# Retrieval-Augmented Multimodal Language Model with Heterogeneous Evidence Integration for Clinical Prognosis

arXiv: https://arxiv.org/abs/2610.13510 · HTML: https://mlsanity.com/paper/2610.13510/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Medical vision-language models, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 24 of its 35 references that are in the map.
- First posted on arXiv 9 Oct 2026.

## Placed in

- Multimodal LLMs / vision-language models / [Medical vision-language models](https://mlsanity.com/mllm/medical-vision-language-models/)
- Retrieval-augmented generation / [Medical and biomedical RAG](https://mlsanity.com/rag/medical-and-biomedical-rag/)

## Builds on (most cited first, 24 of 24)

- [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://mlsanity.com/paper/2005.11401/) (2020)
- [Visual Instruction Tuning](https://mlsanity.com/paper/2304.08485/) (2023)
- [Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities](https://mlsanity.com/paper/2507.06261/) (2025)
- [Large Language Models Encode Clinical Knowledge](https://mlsanity.com/paper/2212.13138/) (2022)
- [LLaVA-Med: Training a Large Language-and-Vision Assistant for Biomedicine in One Day](https://mlsanity.com/paper/2306.00890/) (2023)
- [On Layer Normalization in the Transformer Architecture](https://mlsanity.com/paper/2002.04745/) (2020)
- [Revisiting Deep Learning Models for Tabular Data](https://mlsanity.com/paper/2106.11959/) (2021)
- [MedGemma Technical Report](https://mlsanity.com/paper/2507.05201/) (2025)
- [Med-BERT: pre-trained contextualized embeddings on large-scale structured electronic health records for disease prediction](https://mlsanity.com/paper/2005.12833/) (2020)
- [BEHRT: Transformer for Electronic Health Records](https://mlsanity.com/paper/1907.09538/) (2019)
- [GEM: Empowering MLLM for Grounded ECG Understanding with Time Series and Images](https://mlsanity.com/paper/2503.06073/) (2025)
- [ECG-Chat: A Large ECG-Language Model for Cardiac Disease Diagnosis](https://mlsanity.com/paper/2408.08849/) (2024)
- [Teach Multimodal LLMs to Comprehend Electrocardiographic Images](https://mlsanity.com/paper/2410.19008/) (2024)
- [ECG Semantic Integrator (ESI): A Foundation ECG Model Pretrained with LLM-Enhanced Cardiological Text](https://mlsanity.com/paper/2405.19366/) (2024)
- [Deep neural networks can predict mortality from 12-lead electrocardiogram voltage data](https://mlsanity.com/paper/1904.07032/) (2019)
- [TimeMaster: Training Time-Series Multimodal LLMs to Reason via Reinforcement Learning](https://mlsanity.com/paper/2506.13705/) (2025)
- [EMERGE: Enhancing Multimodal Electronic Health Records Predictive Modeling with Retrieval-Augmented Generation](https://mlsanity.com/paper/2406.00036/) (2024)
- [Boosting Masked ECG-Text Auto-Encoders as Discriminative Learners](https://mlsanity.com/paper/2410.02131/) (2024)
- [Enhancing clinical decision support with physiological waveforms -- a multimodal benchmark in emergency care](https://mlsanity.com/paper/2407.17856/) (2024)
- [Q-Heart: ECG Question Answering via Knowledge-Informed Multimodal LLMs](https://mlsanity.com/paper/2505.06296/) (2025)
- [Electrocardiogram-Language Model for Few-Shot Question Answering with Meta Learning](https://mlsanity.com/paper/2410.14464/) (2024)
- [HeartBERT: A Self-Supervised ECG Embedding Model for Efficient and Effective Medical Signal Analysis](https://mlsanity.com/paper/2411.11896/) (2024)
- [High-Accuracy ECG Image Interpretation using Parameter-Efficient LoRA Fine-Tuning with Multimodal LLaMA 3.2](https://mlsanity.com/paper/2501.18670/) (2025)
- [UniPACT: A Multimodal Framework for Prognostic Question Answering on Raw ECG and Structured EHR](https://mlsanity.com/paper/2601.17916/) (2026)

## Abstract (arXiv)

Clinical prognosis requires integrating structured measurements, physiological waveforms, longitudinal history, and population context. We introduce PRISM (Population-Retrieved Inference for Structured Multimodal Prognosis), a multimodal language model that combines per-feature numerical tokenization, recency-weighted aggregation of longitudinal ECGs, and a dual-modality population memory bank. Retrieval operates in representation space: the model receives mean-pooled embeddings of similar training-cohort encounters, while no individual retrieved record enters the prompt. On the 1,443-task MDS-ED benchmark, PRISM achieves AUROCs of 86.2% for diagnosis, 93.1% for deterioration, 91.8% for ICU admission, and 92.8% for mortality. These results use a 96-hour pre-arrival window unavailable to reported baselines. With benchmark-window inputs only, PRISM reaches 0.8895 AUROC, compared with 0.8937 for UniPACT and 0.8889 for MDS-ED. A matched-input compact fusion model using the same five evidence streams and frozen encoders reaches 0.7652 AUROC, versus 0.9092 for PRISM. Replacing numerical EHR encoding with text serialization reduces AUROC by 3.66 points; removing pre-arrival history reduces it by 1.97 points; and zeroing retrieval reduces it by 0.99 points. Genuine retrieval outperforms random-neighbor retrieval by 0.97 macro-average points, although a paired task-level sign test does not establish a uniform advantage (p=0.067). On the external MC-MED cohort, supervised fine-tuning reaches 0.8611 AUROC, compared with 0.8227 for locally trained gradient boosting and 0.7841 for source-only PRISM. These results support structure-aware multimodal integration and supervised external adaptation.