# MedOrchestra: A Hybrid Cloud-Local LLM Approach for Clinical Data Interpretation

arXiv: https://arxiv.org/abs/2505.23806 · HTML: https://mlsanity.com/paper/2505.23806/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 11 of its 30 references that are in the map.
- First posted on arXiv 27 May 2025.

## Builds on (most cited first, 11 of 11)

- [Toolformer: Language Models Can Teach Themselves to Use Tools](https://mlsanity.com/paper/2302.04761/) (2023)
- [Decomposed Prompting: A Modular Approach for Solving Complex Tasks](https://mlsanity.com/paper/2210.02406/) (2022)
- [Guiding Clinical Reasoning with Large Language Models via Knowledge Seeds](https://mlsanity.com/paper/2403.06609/) (2024)
- [Multimodal Large Language Model driven Radiology Report Generation with Clinical Knowledge Enhancement](https://mlsanity.com/paper/2403.06728/) (2024)
- [A Short Review of Ethical Challenges in Clinical Natural Language Processing](https://mlsanity.com/paper/1703.10090/) (2017)
- [Fine-Tuning a Local LLaMA-3 Large Language Model for Automated Privacy-Preserving Physician Letter Generation in Radiation Oncology](https://mlsanity.com/paper/2408.10715/) (2024)
- [How essential are unstructured clinical narratives and information fusion to clinical trial recruitment?](https://mlsanity.com/paper/1502.04049/) (2015)
- [Aligning Large Language Models for Clinical Tasks](https://mlsanity.com/paper/2309.02884/) (2023)
- [Surf at MEDIQA 2019: Improving Performance of Natural Language Inference in the Clinical Domain by Adopting Pre-trained Language Model](https://mlsanity.com/paper/1906.07854/) (2019)
- [Fine-Tuning In-House Large Language Models to Infer Differential Diagnosis from Radiology Reports](https://mlsanity.com/paper/2410.09234/) (2024)
- [Exploring Large Language Models in Healthcare: Insights into Corpora Sources, Customization Strategies, and Evaluation Metrics](https://mlsanity.com/paper/2502.11861/) (2025)

## Abstract (arXiv)

Deploying large language models (LLMs) in clinical settings faces critical trade-offs: cloud LLMs, with their extensive parameters and superior performance, pose risks to sensitive clinical data privacy, while local LLMs preserve privacy but often fail at complex clinical interpretation tasks. We propose MedOrchestra, a hybrid framework where a cloud LLM decomposes complex clinical tasks into manageable subtasks and prompt generation, while a local LLM executes these subtasks in a privacy-preserving manner. Without accessing clinical data, the cloud LLM generates and validates subtask prompts using clinical guidelines and synthetic test cases. The local LLM executes subtasks locally and synthesizes outputs generated by the cloud LLM. We evaluate MedOrchestra on pancreatic cancer staging using 100 radiology reports under NCCN guidelines. On free-text reports, MedOrchestra achieves 70.21% accuracy, outperforming local model baselines (without guideline: 48.94%, with guideline: 56.59%) and board-certified clinicians (gastroenterologists: 59.57%, surgeons: 65.96%, radiologists: 55.32%). On structured reports, MedOrchestra reaches 85.42% accuracy, showing clear superiority across all settings.