Checklist for responsible deep learning modeling of medical images based on COVID-19 detection studies

TL;DR

This paper systematically analyzes deep learning models for COVID-19 detection from medical images, identifying common mistakes and proposing a checklist to ensure responsible and reliable model development.

Contribution

It provides a comprehensive review of existing models, highlights typical errors, and introduces a checklist for responsible deep learning modeling in COVID-19 radiography.

Findings

Many models contain data and explanation errors

Lack of domain understanding leads to unreliable models

A proposed checklist improves model reliability

Abstract

The sudden outbreak and uncontrolled spread of COVID-19 disease is one of the most important global problems today. In a short period of time, it has led to the development of many deep neural network models for COVID-19 detection with modules for explainability. In this work, we carry out a systematic analysis of various aspects of proposed models. Our analysis revealed numerous mistakes made at different stages of data acquisition, model development, and explanation construction. In this work, we overview the approaches proposed in the surveyed Machine Learning articles and indicate typical errors emerging from the lack of deep understanding of the radiography domain. We present the perspective of both: experts in the field - radiologists and deep learning engineers dealing with model explanations. The final result is a proposed checklist with the minimum conditions to be met by a reliable COVID-19 diagnostic model.