AI models achieve 98% F1-score for COVID-19 lesion prediction

Researchers have compared four deep learning segmentation architectures—Unet, PSPNet, Linknet, and FPN—each integrated with six pre-trained encoders, to predict COVID-19 lesions in CT scans. The study utilized three distinct COVID-19 CT segmentation datasets for both binary and multi-class experiments. Results showed that deep learning models achieved high accuracy, with a maximum F1-Score of 98% for binary segmentation and scores of 75% and 77% for multi-class segmentation on different datasets, highlighting AI's role in enhancing pandemic disease diagnostics. AI

Summary written by gemini-2.5-flash-lite from 1 source. How we write summaries →

IMPACT This research demonstrates AI's potential to improve diagnostic accuracy for diseases like COVID-19 through advanced image segmentation techniques.

RANK_REASON The cluster contains an academic paper detailing a comparative analysis of AI models for medical image segmentation. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• COVID-19
• DenseNet 121
• EfficientNet B0
• Inception ResNet V2
• PSPNet
• SeresNet 101
• VGG 19

• paper
• product