Method, system, medium, equipment and terminal for predicting efficacy of locally advanced rectal cancer
By using multi-parameter MRI images and multi-target ROI methods, combined with radiomics feature extraction and ensemble learning models, the limitations of existing technologies have been overcome, enabling more accurate prediction of treatment efficacy for locally advanced rectal cancer, thus improving treatment outcomes and patients' quality of life.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- WEST CHINA HOSPITAL SICHUAN UNIV
- Filing Date
- 2022-07-27
- Publication Date
- 2026-07-21
AI Technical Summary
Existing technologies for predicting the efficacy of treatment in locally advanced rectal cancer have limitations due to the use of a single imaging modality and a single region of interest (ROI). They do not make full use of ADC and DWI-weighted MRI images and neglect the distinction between tumors within the rectal mesentery and positive lymph nodes.
Using multi-parameter MRI images and multi-target ROIs, DWI-MRI and ADC-MRI were registered to T2w-MRI through image registration. Radiomics features were extracted using the pyradiomics toolkit, and a pathological response prediction model was built using a stacking ensemble learning approach.
It improves the accuracy of efficacy prediction, enabling more accurate prediction of responses in patients with locally advanced rectal cancer, guiding personalized treatment plans, and improving patients' quality of life and survival.
Smart Images

Figure CN115205276B_ABST