AI analysis method and system based on unidirectional mirror image

Abstract

The present invention provides an AI analysis method and system based on CT scan images, which relates to the field of AI technology, including using a pre-trained deep convolutional neural network to extract multi-scale features of CT scan images, and performing feature fusion through an adaptive feature weight allocation module. A dual-branch detection network is used to detect lesion areas on the fused feature map, and the detected lesion areas are classified. Three-dimensional reconstruction is performed based on a continuous CT scan image sequence, and the detection results are mapped to a three-dimensional model to achieve accurate positioning and three-dimensional visualization of the lesion area, and its three-dimensional coordinate position information is recorded. The present invention can improve the accuracy and efficiency of CT scan image analysis and assist doctors in diagnosis.

Images