🤖 AI Summary
Large language models (LLMs) deployed in clinical decision support systems (CDSS) face critical bottlenecks in accuracy, inference efficiency, and interpretability—especially under resource-constrained, privacy-sensitive, and regulatory-compliant clinical environments.
Method: We propose a lightweight, domain-adapted architecture integrating QLoRA fine-tuning with medical-domain-specific retrieval-augmented generation (RAG). Built upon Llama 3.2-3B-Instruct, it incorporates hospital-specific clinical data, ontology-aligned medical knowledge graphs, and FAISS/Chroma-based vector retrieval, coupled with 4-bit NF4 quantization for high-fidelity model compression.
Contribution/Results: Our novel QLoRA-RAG co-design achieves parameter efficiency (75% GPU memory reduction), low-latency inference (<800 ms), offline edge deployability, and clinically verifiable outputs. Evaluated on multiple medical benchmarks, it attains F1 ≥ 0.82—meeting clinical usability thresholds—and significantly improves accuracy, privacy preservation, and scalability across disease prediction, treatment recommendation, and clinical note summarization tasks.
📝 Abstract
This research paper investigates the application of Large Language Models (LLMs) in healthcare, specifically focusing on enhancing medical decision support through Retrieval-Augmented Generation (RAG) integrated with hospital-specific data and fine-tuning using Quantized Low-Rank Adaptation (QLoRA). The system utilizes Llama 3.2-3B-Instruct as its foundation model. By embedding and retrieving context-relevant healthcare information, the system significantly improves response accuracy. QLoRA facilitates notable parameter efficiency and memory optimization, preserving the integrity of medical information through specialized quantization techniques. Our research also shows that our model performs relatively well on various medical benchmarks, indicating that it can be used to make basic medical suggestions. This paper details the system's technical components, including its architecture, quantization methods, and key healthcare applications such as enhanced disease prediction from patient symptoms and medical history, treatment suggestions, and efficient summarization of complex medical reports. We touch on the ethical considerations-patient privacy, data security, and the need for rigorous clinical validation-as well as the practical challenges of integrating such systems into real-world healthcare workflows. Furthermore, the lightweight quantized weights ensure scalability and ease of deployment even in low-resource hospital environments. Finally, the paper concludes with an analysis of the broader impact of LLMs on healthcare and outlines future directions for LLMs in medical settings.