# M3: Conversational LLMs Simplify Secure Clinical Data Access, Understanding, and Analysis

arXiv: https://arxiv.org/abs/2507.01053 · HTML: https://mlsanity.com/paper/2507.01053/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 13 of its 41 references that are in the map.
- First posted on arXiv 27 Jun 2025.

## Builds on (most cited first, 13 of 13)

- [Large Language Models Encode Clinical Knowledge](https://mlsanity.com/paper/2212.13138/) (2022)
- [Spider: A Large-Scale Human-Labeled Dataset for Complex and Cross-Domain Semantic Parsing and Text-to-SQL Task](https://mlsanity.com/paper/1809.08887/) (2018)
- [Can LLM Already Serve as A Database Interface? A BIg Bench for Large-Scale Database Grounded Text-to-SQLs](https://mlsanity.com/paper/2305.03111/) (2023)
- [Seq2SQL: Generating Structured Queries from Natural Language using Reinforcement Learning](https://mlsanity.com/paper/1709.00103/) (2017)
- [AmbigQA: Answering Ambiguous Open-domain Questions](https://mlsanity.com/paper/2004.10645/) (2020)
- [Benchmarking Retrieval-Augmented Generation for Medicine](https://mlsanity.com/paper/2402.13178/) (2024)
- [Text-to-SQL Generation for Question Answering on Electronic Medical Records](https://mlsanity.com/paper/1908.01839/) (2019)
- [EHRSQL: A Practical Text-to-SQL Benchmark for Electronic Health Records](https://mlsanity.com/paper/2301.07695/) (2023)
- [Evaluating Cross-Domain Text-to-SQL Models and Benchmarks](https://mlsanity.com/paper/2310.18538/) (2023)
- [TrustSQL: Benchmarking Text-to-SQL Reliability with Penalty-Based Scoring](https://mlsanity.com/paper/2403.15879/) (2024)
- [Overview of the EHRSQL 2024 Shared Task on Reliable Text-to-SQL Modeling on Electronic Health Records](https://mlsanity.com/paper/2405.06673/) (2024)
- [SQLucid: Grounding Natural Language Database Queries with Interactive Explanations](https://mlsanity.com/paper/2409.06178/) (2024)
- [BiomedSQL: Text-to-SQL for Scientific Reasoning on Biomedical Knowledge Bases](https://mlsanity.com/paper/2505.20321/) (2025)

## Built on by (most cited first, 3 of 3)

- [Rethinking Tokenization for Clinical Time Series: When Less is More](https://mlsanity.com/paper/2512.05217/) (2025)
- [Reliable Curation of EHR Dataset via Large Language Models under Environmental Constraints](https://mlsanity.com/paper/2511.00772/) (2025)
- [Coding-Free and Privacy-Preserving Agentic Framework for Data-Driven Clinical Research](https://mlsanity.com/paper/2604.12258/) (2026)

## Abstract (arXiv)

Large-scale clinical databases offer opportunities for medical research, but their complexity creates barriers to effective use. The Medical Information Mart for Intensive Care (MIMIC-IV), one of the world's largest open-source electronic health record databases, traditionally requires both SQL proficiency and clinical domain expertise. We introduce M3, a system that enables natural language querying of MIMIC-IV data through the Model Context Protocol. With a single command, M3 retrieves MIMIC-IV from PhysioNet, launches a local SQLite instance or connects to hosted BigQuery, and allows researchers to pose clinical questions in plain English. We evaluated M3 using samples from the EHRSQL 2024 benchmark with two language models. On one hundred answerable questions, the proprietary Claude Sonnet 4 achieved 94% accuracy and the open-weights gpt-oss-20B (deployable locally on consumer hardware) achieved 93%; on a matched sample of one hundred unanswerable questions, where correct behavior is to abstain rather than produce SQL, gpt-oss-20B correctly abstained on 69%. Both models translate natural language into SQL, execute queries against MIMIC-IV, and return structured results alongside the underlying query for verification. Error analysis revealed that most failures stemmed from complex temporal reasoning or ambiguous question phrasing rather than fundamental architectural limitations. The comparable performance of a smaller open-weights model demonstrates that privacy-preserving local deployment is viable for sensitive clinical data analysis. M3 lowers technical barriers to critical care data analysis and is designed with security measures including OAuth2 authentication, query validation, and audit logging.