An incomplete multi-omics cancer subtype identification method, system, device and medium
By constructing granular feature modules and module-level skeleton representations, and combining cross-omics consensus structures and view availability masks, the problem of structural bias and missing patterns in incomplete multi-omics data is solved, improving the stability and biological consistency of cancer subtype identification.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- JIANGNAN UNIV
- Filing Date
- 2026-05-27
- Publication Date
- 2026-07-24
AI Technical Summary
Existing multi-omics cancer subtype identification methods are prone to structural bias in scenarios with incomplete data, insufficient preservation of the internal structure of omics, susceptibility of cross-omics fusion to missing patterns, and limited stability of subtype results.
By constructing a granular feature module, extracting module-level skeleton representation, performing entry-level missing data recovery, and performing skeleton consensus alignment and adaptive view weighting in the cross-omics fusion stage, combined with view availability masking for sample-level mask-aware fusion, the inherent structural consistency of multi-omics data is prioritized.
This improved the robustness, stability, and biological consistency of cancer subtype identification results, enhancing their clinical application value.
Smart Images

Figure CN122266484B_ABST