{"ID":2830997,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2512.08216","arxiv_id":"2512.08216","title":"Tumor-anchored deep feature random forests for out-of-distribution detection in lung cancer segmentation","abstract":"Accurate segmentation of lung tumors from 3D computed tomography (CT) scans is essential for automated treatment planning and response assessment. Despite self-supervised pretraining on numerous datasets, state-of-the-art transformer backbones remain susceptible to out-of-distribution (OOD) inputs, often producing confidently incorrect segmentations with potential for risk in clinical deployment. Hence, we introduce RF-Deep, a lightweight post-hoc random forests-based framework that leverages deep features trained with limited outlier exposure, requiring as few as 40 labeled scans (20 in-distribution and 20 OOD), to improve scan-level OOD detection. RF-Deep repurposes the hierarchical features from the pretrained-then-finetuned segmentation backbones, aggregating features from multiple regions-of-interest anchored to predicted tumor regions to capture OOD likelihood. We evaluated RF-Deep on 2,232 CT volumes spanning near-OOD (pulmonary embolism, COVID-19 negative) and far-OOD (kidney cancer, healthy pancreas) datasets. RF-Deep achieved AUROC \u003e~93 on the challenging near-OOD datasets, where it outperformed the next best method by 4--7 percentage points, and produced near-perfect detection (AUROC \u003e~99) on far-OOD datasets. The approach also showed transferability to two blinded validation datasets under the ensemble configuration (COVID-19 positive and breast cancer; AUROC \u003e~94). RF-Deep maintained consistent performance across backbones of different depths and pretraining strategies, demonstrating applicability of post-hoc detectors as a safety filter for clinical deployment of tumor segmentation pipelines.","short_abstract":"Accurate segmentation of lung tumors from 3D computed tomography (CT) scans is essential for automated treatment planning and response assessment. Despite self-supervised pretraining on numerous datasets, state-of-the-art transformer backbones remain susceptible to out-of-distribution (OOD) inputs, often producing conf...","url_abs":"https://arxiv.org/abs/2512.08216","url_pdf":"https://arxiv.org/pdf/2512.08216v3","authors":"[\"Aneesh Rangnekar\",\"Harini Veeraraghavan\"]","published":"2025-12-09T03:49:50Z","proceeding":"eess.IV","tasks":"[\"eess.IV\",\"cs.CV\",\"cs.LG\"]","methods":"[\"Transformer\"]","has_code":false}