SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2601.13385 · 2026

Organ-Aware Attention Improves CT Triage and Classification

Geoffrey Rubin, Joseph Lo, Lavsen Dahal, Yubraj Bhandari

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 12 functions out of this paper's own repositories and ran 8 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.

RepositoryRoleRan
lavsendahal/oracle-ct — 8 of 12
FunctionStatusWhere it lives
_inflate_tri_to_vol Ran lavsendahal/oracle-ct/models/dinov3_oracle_ct.py
code served (permissive licence) · get_code("54fb2129472508bb")
create_roi_mask Ran lavsendahal/oracle-ct/models/dinov3_oracle_ct.py
code served (permissive licence) · get_code("a921fb96f6572812")
dilate_mask_adaptive Ran lavsendahal/oracle-ct/models/dinov3_oracle_ct.py
code served (permissive licence) · get_code("ac60d552e4ba0773")
inv_sigmoid_temp Ran lavsendahal/oracle-ct/models/dinov3_oracle_ct.py
code served (permissive licence) · get_code("c16081b0511f7f02")
make_trislices Ran lavsendahal/oracle-ct/models/dinov3_oracle_ct.py
code served (permissive licence) · get_code("57e297dc68e34e85")
masked_attention_pool Ran lavsendahal/oracle-ct/models/dinov3_oracle_ct.py
code served (permissive licence) · get_code("4972a3f1e88d07ba")
masks_3d_to_tri Ran lavsendahal/oracle-ct/models/dinov3_oracle_ct.py
code served (permissive licence) · get_code("b9347bce2263acaa")
to_logit Ran lavsendahal/oracle-ct/models/dinov3_oracle_ct.py
code served (permissive licence) · get_code("f9668e44eda57a80")
OracleCT_DINOv3_MaskedUnaryAttn Not yet run lavsendahal/oracle-ct/models/dinov3_oracle_ct.py
code served (permissive licence) · get_code("f197a9d6e3a45f95")
get_all_disease_configs Not yet run lavsendahal/oracle-ct/models/dinov3_oracle_ct.py
code served (permissive licence) · get_code("6aee030c0668d7ac")
get_all_diseases Not yet run lavsendahal/oracle-ct/models/dinov3_oracle_ct.py
code served (permissive licence) · get_code("c4f5e947122507dd")
get_attention_mask_for_disease Not yet run lavsendahal/oracle-ct/models/dinov3_oracle_ct.py
code served (permissive licence) · get_code("036217d411ecfdcc")

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

There is an urgent need for triage and classification of highvolume medical imaging modalities such as computed tomography (CT), which can improve patient care and mitigate radiologist burnout. Study-level CT triage requires calibrated predictions with localized evidence; however, off-the-shelf Vision-Language Models (VLM) struggle with 3D anatomy, protocol shifts, and noisy report supervision. This study used the two largest publicly available chest CT datasets-CT-RATE and RADCHEST-CT (held-out external test set). Our carefully tuned supervised baseline (instantiated as a simple Global Average Pooling head) establishes a new supervised state of the art, surpassing all reported linear-probe VLMs. Building on this baseline, we present ORACLE-CT, an encoder-agnostic, organ-aware head that pairs Organ-Masked Attention (mask-restricted, per-organ pooling that yields spatial evidence) with Organ-Scalar Fusion (lightweight fusion of normalized volume and mean-HU cues). In the chest setting, ORACLE-CT's masked attention model achieves AUROC 0.86 on CT-RATE; in the abdomen setting, on MERLIN (30 findings), our supervised baseline exceeds a reproduced zero-shot VLM baseline obtained by running publicly released weights through our pipeline, and adding masked attention plus scalar fusion further improves performance to AUROC 0.85. Together, these results deliver state-of-the-art supervised classification performance across both chest and abdomen CT under a unified evaluation protocol. The source code is available at https://github.com/lavsendahal/oracle-ct.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2601.13385")
get_code_for_paper("2601.13385")
have("2601.13385")

Connect an agent — have() is free.