Multi-period outcome prediction method based on improved ResNet encoder

By combining an improved ResNet encoder and Focal Loss, the problem of insufficient feature extraction in multi-period prediction models is solved, achieving high-precision multi-period outcome prediction and improving the performance of the model and the continuity of prediction results.

CN120953660APending Publication Date: 2025-11-14XIEHE HOSPITAL ATTACHED TO TONGJI MEDICAL COLLEGE HUAZHONG SCI & TECH UNIV
0 Cites 0 Cited by

Patent Information

Application Number
CN202510986903.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-07-17
Publication Date
2025-11-14

AI Technical Summary

Technical Problem

Existing technologies lack feature extraction encoder structures for multi-class head cascade structures and loss inter-period fusion combinations, resulting in insufficient performance of multi-period prediction models, making it difficult to meet the needs of personalized follow-up and intervention.

Method used

An improved ResNet encoder is used to construct a multi-period regression prediction model through input convolutional layers, residual blocks, average pooling layers, and linear transformation layers. Combining the loss function of the multi-period regression prediction model and Focal Loss, a cascaded classification head is designed to guide the time series information and constrain the continuity of the prediction results across periods.

Benefits of technology

It enables the output of high-precision short-term, medium-term, and long-term outcome probabilities from a single acute-phase CT image, improving the performance of multi-phase outcome prediction models and ensuring the consistency and accuracy of multi-phase prediction results.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention relates to the technical field of artificial intelligence and medical image processing, in particular to a multi-stage outcome prediction method based on an improved ResNet encoder, and the method comprises the following steps: collecting a chest CT image of an acute-stage lung disease; an improved ResNet encoder module used for extracting high-dimensional image features in the chest CT image is established through an input convolution layer, four residual blocks, an average pooling layer and a linear transformation layer; and outputting short-term, medium-term and long-term outcome probabilities based on the high-dimensional image features output by the improved ResNet encoder module by using a pre-established multi-stage outcome prediction model. According to the method, a plurality of classification heads are designed to be fused in a cascade mode to realize guidance of time sequence information, an improved cross-period fusion loss function based on Focal Loss is introduced, continuous constraint among prediction results in different periods is realized, and a multi-classification-head cascade structure and a feature extraction encoder structure of a loss cross-period fusion combination are combined, so that the prediction accuracy is improved. And therefore, the performance of the multi-period prediction model can be played to the optimum.
Need to check novelty before this filing date? Find Prior Art