Summary

The provided text includes five distinct academic works focusing on machine learning, AI, and deep generative methods in healthcare and radiology. These papers collectively advance the state-of-the-art in interpreting medical data through natural language, particularly in the context of radiology reports generated by Large Language Models.

The first paper highlights the potential for predictive analytics in critical care by analyzing patient outcomes during hospitalization, which improves survival rates by predicting circulatory failure. Second, the research explores how models can leverage temporal patterns to enhance biomedical vision-language processing, offering new insights into medical image analysis. The third paper examines the limitations of recent language models like GPT-4 in radiology, specifically regarding image-to-text generation accuracy and reliability for medical professionals. Fourth, the fifth paper introduces MAIRA-2, a specific system grounded in radiology reports, demonstrating improved accuracy in generating and verifying medical imaging evidence. Together, these contributions demonstrate significant progress in automating and refining AI-based medical workflows.

Title

Stephanie L. Hyland

Description

Research page of Stephanie L. Hyland. Theme based on [*folio](https://github.com/bogoli/-folio) design.

Keywords

learning, daniel, castro, maria, matthew, vision, radiology, machine, title, author, year, language, processing, aditya, conference, research, medicine

NS Lookup

A 185.199.108.153, A 185.199.111.153, A 185.199.109.153, A 185.199.110.153

Dates

Created 2026-04-15

Updated 2026-04-15

Summarized 2026-04-15