PulseAugur
EN
LIVE 13:13:41

AI models tackle template collapse and improve CT scan report generation

Researchers have developed two new AI models aimed at improving the accuracy and efficiency of generating reports from 3D CT scans. One model, CLarGen, addresses the issue of "Template Collapse" where AI models produce generic reports that miss critical findings, by decoupling detection from synthesis and improving clinical accuracy. The other model, Astra, is a generalizable foundation model trained on a large dataset that harmonizes reporting styles and improves diagnostic consistency, accelerating report drafting and enhancing completeness in clinical workflows. AI

IMPACT These models aim to improve diagnostic accuracy and efficiency in medical reporting, potentially accelerating clinical workflows and aiding in the detection of critical findings.

RANK_REASON Two distinct research papers on AI models for medical report generation were published on arXiv.

Read on arXiv cs.CL →

AI-generated summary · Google Gemini · from 5 sources. How we write summaries →

COVERAGE [5]

  1. arXiv cs.AI TIER_1 English(EN) · Tom Maye-Lasserre, Yitong Li, Bailiang Jian, Morteza Ghahremani, Benedikt Wiestler, Christian Wachinger ·

    Generating Reports or Repeating Templates? Measuring and Mitigating Template Collapse in 3D CT Report Generation

    arXiv:2605.30984v1 Announce Type: cross Abstract: Modern 3D medical vision-language models (VLMs) can generate fluent radiology-style text while exhibit critically low pathology detection and output diversity, collapsing to generic templates that under-report rare yet critical fi…

  2. arXiv cs.CL TIER_1 English(EN) · Christian Wachinger ·

    Generating Reports or Repeating Templates? Measuring and Mitigating Template Collapse in 3D CT Report Generation

    Modern 3D medical vision-language models (VLMs) can generate fluent radiology-style text while exhibit critically low pathology detection and output diversity, collapsing to generic templates that under-report rare yet critical findings. We identify this failure mode as Template …

  3. arXiv cs.CV TIER_1 English(EN) · Weicheng Dai, Chenyu Wang, Andy Li, Shantanu Ghosh, Kayhan Batmanghelich ·

    Flexible Control of 3D CT Generation via Text and Semantically-Defined Segmentation Prompts

    arXiv:2606.00967v1 Announce Type: new Abstract: Generative models for volumetric medical images have found many applications in medical imaging, ranging from data augmentation to serving as priors for inverse problems. For these applications, generating high-resolution 3D images …

  4. arXiv cs.CV TIER_1 English(EN) · Zhuhao Wang, Fang Chen, Chaohui Yu, Zihan Li, Yuchao Zheng, Jing Wang, Xuan Yang, Jia Guo, Zhenlu Yang, Xingju Zheng, Yihua Sun, Haojie Han, Xiaoxiao Qin, Zhan Feng, Wenbo Xiao, Chao Zhu, Yuehua Li, Shipeng Zhang, Hao Luo, Yunsong Peng, Fan Wang, Hongen … ·

    Astra: a generalizable report generation foundation model for 3D computed tomography

    arXiv:2605.31437v1 Announce Type: new Abstract: CT interpretation requires radiologists to review hundreds of volumetric slices per examination, making reporting time-consuming and highly expertise-dependent. Automated CT report generation offers a promising route to improving cl…

  5. arXiv cs.CV TIER_1 English(EN) · Hongen Liao ·

    Astra: a generalizable report generation foundation model for 3D computed tomography

    CT interpretation requires radiologists to review hundreds of volumetric slices per examination, making reporting time-consuming and highly expertise-dependent. Automated CT report generation offers a promising route to improving clinical efficiency, yet the field still lacks a g…