CN113435607A - Disease screening method based on federal learning - Google Patents

Abstract

The invention discloses a disease screening method based on federal learning, which comprises the following steps: 1) establishing a shared data set; 2) pre-training; 3) calculating an accuracy average value; 4) calculating the data quantity; 5) and (6) converging. The invention belongs to the technical field of federal learning and deep learning, and particularly relates to a disease screening method based on federal learning, which combines knowledge of federal learning and deep learning, can more fully utilize data of various regions, and can efficiently and accurately diagnose diseases.

Description

Disease screening method based on federal learning

Technical Field

The invention belongs to the technical field of federal learning and deep learning, and particularly relates to a disease screening method based on federal learning.

Background

Medical images are images that reflect the internal structure or internal function of an anatomical region and are composed of a set of image elements, pixels (2D) or voxels (3D). Medical images are discrete image representations produced by sampling or reconstruction that can map values to different spatial locations. The number of pixels is used to describe the medical imaging under a certain imaging device and is an expression for describing the anatomy and its functional details. The specific values expressed by the pixels are determined by the imaging equipment, imaging protocol, image reconstruction, and post-processing. But sometimes the data is examined for even as many as thousands of images per patient, and more images are assumed when the disease is in a concentrated outbreak. Physician diagnosis alone, resulting in slow efficiency. The image data is processed by combining the federal learning and the deep learning, so that the data can be comprehensively utilized, the image can be rapidly analyzed, and the accurate diagnosis can be realized.

The closest techniques to the present invention are:

1. the image classification algorithm based on deep learning comprises the following steps: image classification is to distinguish different image types according to semantic information of the images, and is an important basic problem in computer vision. In deep learning, a Convolutional Neural Network (CNN) is mainly used for image classification, pixel information of an image is used as input, feature extraction and high-level abstraction are performed through Convolution operation, and model output is directly the result of image recognition. The current common image classification CNN networks include Lenet, Alxnet, Vgg series, Resnet series, inclusion series, densnet series, Googlenet, and the like. However, the method cannot comprehensively utilize data of hospitals in various regions for comprehensive processing, so that the method has low accuracy and poor performance.

2. The DeCoVNet network is a weakly supervised learning algorithm. The method has a heavier practical significance in the case of uneven distribution of disease cases and small overall number. The method has the characteristics of high operation speed and less requirement on data label solving, but still has the defect of low accuracy so as to easily generate misdiagnosis.

Disclosure of Invention

In order to overcome the defects of the prior art, the invention provides a disease screening method based on federal learning, which combines the knowledge of federal learning and deep learning, can make full use of data of various regions, and can efficiently and accurately diagnose diseases.

The technical scheme adopted by the invention is as follows: the invention relates to a disease screening method based on federal learning, which comprises the following steps:

1) the server builds a shared data set and sends an initial model (U-Net + +, DecoVNet) to the client participating in training;

2) when the client runs the first epoch, a mask with a mark and a training set with a mark are obtained by using unsupervised learning and training to pre-train the U-Net + + model; obtaining masks of all training sets by using the U-Net + + model after all training sets are pre-trained; the mask, the training set and the label of all the training sets enter a judgment model DecoVNet together, and the training sets are trained; the mask and the test set of all the test sets enter a judgment model DecoVNet for testing; calculating local precision and uploading the local precision to a server;

3) the server collects the precision of the client sides participating in training and calculates the precision average value;

4) and before the second epoch is operated, if the precision of a certain client is smaller than the precision average value, sending m pieces of data to the client. The formula is as follows: m ═ i × n, where i ═ c (acc)_avg-acc_i)/(acc_avg-acc_min) N is the size of the shared data set, and the data set issued by the shared data set is not larger than the data volume of the client;

5) after the second epoch, the client having the issued shared data set trains the shared data set and the own data set together to the model until the whole model converges. .

The invention with the structure has the following beneficial effects: according to the disease screening method based on the federal learning, a model is established for a medical image through a deep learning convolutional neural network, so that the disease diagnosis accuracy is effectively improved; the node data of each hospital is comprehensively utilized by using federal learning, so that the generalization capability of the model is improved; a dynamic fusion strategy is provided, and the overall accuracy of the system is improved.

Drawings

FIG. 1 is a block diagram of the Federal learning fusion method of the invention based on the Federal learning disease screening method;

FIG. 2 is a block diagram of a medical image deep learning method based on the federal learning disease screening method of the present invention.

The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the principles of the invention and not to limit the invention.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments; all other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

As shown in fig. 1-2, the federal learning based disease screening method of the present invention comprises the following steps:

1) the server builds a shared data set and sends an initial model (U-Net + +, DecoVNet) to the client participating in training;

2) when the client runs the first epoch, a mask with a mark and a training set with a mark are obtained by using unsupervised learning and training to pre-train the U-Net + + model; obtaining masks of all training sets by using the U-Net + + model after all training sets are pre-trained; the mask, the training set and the label of all the training sets enter a judgment model DecoVNet together, and the training sets are trained; the mask and the test set of all the test sets enter a judgment model DecoVNet for testing; calculating local precision and uploading the local precision to a server;

3) the server collects the precision of the client sides participating in training and calculates the precision average value;

4) and before the second epoch is operated, if the precision of a certain client is smaller than the precision average value, sending m pieces of data to the client. The formula is as follows: m ═ i × n, where i ═ c (acc)_avg-acc_i)/(acc_avg-acc_min) N is the size of the shared data set, and the data set issued by the shared data set is not larger than the data volume of the client;

5) after the second epoch, the client having the issued shared data set trains the shared data set and the own data set together to the model until the whole model converges.

When the method is used specifically, a user firstly sets a shared data set, then in the first epoch, after n iterations are carried out on a client, the precision is uploaded to a server, and the server collects all the precision of the client participating in training and calculates the average value. And if the precision of the ith client is lower than the precision average value, the data volume of the issued shared data is obtained (the data volume is not larger than the data volume of the client), and if the precision of the ith client is higher than the precision average value, the shared data value is not issued. And finally, the clients with the shared data sets in the remaining epochs are trained until the model converges. Aiming at the characteristic that a large number of medical images are diagnosed every day in a densely populated city, doctors face the condition that the judging speed of a large number of medical images is low, so that an artificial intelligence model is urgently needed to help the doctors judge diseases. The invention provides a model which is characterized in that a U-Net + + model is used for segmenting a medical image (judging a disease region), and then a judgment model (DecoVNet) is used for judging whether a disease exists in a segmented part, so that the overall working process is the working process of the invention, and the step is repeated when the medical image is used next time.

It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus 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 apparatus.

Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.

The present invention and its embodiments have been described above, and the description is not intended to be limiting, and the drawings are only one embodiment of the present invention, and the actual structure is not limited thereto. In summary, those skilled in the art should appreciate that they can readily use the disclosed conception and specific embodiments as a basis for designing or modifying other structures for carrying out the same purposes of the present invention without departing from the spirit and scope of the invention as defined by the appended claims.

Claims (1)

1. A federal learning-based disease screening method, characterized in that; the method comprises the following steps:

1) the server builds a shared data set and sends an initial model (U-Net + +, DecoVNet) to the client participating in training;

2) when the client runs the first epoch, a mask with a mark and a training set with a mark are obtained by using unsupervised learning and training to pre-train the U-Net + + model; obtaining masks of all training sets by using the U-Net + + model after all training sets are pre-trained; the mask, the training set and the label of all the training sets enter a judgment model DecoVNet together, and the training sets are trained; the mask and the test set of all the test sets enter a judgment model DecoVNet for testing; calculating local precision and uploading the local precision to a server;

3) the server collects the precision of the client sides participating in training and calculates the precision average value;

4) and before the second epoch is operated, if the precision of a certain client is smaller than the precision average value, sending m pieces of data to the client. The formula is as follows: m ═ i × n, where i ═ c (acc)_avg-acc_i)/(acc_avg-acc_min) N is the size of the shared data set, and the data set issued by the shared data set is not larger than the data volume of the client;

5) after the second epoch, the client having the issued shared data set trains the shared data set and the own data set together to the model until the whole model converges.

Images