934 papers - avg viability 6.2
Current research in medical AI is increasingly focused on enhancing clinical decision-making through advanced retrieval and reasoning frameworks. Recent work is transforming static pathology archives into dynamic, searchable libraries, enabling real-time case exploration and cohort construction, which drastically reduces the time and cost associated with traditional methods. In parallel, the development of specialized large language models for tasks like Alzheimer's disease prediction and surgical decision support is demonstrating improved accuracy and interpretability, crucial for clinical applications. Additionally, innovative evaluation metrics are being introduced to ensure the reliability of vision-language models in pathology, addressing concerns over hallucinations and output consistency. The emergence of lightweight, offline models is also noteworthy, as they promise to mitigate privacy risks and operational costs associated with cloud-based systems. Collectively, these advancements signal a shift toward more integrated, efficient, and clinically relevant AI solutions that can directly impact patient care and operational workflows in healthcare settings.
PathoScribe transforms static pathology archives into an interactive, LLM-driven living library for enhanced clinical decision-making.
TAP-GPT is a domain-adapted tabular LLM for accurate Alzheimer's disease prediction using multimodal biomedical data.
Surg-R1 is a hierarchical reasoning foundation model designed to enhance surgical decision support through interpretable predictions and clinical validation.
PathGLS is a novel evaluation framework for pathology vision-language models that quantifies hallucination rates and robustness without ground truth.
AdaRAG-CT enhances automated radiology report generation by overcoming visual representation bottlenecks with adaptive retrieval techniques.
ACE-LoRA enhances medical vision-language models with parameter-efficient adaptation for improved diagnostic accuracy.
Meissa is a lightweight, offline multi-modal medical language model that enhances clinical decision-making with agentic capabilities.
MetaboNet offers a standardized, consolidated dataset for type 1 diabetes management, poised to become the benchmark for AI-driven diabetes intervention technologies.
Real-time fetal ultrasound analysis on mobile devices, outperforming larger models with a novel knowledge distillation technique, enabling accessible prenatal care.
PACE-RAG is a personalized drug recommendation system that integrates patient context with clinical prescribing patterns for optimal treatment decisions.