乳腺癌病理学图像数据集

该数据集包含162个完整的乳腺癌标本幻灯片图像，源自公开网站，经40倍放大扫描后，从中提取出277,524张50×50像素的子图像。数据规模大、分辨率高，具有良好的病理代表性，适用于乳腺癌的自动诊断与分析。主要用于训练医疗大模型，辅助病理图像识别，提升乳腺癌早期检测的准确性，推动医学人工智能在临床诊疗中的应用。

This dataset consists of 162 complete breast cancer specimen slide images sourced from public websites. After being scanned at 40× magnification, 277,524 50×50-pixel sub-images were extracted from these slides. Boasting large scale, high resolution and excellent pathological representativeness, the dataset is suitable for automatic diagnosis and analysis of breast cancer. It is mainly used to train medical large language models (LLMs), assist pathological image recognition, improve the accuracy of early breast cancer detection, and promote the application of medical artificial intelligence in clinical diagnosis and treatment.

搜集汇总

背景与挑战

背景概述

该数据集包含162个乳腺癌标本的病理学图像，从中提取了27万余张高分辨率子图像，适用于乳腺癌自动诊断模型的训练与分析，具有较高的医疗科研价值。