# Automated Model Design and Benchmarking of 3D Deep Learning Models for COVID-19 Detection with Chest CT Scans

Source: [Automated Model Design and Benchmarking of 3D Deep Learning Models for COVID-19 Detection with Chest CT Scans](<https://www.sota2.com/research/paper/68fa694f1e209131cea2170f>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-25T22:57:51.173Z

Authors: Xin He, Shihao Wang, Xiaowen Chu, Shaohuai Shi, Jiangping Tang, Xin Liu, Chenggang Yan, Jiyong Zhang, Guiguang Ding

Published: 2021-01-14T03:45:01.000Z

## Abstract

The COVID-19 pandemic has spread globally for several months. Because its transmissibility and high pathogenicity seriously threaten people's lives, it is crucial to accurately and quickly detect COVID-19 infection. Many recent studies have shown that deep learning (DL) based solutions can help detect COVID-19 based on chest CT scans. However, most existing work focuses on 2D datasets, which may result in low quality models as the real CT scans are 3D images. Besides, the reported results span a broad spectrum on different datasets with a relatively unfair comparison. In this paper, we first use three state-of-the-art 3D models (ResNet3D101, DenseNet3D121, and MC3\\\_18) to establish the baseline performance on the three publicly available chest CT scan datasets. Then we propose a differentiable neural architecture search (DNAS) framework to automatically search for the 3D DL models for 3D chest CT scans classification with the Gumbel Softmax technique to improve the searching efficiency. We further exploit the Class Activation Mapping (CAM) technique on our models to provide the interpretability of the results. The experimental results show that our automatically searched models (CovidNet3D) outperform the baseline human-designed models on the three datasets with tens of times smaller model size and higher accuracy. Furthermore, the results also verify that CAM can be well applied in CovidNet3D for COVID-19 datasets to provide interpretability for medical diagnosis.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [COVID-19 3D CT classification on Covid-CTset Iran (test)](<https://www.sota2.com/research/sota/covid-19-3d-ct-classification-on-covid-ctset-iran-test.md>): COVID-19 3D CT classification; Covid-CTset Iran (test); Accuracy: 96.88.

- [COVID-19 3D CT classification on CC-CCII China (test)](<https://www.sota2.com/research/sota/covid-19-3d-ct-classification-on-cc-ccii-china-test.md>): COVID-19 3D CT classification; CC-CCII China (test); Accuracy: 88.69.

- [COVID-19 3D CT classification on MosMed Russia (test)](<https://www.sota2.com/research/sota/covid-19-3d-ct-classification-on-mosmed-russia-test.md>): COVID-19 3D CT classification; MosMed Russia (test); Accuracy: 82.29.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2101.05442>)

[PDF](<https://arxiv.org/pdf/2101.05442.pdf>)
