Computer-Aided Pathology Image Classification and Segmentation Joint Analysis Model
Shuhan Yang, Lin Chen · 2025
With the development of image analysis technology, computer-aided medical image analysis has become an important research field and has been widely used in disease prediction, case analysis, and image enhancement. In this study, the novel joint analysis model is proposed. At the theoretical and modeling level, the model includes two key innovations: a new small-sample image classification algorithm and a new priori-driven deep learning-guided image segmentation algorithm. In the image classification stage, this study used kernel function-based feature scaling technology and multi-modal information fusion methods to retain details to the greatest extent and achieve accurate classification of pathological images in small sample conditions. In the image segmentation stage, this study constructed a new segmentation algorithm through a curvature regularization model and an improved deep neural network, which improved the smoothness and detail retention capabilities of image segmentation. This study is tested on two sets of mainstream data sets, including 800 classification samples and 500 segmentation samples. The experimental results show that compared with the existing methods, the model has improved the classification accuracy by 12.5% and the segmentation IoU index by 15.2 %. The experimental results intuitively verify the accuracy of the proposed algorithm.