Medical symptom detection method and system based on 3D variable convolution and sequential feature fusion, terminal and storage medium

Abstract

The invention provides a medical symptom detection method and system based on 3D variable convolution and time sequence feature fusion, a terminal and a storage medium. The method comprises the stepsof obtaining 3D input data or pseudo 3D input data of a medical image; constructing a variable convolutional network model, and inputting the standard medical symptom detection data into a variable convolutional neural network for training to obtain a trained variable convolutional neural network model; inputting the 3D input data or pseudo 3D input data of the medical image into a trained variable convolutional neural network model; modeling prediction data output by the variable convolutional neural network model by using time sequence fusion to form prediction data fused with time sequencefeatures based on different scales; performing multi-scale pyramid progressive fusion on the prediction data fused with the time sequence features and having different resolutions to obtain candidateboxes of medical signs; according to the method and the device, the problems of calculation efficiency of a 3D network and loss and insufficient utilization of multi-layer related information of pseudo 3D in the prior art are solved.

Description

Technical field [0001] The field of medical imaging and computer-aided technology of this application, in particular, relates to a medical sign detection method, system, terminal and storage medium based on 3D variable convolution and time series feature fusion. Background technique [0002] The detection of medical signs is a very important issue in the field of disease diagnosis. The traditional diagnosis of suspected malignant areas requires cutting the patient’s tissue from the lesion for biopsy. However, this process has high requirements on the position and angle of the slice, and the degree of trauma to the patient also needs to be considered. [0003] With the development of many medical imaging technologies and the improvement of medical equipment, an opportunity is provided to solve this problem. In the past two decades, computer vision and artificial intelligence have developed rapidly, and there have been many computer-aided diagnosis systems helping doctors to make d...

AI Technical Summary

Problems solved by technology

[0006] Aiming at the deficiencies of the existing technology, this application provides a medical sign detection method, system, terminal and storage medium based on 3D variable convolution and time series feature fusion, which solves the problem of computational efficiency of 3D networks and pseudo 3D in the prior art. Issues such as loss of relevant information at multiple levels and underutilization of

Images