Table 1 Overview of all 150 Residual U-Nets (ResU-Nets) and 30 intensity thresholds (Intens thresh) trained in this work, with their respective input images (micro-CT, micro-PET, or micro-PET-CT), and the tumour type they were trained to segment (invasive carcinoma of no special type (NST), or invasive lobular carcinoma (ILC) and ductal carcinoma in situ (DCIS))
Model input | Tumour type | # ResU-Nets | # ResU-Nets/test fold | # Intens thresh | # Intens thresh/test fold |
|---|---|---|---|---|---|
micro-CT | NST | 25 | 5 | 5 | 1 |
micro-PET | NST | 25 | 5 | 5 | 1 |
micro-PET-CT | NST | 25 | 5 | 5 | 1 |
micro-CT | ILC + DCIS | 25 | 5 | 5 | 1 |
micro-PET | ILC + DCIS | 25 | 5 | 5 | 1 |
micro-PET-CT | ILC + DCIS | 25 | 5 | 5 | 1 |
