Cell image detection method based on transfer learning

Abstract

The invention provides a cell image detection method based on transfer learning, and the method comprises the steps: collecting a cell image through a Fourier laminated microscopic imaging system, carrying out fusion through frequency spectrum iteration, obtaining a large-view-field and high-resolution cell image, constructing a cell density estimation network through VGG and FPN network models, marking a cell center position in the cell image to obtain a cell density map, inputting the cell density map into a training model, constructing a cell detection network model by taking the trained network model as a backbone network, carrying out transfer learning, obtaining a cell detection map, inputting the cell detection map into the cell detection network model, extracting a candidate region by adopting an RPN, and carrying out position regression and classification on cells through a regression device and a classifier to finally obtain a cell prediction result. According to the method, based on transfer learning, the network model of common features of the similar data sets can be extracted through transfer learning training, the problem of insufficient training samples is solved, and meanwhile, the accuracy of model output is ensured.

Description

technical field [0001] The invention relates to the technical field of cell image detection, in particular to a cell image detection method based on migration learning. Background technique [0002] With the advent of the information age, in medicine, the collection, storage and processing of images can be realized with the help of computer systems. Digital medical images make data storage more convenient and safe, and at the same time promote the development of digital image processing. With the advancement of the times and technology, the processing of medical images is developing in the direction of digitization, automation and intelligence. [0003] Microscopic examination is a detection method widely used in clinical medicine. The number and morphological changes of cells in the sample to be tested can provide a lot of useful information for clinical diagnosis. In order to improve the accuracy of diagnosis, it is generally necessary to inspect a large sample area when ...

AI Technical Summary

Problems solved by technology

[0006] After obtaining the high-resolution image data, it is finally necessary to consider the detection of cells. According to different detection methods, the current cell detection methods are generally divided into traditional cell detection methods and cell detection methods based on deep learning; deep learning integrates feature learning into the In the process of building the model, it can learn richer feature information from massive training data sets. Therefore, in application scenarios that meet specific conditions, it has achieved recognition or classification performance that surpasses traditional algorithms. However, the current deep learning algorithm It is very dependent on the size of the training set. When the training samples are insufficient, it will lead to overfitting of the network model, and effective features cannot be extracted. In the field of medical image processing, due to the difficulty in obtaining samples, the difficulty in labeling, and the lack of researchers Due to various reasons, data sets are often very scarce, among which the lack of large field of view and high-resolution image data is the most prominent, which is a major problem in cell image detection in medicine.

Images