image and region feature information of the target feature object in the partial image;
and taking the edge characteristic information and the area characteristic information as target image data.
4. The medical data processing method of claim 1, wherein the step of performing audio denoising on the audio data to obtain target audio data comprises:
pre-emphasis is carried out on the audio data to obtain audio data to be processed;
framing the audio data to be processed to obtain framed audio data, and windowing the framed audio data to obtain audio data to be denoised;
and carrying out end point detection on the audio data to be denoised, and acquiring target audio data according to a detection result.
5. The medical data processing method according to claim 1, wherein the step of performing feature extraction on the image data in the medical data to be processed according to the fuzzy C-means clustering algorithm to obtain image feature data comprises:
setting the clustering number corresponding to the image data and the clustering center of fuzzy classification according to the fuzzy C-means clustering algorithm;
acquiring membership functions corresponding to different data samples in the image data;
calculating fuzzy classification corresponding to the image data according to the clustering number, the clustering center of the fuzzy classification and the membership function through the following formula;
wherein j is the jth fuzzy classification, C is the number of clusters, and n is the data sample xiNumber of (d), muj（xi) Is a membership function of the ith data sample to the jth fuzzy classification, and
，mjthe cluster center of the jth fuzzy classification is defined, and p is a fuzzy degree control coefficient;
and performing feature extraction on the image data according to the fuzzy classification to obtain image feature data.
6. The medical data processing method according to claim 1, wherein the step of performing feature extraction on the audio data in the medical data to be processed according to the ant colony algorithm-based fuzzy C-means clustering algorithm to obtain audio feature data comprises:
acquiring a plurality of audio characteristic vectors corresponding to audio data in the medical data to be processed;
taking a plurality of audio feature vectors as ant colonies, and determining clustering centers and clustering numbers corresponding to the audio feature vectors according to an ant colony algorithm;
calculating Euclidean distances between the audio feature vectors and the clustering center, and determining a membership function according to the Euclidean distances;
determining fuzzy classification corresponding to the audio data according to the cluster number, the cluster center and the membership function;
and performing feature extraction on the audio data according to the fuzzy classification to obtain audio feature data.
7. A medical data processing apparatus, characterized in that the medical data processing apparatus comprises:
the data acquisition module is used for acquiring image data and audio data contained in the medical data to be processed;
the data processing module is used for optimizing image characteristics of the image data to obtain target image data and carrying out audio denoising on the audio data to obtain target audio data;
the algorithm determining module is used for respectively extracting the characteristics of the target image data and the target audio data and determining a target fuzzy clustering algorithm according to the extracted data characteristic information;
the characteristic extraction module is used for extracting the characteristics of the medical data to be processed based on the target fuzzy clustering algorithm to obtain the characteristic data to be classified;
the data classification module is used for classifying the characteristic data to be classified to obtain a classification result;
the algorithm determining module is further configured to perform feature extraction on the target audio data by using a mel-frequency cepstrum coefficient algorithm, perform feature extraction on the target image data by using a histogram method, and determine a target fuzzy clustering algorithm according to extracted data feature information;
the characteristic extraction module is also used for extracting the characteristics of the image data in the medical data to be processed according to a fuzzy C-mean clustering algorithm to obtain image characteristic data; performing feature extraction on the audio data in the medical data to be processed according to a fuzzy C-means clustering algorithm based on an ant colony algorithm to obtain audio feature data; and taking the image characteristic data and the audio characteristic data as characteristic data to be classified.
8. A medical data processing apparatus, characterized in that the apparatus comprises: memory, a processor and a medical data processing program stored on the memory and executable on the processor, the medical data processing program being configured to implement the steps of the medical data processing method according to any one of claims 1 to 6.
9. A storage medium, characterized in that the storage medium has stored thereon a medical data processing program which, when executed by a processor, implements the steps of the medical data processing method according to any one of claims 1 to 6.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010991493.5A CN111931866B (en) | 2020-09-21 | 2020-09-21 | Medical data processing method, device, equipment and storage medium |
| PCT/CN2020/125445 WO2021151337A1 (en) | 2020-09-21 | 2020-10-30 | Medical data processing method, apparatus, device, and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010991493.5A CN111931866B (en) | 2020-09-21 | 2020-09-21 | Medical data processing method, device, equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111931866A CN111931866A (en) | 2020-11-13 |
| CN111931866B true CN111931866B (en) | 2021-01-01 |
Family
ID=73333984
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010991493.5A Active CN111931866B (en) | 2020-09-21 | 2020-09-21 | Medical data processing method, device, equipment and storage medium |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN111931866B (en) |
| WO (1) | WO2021151337A1 (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115098596B (en) * | 2022-05-25 | 2023-04-25 | 开普数智科技(广东)有限公司 | Government affair related data carding method, government affair related data carding device, government affair related data carding equipment and readable storage medium |
| CN116030988A (en) * | 2022-12-28 | 2023-04-28 | 北京融威众邦电子技术有限公司 | Intelligent service equipment for hospitals and application method thereof |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103955707A (en) * | 2014-05-04 | 2014-07-30 | 电子科技大学 | Mass image sorting system based on deep character learning |
| CN110532424A (en) * | 2019-09-26 | 2019-12-03 | 西南科技大学 | A kind of lungs sound tagsort system and method based on deep learning and cloud platform |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7386165B2 (en) * | 2004-02-06 | 2008-06-10 | Siemens Medical Solutions Usa, Inc. | System and method for a sparse kernel expansion for a Bayes classifier |
| US20060269135A1 (en) * | 2005-03-15 | 2006-11-30 | Ramsay Thomas E | System and method for identifying objects of interest in image data |
| CN100422999C (en) * | 2006-09-14 | 2008-10-01 | 浙江大学 | Transmedia searching method based on content correlation |
| CN101470897B (en) * | 2007-12-26 | 2011-04-20 | 中国科学院自动化研究所 | Sensitive film detection method based on audio/video amalgamation policy |
| CN107016131A (en) * | 2017-05-19 | 2017-08-04 | 北方工业大学 | Machine learning algorithm based on enhanced clustering and application of algorithm |
| CN110533658A (en) * | 2019-09-02 | 2019-12-03 | 山东大学齐鲁医院 | Intelligent pulmonary emphysema diagnostic message processing system and method, information data processing terminal |
-
2020
- 2020-09-21 CN CN202010991493.5A patent/CN111931866B/en active Active
- 2020-10-30 WO PCT/CN2020/125445 patent/WO2021151337A1/en active Application Filing
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103955707A (en) * | 2014-05-04 | 2014-07-30 | 电子科技大学 | Mass image sorting system based on deep character learning |
| CN110532424A (en) * | 2019-09-26 | 2019-12-03 | 西南科技大学 | A kind of lungs sound tagsort system and method based on deep learning and cloud platform |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111931866A (en) | 2020-11-13 |
| WO2021151337A1 (en) | 2021-08-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Sun et al. | Cervical cancer diagnosis based on random forest | |
| Bemis et al. | Probabilistic segmentation of mass spectrometry (MS) images helps select important ions and characterize confidence in the resulting segments | |
| CN111931866B (en) | Medical data processing method, device, equipment and storage medium | |
| Xu et al. | Texture-specific bag of visual words model and spatial cone matching-based method for the retrieval of focal liver lesions using multiphase contrast-enhanced CT images | |
| Kavitha et al. | ECM-CSD: an efficient classification model for cancer stage diagnosis in CT lung images using FCM and SVM techniques | |
| Ahammed et al. | A machine learning approach for skin disease detection and classification using image segmentation | |
| Fuchs et al. | Computational pathology analysis of tissue microarrays predicts survival of renal clear cell carcinoma patients | |
| US11538577B2 (en) | System and method for automated diagnosis of skin cancer types from dermoscopic images | |
| Gopalakrishnan et al. | Multilevel thresholding based follicle detection and classification of polycystic ovary syndrome from the ultrasound images using machine learning | |
| do Nascimento et al. | Lymphoma images analysis using morphological and non-morphological descriptors for classification | |
| US11666286B2 (en) | Classification of polyps using learned image analysis | |
| CN110827236A (en) | Neural network-based brain tissue layering method and device, and computer equipment | |
| CN115204238B (en) | PPG signal identity recognition method for wearable equipment and wearable equipment | |
| WO2021253709A1 (en) | Electrocardiogram heartbeat classification method and apparatus, and electronic device and medium | |
| CN112560993A (en) | Data screening method and device, electronic equipment and storage medium | |
| Kumarganesh et al. | An enhanced medical diagnosis sustainable system for brain tumor detection and segmentation using ANFIS classifier | |
| CN112651450A (en) | Medical image classification method based on multi-example deep learning | |
| Özbay et al. | Interpretable pap-smear image retrieval for cervical cancer detection with rotation invariance mask generation deep hashing | |
| Abdullah et al. | Round randomized learning vector quantization for brain tumor imaging | |
| Wang et al. | Community-acquired pneumonia recognition by wavelet entropy and cat swarm optimization | |
| CN110648318A (en) | Auxiliary analysis method and device for skin diseases, electronic equipment and storage medium | |
| CN117036894B (en) | Multi-mode data classification method and device based on deep learning and computer equipment | |
| Zhu et al. | Improved maximally stable extremal regions based method for the segmentation of ultrasonic liver images | |
| CN111047580B (en) | Acoustic image quantitative analysis and comprehensive quantitative analysis method for ultrasonic molecular image research | |
| Breen et al. | Efficient subtyping of ovarian cancer histopathology whole slide images using active sampling in multiple instance learning |
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 |