COVID-19 Classification on Segmented lungs (+ve 410; -ve 500) (test)
99.26Accuracy[7]
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| [7]Trade-offs=Test dataset too low, contains only 91 images possibly leading to loss of generalization. Performed data augmentation further leading to loss of generalization.2026.04 | 99.26 | 98.82 | 98.53 | 99.25 | |
| [22]Trade-offs=Dataset small, average performance.2026.04 | 84.23 | — | — | 81.31 | |
| [20]Trade-offs=Impractical from utility point of view as accuracy drops drastically for patients having mild or moderate symptoms. Study is based on smart data, could lead to loss of generalization.2026.04 | 81 | 76.8 | — | 80 |