PulseAugur
EN
LIVE 04:10:54

AI models improve factuality in 3D brain and liver MRI report generation

Researchers have developed new methods for generating more accurate radiology reports from 3D medical imaging. One approach, PIRTA, uses retrieval-augmented generation to improve the factuality of brain MRI reports by leveraging similar image-text pairs to ground LLM generation. Another method, MRI2Rep, focuses on 3D liver MRI, converting free-text reports into structured diagnostic sequences and achieving high case-level sensitivity and lesion-level F1 scores, though human evaluation showed AI reports were rated as clinically acceptable less often than original reports. AI

IMPACT Advances in AI-powered medical report generation could lead to faster, more accurate diagnoses and reduce radiologist workload.

RANK_REASON Two distinct research papers on arXiv detailing novel methods for AI-driven medical report generation from 3D MRI scans.

Read on arXiv cs.LG →

AI-generated summary · Google Gemini · from 2 sources. How we write summaries →

AI models improve factuality in 3D brain and liver MRI report generation

COVERAGE [2]

  1. arXiv cs.LG TIER_1 English(EN) · Junhyeok Lee, Yujin Oh, Dahyoun Lee, Hyon Keun Joh, Chul-Ho Sohn, Sung Hyun Baik, Cheol Kyu Jung, Jung Hyun Park, Kyu Sung Choi, Byung-Hoon Kim, Jong Chul Ye ·

    Improving Factuality of 3D Brain MRI Report Generation with Paired Image-domain Retrieval and Text-domain Augmentation

    arXiv:2411.15490v2 Announce Type: replace-cross Abstract: Acute ischemic stroke (AIS) requires time-critical decision-making, where inaccurate interpretation of neuroimaging findings can lead to irreversible disability. Diffusion-weighted imaging (DWI) and apparent diffusion coef…

  2. arXiv cs.CV TIER_1 English(EN) · Xinran Li, Junlin Yang, Annabella Shewarega, Zongwei Zhou, Julius Chapiro, James S. Duncan, Lawrence H. Staib ·

    MRI2Rep: Autoregressive Structured Report Generation for 3D Liver MRI

    arXiv:2606.25279v1 Announce Type: new Abstract: Manual reporting of 3D MRI studies is time-consuming, yet end-to-end structured report generation for 3D liver MRI remains underexplored due to volumetric complexity and scarce paired data. We propose MRI2Rep, an autoregressive fram…