09 September 2025 : Clinical Research
Early Detection of Lung Metastases in Breast Cancer Using YOLOv10 and Transfer Learning: A Diagnostic Accuracy Study
Hakan Gokalp Taş
ABCE 1*, Mehmet Bilge Han Taş CEF 2, Eyyup Yildiz CEF 2, Sonay Aydin AE 3
DOI: 10.12659/MSM.948195
Med Sci Monit 2025; 31:e948195
Table 1 YOLOv10 model performance and loss metrics.
| YOLOv10-N | YOLOv10-S | YOLOv10-M | YOLOv10-B | YOLOv10-L | YOLOv10-X | |
|---|---|---|---|---|---|---|
| mAP50 | 82.5% | 82.4% | 79.2% | 81.8% | 80.4% | 81.0% |
| mAP50–95 | 43.5% | 42.8% | 39.3% | 40.6% | 40.5% | 40.1% |
| Precision | 87.7% | 86.8% | 81.2% | 87.9% | 84.6% | 84.4% |
| Recall | 76.9% | 77.3% | 76.0% | 73.6% | 75.6% | 74.3% |
| F1-Score | 78.5% | 77.8% | 76.0% | 77.9% | 77.5% | 78.2% |
| Train Box Loss OM | 1.37 | 1.28 | 1.20 | 1.18 | 1.17 | 1.16 |
| Train Class Loss OM | 0.72 | 0.64 | 0.59 | 0.57 | 0.56 | 0.55 |
| Train DFL OM | 0.93 | 0.91 | 0.90 | 0.90 | 0.90 | 0.89 |
| Validation Box Loss OM | 1.63 | 1.67 | 1.73 | 1.74 | 1.72 | 1.76 |
| Validation Class Loss OM | 0.86 | 0.85 | 0.87 | 0.85 | 0.86 | 0.83 |
| Validation DFL OM | 0.99 | 0.98 | 1.02 | 1.01 | 1.02 | 1.02 |