JANUS: Anatomy-Conditioned Gating for Robust CT Triage Under Distribution Shift

2026-05-13 · Lavsen Dahal, Yubraj Bhandari, Geoffrey Rubin, Joseph Y. Lo arxiv

Automated CT triage requires models that are simultaneously accurate across diverse pathologies and reliable under institutional shift. While Vision Transformers provide strong visual representations, many clinically significant findings are defined by quantitative imaging biomarkers rather than appearance alone. We introduce JANUS, a physiology-guided dual-stream architecture that conditions visual embeddings on macro-radiomic priors via Anatomically Guided Gating. On the MERLIN test set (N=5082), JANUS attains macro-AUROC 0.88 and AUPRC 0.74, outperforming all reproduced baselines. It generalizes to an external dataset N=2000; AUROC 0.87), with the largest gains on findings defined by size and attenuation as well as improved calibration on both datasets. We further quantify prediction suppression using the Physiological Veto Rate (PVR), showing that under domain shift JANUS reduces high-confidence false positives substantially more often than true positives. Together, these results are consistent with physically grounded conditioning that improves both discrimination and reliability in CT triage. Code is made publicly available at github repository https://github.com/lavsendahal/janus and model weights are at https://huggingface.co/lavsendahal/janus.

📄 PDF Abstract BibTeX arXiv:2605.13813

Code (0)

등록된 구현이 없습니다.

Janus: A Benchmark for Goal-Conditioned Information Distortion in LLMs

2026-06-09 · Polydoros Giannouris, Mohsinul Kabir, Sophia Ananiadou arxiv

LLM deception is often evaluated through direct markers such as fabricated claims, explicit lies, or strategic concealment. However, many real-world misleading communications do not depend on false statements, rather, th…

Beyond Patient Invariance: Learning Cardiac Dynamics via Action-Conditioned JEPAs

2026-04-24 · Jose Geraldo Fernandes, Luiz Facury, Pedro Robles Dutenhefner, Wagner Meira arxiv

Self-supervised learning in healthcare has largely relied on invariance-based objectives, which maximize similarity between different views of the same patient. While effective for static anatomy, this paradigm is fundam…

Self-Supervised Learning

TriageSim: A Conversational Emergency Triage Simulation Framework from Structured Electronic Health Records

2026-03-02 · Dipankar Srirag, Quoc Dung Nguyen, Aditya Joshi, Padmanesan Narasimhan 외 arxiv

Research in emergency triage is restricted to structured electronic health records (EHR) due to regulatory constraints on nurse-patient interactions. We introduce TriageSim, a simulation framework for generating persona-…

Janus-Pro: Unified Multimodal Understanding and Generation with Data and Model Scaling

2025-01-29 · Xiaokang Chen, Zhiyu Wu, Xingchao Liu, Zizheng Pan 외

In this work, we introduce Janus-Pro, an advanced version of the previous work Janus. Specifically, Janus-Pro incorporates (1) an optimized training strategy, (2) expanded training data, and (3) scaling to larger model s…

Image Generation Instruction Following Text to Image Generation+2

MedObvious: Exposing the Medical Moravec's Paradox in VLMs via Clinical Triage

2026-03-24 · Ufaq Khan, Umair Nawaz, L D M S S Teja, Numaan Saeed 외 arxiv

Vision Language Models (VLMs) are increasingly used for tasks like medical report generation and visual question answering. However, fluent diagnostic text does not guarantee safe visual understanding. In clinical practi…

Visual Question Answering Medical Report Generation