From visual saliency to anatomical explanation: generating textual diagnostic reports from brain MRIs with saliency-grounded LLMs

作者
Paul Valery Nguezet,Elie Tagne Fute,Yusuf Brima,Benoît Martin Azanguezet Quimatio,Marcellin Atemkeng
出处
期刊:Biodata Mining [BioMed Central]
标识
DOI:10.1186/s13040-026-00601-w
摘要

The opaque nature of deep learning models remains a significant barrier to their clinical adoption in medical imaging. This paper presents a systematic empirical benchmark of visual saliency attribution, atlas-based visualisation, and LLM-based reporting for brain tumour MRI classification, integrated into a unified pipeline, leveraging large language models (LLMs) to deliver human-interpretable diagnostic narratives. The proposed framework operates through three coupled stages. First, nine CNN architectures are extended with a dual-output hybrid formulation that simultaneously optimises a classification head and a segmentation head, enabling spatially richer feature learning. Second, visual saliency attribution methods, namely Grad-CAM, Grad-CAM++, and ScoreCAM, are applied to generate class-discriminative heatmaps, which are subsequently refined into coarse binary masks via an adaptive percentile thresholding pipeline. Third, the resulting masks are projected onto the Harvard–Oxford cortical atlas as an illustrative anatomical overlay, offering an approximate visual reference for the tumour’s cortical neighbourhood, and the extracted findings are encoded into a structured JSON file that conditions three LLMs (Grok3, Mistral, and LLaMA) to generate coherent, radiological-style diagnostic reports. Evaluated on a dataset of 4,834 contrast-enhanced T1-weighted brain MRI images spanning three tumour classes, InceptionResNetV2 achieved the highest classification performance, and Grad-CAM++ yielded the best segmentation overlap. Among the language models, Grok3 led in lexical diversity and coherence, while LLaMA achieved the highest readability score. By integrating visual, anatomical, and linguistic modalities into a unified pipeline, the framework attempts to produce technically grounded and meaningfully interpretable explanations, offering a step toward more transparent artificial intelligence-assisted brain tumour diagnosis.

科研通智能强力驱动
Strongly Powered by AbleSci AI
科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
www完成签到 ,获得积分10
刚刚
zuijiasunyou完成签到,获得积分10
刚刚
李钢完成签到,获得积分10
刚刚
小猪完成签到,获得积分10
1秒前
Penalty完成签到 ,获得积分10
1秒前
1秒前
1160005485完成签到 ,获得积分10
3秒前
去明天完成签到 ,获得积分10
3秒前
草根喵完成签到,获得积分10
3秒前
hyh完成签到,获得积分10
5秒前
2Y完成签到,获得积分10
5秒前
downdown完成签到,获得积分10
7秒前
Camellia完成签到 ,获得积分10
7秒前
占那个完成签到 ,获得积分10
8秒前
小杨完成签到,获得积分10
8秒前
浣熊完成签到 ,获得积分10
9秒前
橙C完成签到,获得积分10
9秒前
twebery完成签到,获得积分10
12秒前
ff完成签到 ,获得积分10
13秒前
啊泽ovo完成签到,获得积分10
14秒前
123完成签到,获得积分10
14秒前
15秒前
罗春燕完成签到 ,获得积分10
16秒前
16秒前
leehoo完成签到,获得积分10
16秒前
沉静野狼完成签到,获得积分10
17秒前
cheng完成签到 ,获得积分10
17秒前
复杂芷文完成签到,获得积分10
18秒前
勤恳的猫完成签到,获得积分10
18秒前
铁头霸霸完成签到 ,获得积分10
19秒前
乐乐乐完成签到,获得积分10
19秒前
陈鹿华完成签到 ,获得积分10
19秒前
19秒前
ncuwzq完成签到,获得积分0
20秒前
内向鼠标完成签到 ,获得积分10
21秒前
机灵的寻芹完成签到,获得积分10
21秒前
keyanxiaobaishu完成签到 ,获得积分10
21秒前
502s完成签到,获得积分10
22秒前
卷毛君完成签到,获得积分10
22秒前
24秒前
高分求助中
(应助此贴封号)【重要!!请各用户(尤其是新用户)详细阅读】【科研通的精品贴汇总】 10000
拉伸高分子熔体中缠结行为的模拟和小角中子散射研究 1700
The World Atlas of Wine 9th edition 1000
奈特基础组织学:配套组织病理学(第4版)(Netter's Essential Histology With Correlated Histopathology) 800
John McDowells Bildungstheorie 800
八旗制度の研究 600
Hidden Empire of Finance: How Wall Street Profits from Our Cities and Fuels Global Inequality 500
热门求助领域 (近24小时)
化学 材料科学 医学 生物 纳米技术 计算机科学 工程类 有机化学 化学工程 生物化学 物理 复合材料 内科学 催化作用 光电子学 人工智能 细胞生物学 无机化学 物理化学 数学
热门帖子
关注 科研通微信公众号,转发送积分 8166570
求助须知:如何正确求助?哪些是违规求助? 9559581
关于积分的说明 21750904
捐赠科研通 7757397
什么是DOI,文献DOI怎么找? 3429170
关于科研通互助平台的介绍 2553530
邀请新用户注册赠送积分活动 2473923