Research map / Label-efficient and robust learning / Self-supervised and contrastive
AbdCTBench: Learning Clinical Biomarker Representations from Abdominal Surface Geometry
Muhammad Ahmed Chaudhry, Suhana Bedi, Pola Lydia Lagari, Brian T Layden, William Galanter, Ayis Pyrros, Sanmi Koyejo
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| train a shared-backbone multi-task model on all 10 biomarkers instead of single-task (ablation) | ResNet-18 · Age prediction | MAE: 6.472 -> 14.529 | Hurt |
| train multi-task instead of single-task (ablation) | ResNet-18 · Mortality | AUROC: 0.839 -> 0.615 | Hurt |
| train multi-task instead of single-task (ablation) | ResNet-18 · HCC-18 | AUROC: 0.799 -> 0.590 | Hurt |
| train multi-task instead of single-task (ablation) | ResNet-34 · Age prediction | MAE: 6.486 -> 21.834 | Hurt |
| use medical-domain RadImageNet pretraining (ResNet-50) vs ImageNet ResNets | ResNet-50 · Mortality | AUROC: 0.810 vs 0.839 (ResNet-18) | No clear effect |
| use DINOv2 self-supervised pretraining (ViT-Small) vs ImageNet CNNs | ViT-Small · Mortality | AUROC: 0.811 vs 0.839 (ResNet-18) | No clear effect |
| use larger architecture (Swin Transformer-Base) vs small CNN | Swin Transformer-Base · Age prediction | MAE: 6.540 vs 6.223 (EfficientNet-B0) | No clear effect |
| stratify evaluation to male patients | EfficientNet-B0 · Age prediction | MAE: 6.63 -> 5.76 | Improved |
| evaluate on male instead of female patients | ResNet-18 · HCC-111 COPD | AUROC: 0.740 -> 0.802 | Improved |
| evaluate on female instead of male patients | ResNet-18 · HCC-18 | AUROC: 0.773 -> 0.824 | Improved |
| predict cancer comorbidity HCC-12 instead of cardio-metabolic targets | best architectures · HCC-12 | AUROC: 0.591 vs 0.848 (Calcium Score) | Hurt |
Closest papers by meaning
- Touchstone Benchmark: Are We on the Right Way for Evaluating AI Algorithms for Medical Segmentation? (NeurIPS 2024 · similarity 0.86)
- Time-to-Event Pretraining for 3D Medical Imaging (ICLR 2025 · similarity 0.85)
- Versatile Medical Image Segmentation Learned from Multi-Source Datasets via Model Self-Disambiguation (CVPR 2024 · similarity 0.84)
- How Well Do Supervised 3D Models Transfer to Medical Imaging Tasks? (ICLR 2024 · similarity 0.83)
- HIT: Estimating Internal Human Implicit Tissues from the Body Surface (CVPR 2024 · similarity 0.83)
- Why Invariance is Not Enough for Biomedical Domain Generalization and How to Fix It (NeurIPS 2026 · similarity 0.83)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
