Detection and analysis of COVID-19 in medical images using deep learning techniques
Problem
Rapid, automatic detection of COVID-19 from X-ray and CT medical images.
Why it matters
Fast automated screening could support diagnosis during a pandemic.
What the AI did
Applied transfer learning with pre-trained CNNs (VGG16, DenseNet121, ResNet50, ResNet152) via a Fast.AI framework to classify COVID-19 in CT and X-ray images.
Results
Accuracy and F1 above 96% for CT binary classification; 99% accuracy for enhanced VGG16 on X-ray COVID-19/pneumonia detection.
Impact
High reported accuracies on retrospective image datasets; clinical deployment not stated.
Record
| Venue | Scientific Reports |
|---|---|
| Published | 2021 |
| Work type | article |
| Authors | Dandi Yang, Cristhian Martinez-Rendon, Lara Visuña, Hardev Mukeshbhai Khandhar, Chintan Bhatt, Jesús Carretero |
| Citations | 158 (OpenAlex, at import time) |
| References | 37 |
| Model or method | VGG16, DenseNet121, ResNet50, ResNet152 (Fast.AI framework) |
| Organization | not stated in source |
| Primary source | https://doi.org/10.1038/s41598-021-99015-3 |
| Open access | gold |
| Keywords | COVID-19, deep learning, X-ray, CT scan, CNN, transfer learning |
Importance rating
major — millions of lives, an industry or a field
Abstract frames the work within the COVID-19 outbreak, a pandemic-scale health crisis; fast automated screening could aid diagnosis, though results are retrospective dataset studies with no clinical deployment stated.