English

A Survey on Deep Learning and Explainability for Automatic Report Generation from Medical Images

Computer Vision and Pattern Recognition 2022-01-11 v2 Artificial Intelligence Computation and Language Machine Learning

Abstract

Every year physicians face an increasing demand of image-based diagnosis from patients, a problem that can be addressed with recent artificial intelligence methods. In this context, we survey works in the area of automatic report generation from medical images, with emphasis on methods using deep neural networks, with respect to: (1) Datasets, (2) Architecture Design, (3) Explainability and (4) Evaluation Metrics. Our survey identifies interesting developments, but also remaining challenges. Among them, the current evaluation of generated reports is especially weak, since it mostly relies on traditional Natural Language Processing (NLP) metrics, which do not accurately capture medical correctness.

Keywords

Cite

@article{arxiv.2010.10563,
  title  = {A Survey on Deep Learning and Explainability for Automatic Report Generation from Medical Images},
  author = {Pablo Messina and Pablo Pino and Denis Parra and Alvaro Soto and Cecilia Besa and Sergio Uribe and Marcelo andía and Cristian Tejos and Claudia Prieto and Daniel Capurro},
  journal= {arXiv preprint arXiv:2010.10563},
  year   = {2022}
}

Comments

Accepted for publication in ACM CSUR

R2 v1 2026-06-23T19:30:05.314Z