Technology Category
Computer Vision: Multi-modal VisionMachine Learning: Multimodal LearningData Mining & Knowledge Management: Mining of Visual, Multimedia & Multimodal Data
Machine learning-based multimodal prognostic models integrating pathology images and high-throughput omic data for overall survival prediction in cancer: a systematic review
🤖 AI Summary
Predicting overall survival in cancer using multimodal integration of whole-slide pathology images with high-throughput omics data remains methodologically heterogeneous and clinically unvalidated.
Method: We conducted a systematic review following SWiM/PRISMA 2020 guidelines, analyzing 48 studies across 19 cancer types that employed regularized Cox regression, classical machine learning, or deep learning models. Data extraction adhered to CHARMS, and risk of bias was assessed using PROBAST+AI.
Contribution/Results: Multimodal models consistently outperformed unimodal ones (c-index: 0.550–0.857), with deep learning demonstrating enhanced predictive performance. However, all studies exhibited unclear or high risk of bias; external validation was severely limited, and clinical utility assessment was virtually absent. This is the first systematic evaluation of multimodal survival prediction integrating histopathology and multi-omics data. It reveals critical methodological shortcomings—particularly in reproducibility, generalizability, and clinical translatability—and establishes a methodological benchmark and actionable roadmap for developing robust, verifiable, and deployable multimodal prognostic models.
Technology Category
Application Category
📝 Abstract
Multimodal machine learning integrating histopathology and molecular data shows promise for cancer prognostication. We systematically reviewed studies combining whole slide images (WSIs) and high-throughput omics to predict overall survival. Searches of EMBASE, PubMed, and Cochrane CENTRAL (12/08/2024), plus citation screening, identified eligible studies. Data extraction used CHARMS; bias was assessed with PROBAST+AI; synthesis followed SWiM and PRISMA 2020. Protocol: PROSPERO (CRD42024594745).
Forty-eight studies (all since 2017) across 19 cancer types met criteria; all used The Cancer Genome Atlas. Approaches included regularised Cox regression (n=4), classical ML (n=13), and deep learning (n=31). Reported c-indices ranged 0.550-0.857; multimodal models typically outperformed unimodal ones. However, all studies showed unclear/high bias, limited external validation, and little focus on clinical utility.
Multimodal WSI-omics survival prediction is a fast-growing field with promising results but needs improved methodological rigor, broader datasets, and clinical evaluation.
Funded by NPIC, Leeds Teaching Hospitals NHS Trust, UK (Project 104687), supported by UKRI Industrial Strategy Challenge Fund.
Problem
Research questions and friction points this paper is trying to address.
Predicting cancer survival using multimodal machine learning models
Integrating pathology images and omic data for prognosis
Assessing bias and clinical utility in survival prediction
Innovation
Methods, ideas, or system contributions that make the work stand out.
Multimodal machine learning for cancer prognostication
Integration of histopathology and molecular data
Deep learning models outperform unimodal approaches
🔎 Similar Papers
No similar papers found.
