Hallucination in Medical Imaging AI: A Cross-Modality Analytical Framework for Taxonomy, Detection, and Mitigation under Regulatory Constraints
作者: Omar Alshahrani, Muzammil Behzad
分类: cs.AI
发布日期: 2026-06-12
💡 一句话要点
提出跨模态分析框架以解决医学影像AI中的幻觉问题
🎯 匹配领域: 支柱九：具身大模型 (Embodied Foundation Models)
关键词: 医学影像 人工智能 幻觉现象 跨模态分析 FDA监管 临床决策 人机协作
📋 核心要点
- 幻觉现象在医学影像AI中造成严重临床后果，现有方法未能有效识别和缓解这一问题。
- 提出跨模态分析框架，整合幻觉分类、检测与缓解策略，强调与FDA监管的兼容性。
- 研究表明，通用基础模型在幻觉特定基准上表现优于医学专用模型，需加强人机协作。
📝 摘要（中文）
随着AI系统在医学影像领域的快速应用，其失败模式尚未得到充分理解。其中，幻觉现象是最令人担忧的失败模式，表现为临床上看似合理但实际上错误的输出，包括虚构的解剖结构、漏诊、错误的侧别及虚构的测量等。本文综合了同行评审的研究、基准数据集和FDA的监管指导，提出了一种跨模态的幻觉分类、检测和缓解分析框架。研究重点在于如何统一现有的分类法、医学专用基础模型与通用模型的幻觉表现差异，以及有效的缓解策略与FDA监管的兼容性。研究发现，通用模型在幻觉特定基准上表现优于医学专用模型，强调了放射科医生的监督作用。
🔬 方法详解
问题定义：本文旨在解决医学影像AI中的幻觉现象，现有方法未能充分识别和管理这些临床上看似合理但实际上错误的输出，导致潜在的临床风险。
核心思路：通过建立跨模态的幻觉分类、检测与缓解框架，整合不同影像模态的研究成果，提供系统化的解决方案，并强调与FDA监管的兼容性。
技术框架：整体架构包括三个主要模块：幻觉分类、检测机制和缓解策略。分类模块整合了三种不同的分类框架，检测机制利用物理约束和人机协作，而缓解策略则结合了多种有效方法。
关键创新：提出的跨模态分析框架是本研究的核心创新，能够统一不同影像模态的幻觉管理，且强调了放射科医生在AI生成结果中的重要性。
关键设计：采用物理约束的架构设计、Chain-of-Thought提示以及人机协作的安全机制，确保在不同的失败模式下均能有效应对。
🖼️ 关键图片
📊 实验亮点
研究结果表明，通用基础模型在幻觉特定基准上表现优于医学专用模型，具体提升幅度为X%（具体数据未知）。同时，AI生成的高比例标记需要专家校正，强调了人机协作的重要性。
🎯 应用场景
该研究的潜在应用领域包括医学影像分析、临床决策支持系统和AI辅助诊断工具。通过有效管理幻觉现象，能够提高AI系统在临床环境中的可靠性和安全性，进而提升患者的治疗效果和安全性。
📄 摘要（原文）
AI systems are being deployed across medical imaging faster than their failure modes are understood. At this point in time, the failure of greatest clinical concern is hallucination: clinically plausible but factually incorrect outputs, including fabricated anatomical structures, missed findings, incorrect laterality, and invented measurements in generated reports, with direct consequences, for example, for biopsy decisions, staging, and treatment planning. This structured narrative synthesizes peer-reviewed studies, benchmark datasets, and FDA regulatory guidance across five imaging modalities to produce a cross-modality analysis of hallucination taxonomy, etiology, detection, and mitigation. Specifically, we address three questions in this study: (1) how can existing taxonomies be unified across modalities?, (2) how do medical-specialized foundation models hallucinate less than general-purpose ones?, and (3) which mitigation strategies are effective and compatible with FDA lifecycle oversight? We note that three taxonomic frameworks together cover the imaging pipeline in a way no single framework does alone. We also highlight that general-purpose foundation models outperform medical-specialized models on hallucination-specific benchmarks, indicating that narrow domain fine-tuning can introduce overfitting-induced confabulation. At the same time, the oversight of radiologists remains essential; for instance, a very high percentage of of AI-generated flags required expert correction before clinical use. Physics-informed architectural constraints, Chain-of-Thought prompting, and human-in-the-loop safeguards each address different failure modes and is effective when combined. All findings are mapped to the FDA's Total Product Lifecycle and Predetermined Change Control Plan frameworks, which treat hallucination management as a lifecycle obligation rather than a pre-deployment checklist.