{
  "annotation_id": "34cacded-728b-432f-9771-2c4691059174",
  "date_created": "2026-02-17T05:53:07.507000Z",
  "date_modified": "2026-02-17T05:53:07.507000Z",
  "file_hash": "a12b6f3a38bf0401946104e4b9fabb8c43f588dc069a89b8bb047888b2401084",
  "private": false,
  "record": {
    "abstract": "This study presents a unified Attention U-Net architecture trained jointly on MRI (BraTS 2021) and CT (LIDC-IDRI) datasets to investigate the generalizability of a single model across diverse imaging modalities and anatomical sites. Our proposed pipeline incorporates modality-harmonized preprocessing, attention-gated skip connections, and a modality-aware Focal Tversky loss function. To the best of our knowledge, this study is among the first to evaluate a single Attention U-Net trained simultaneously on separate MRI (BraTS) and CT (LIDC-IDRI) tumor datasets, without relying on modality-specific encoders or domain adaptation. The unified model demonstrates competitive performance in terms of Dice coefficient, IoU, and AUC on both domains, thereby establishing a robust and reproducible baseline for future research in cross-modality tumor segmentation.",
    "arxiv_id": "2601.06187",
    "authors": [
      "Nishan Rai",
      "Pushpa R. Dahal"
    ],
    "categories": [
      "cs.CV"
    ],
    "license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
    "title": "A Unified Attention U-Net Framework for Cross-Modality Tumor Segmentation in MRI and CT",
    "url": "https://arxiv.org/abs/2601.06187",
    "version": "v1"
  },
  "schema_id": "dorsal/arxiv",
  "source": {
    "execution_id": "cc9f9358-f2bd-40ce-bbfd-f71d0bf29992",
    "id": "arXiv Dataset",
    "type": "Model",
    "variant": "snapshot-2026-01-17",
    "version": "0.1.0"
  },
  "user_id": 1000002
}