肝癌早期辅助诊断数据集

该数据为医院在肝癌诊断与鉴别诊断过程中产生，汇集了动态增强CT/MRI、普美显MRI特异性对比剂影像、DSA造影等影像学报告以及诊断、用药、人口统计学信息等高质量多模态数据，并关联了金标准病理诊断结果。数据形态包括影像报告与半结构化影像特征和结构化、非结构化临床数据，用于开发和测试肝癌早期辅助诊断AI系统。业务场景应用于放射科、介入科和肝胆外科对肝脏局灶性病变的定性诊断。该数据集可支撑AI模型学习典型与非典型肝癌的影像学特征，与血管瘤、增生结节等良性病变进行精准鉴别，助力医生提高诊断信心与准确率，减少不必要的穿刺活检。This dataset was collected from hospitals during the process of liver cancer diagnosis and differential diagnosis. It encompasses high-quality multimodal data, including imaging reports (such as dynamic contrast-enhanced CT/MRI, Primovist (gadoxetic acid disodium)-enhanced MRI images using specific contrast agents, and digital subtraction angiography (DSA) results), diagnostic records, medication information, and demographic data, and is linked to gold-standard pathological diagnostic outcomes. The dataset includes imaging reports, semi-structured imaging features, as well as structured and unstructured clinical data, and is designed for developing and testing AI-assisted early diagnosis systems for liver cancer. Its application scenarios cover qualitative diagnosis of focal liver lesions in the departments of radiology, interventional radiology, and hepatobiliary surgery. This dataset enables AI models to learn the imaging characteristics of both typical and atypical liver cancers, and accurately distinguish them from benign lesions such as hemangiomas and hyperplastic nodules, thereby helping clinicians improve their diagnostic confidence and accuracy, and reducing unnecessary percutaneous biopsies.

搜集汇总

背景与挑战

背景概述

该数据集整合了医院肝癌诊断过程中产生的多模态数据，包括增强CT/MRI影像、临床报告及病理结果，用于开发AI辅助诊断系统。其可帮助区分肝癌与良性病变，提升诊断准确性并减少不必要的活检。