Deep Dive Readings

Reading: From vision to text: A comprehensive review of natural image captioning in medical diagnosis and radiology report generation.
By Gabriel Reale-Nosei, Elvira Amador-Domínguez, Emílio Serrano.
This paper reviews the use of natural language processing (NLP) and computer vision techniques to generate text from medical images like X-rays, CT scans, and MRIs. It explores how image processing and automated text generation work together to improve medical diagnosis and radiology report creation. As a quick reminder, NLP helps computers understand and generate human language. It covers tasks like spam filtering, sentiment analysis, machine translation, and text generation (think ChatGPT!). For this reading, focus on Chapters 1, 3, and 5, and use the reflective questions to guide your reading. Link
Reading time: ~45 min
Reflective questions
Chapter 1
- What are image captioning and medical image captioning?
- How can it combine Computer vision and NLP in medical radiology?
Chapter 3
- What role does deep learning play in medical image captioning?
- How do atttention-based methods improve image captioning compared to ED and compositional methods?
Chapter 5
- What are the current limitations of AI in medical imaging?
- What ethical concerns are associated with AI in medical imaging?
Source: Gabriel Reale-Nosei, Elvira Amador-Domínguez, Emilio Serrano, From vision to text: A comprehensive review of natural image captioning in medical diagnosis and radiology report generation, Medical Image Analysis, Volume 97, 2024, 103264, ISSN 1361-8415, Link