HiSBreast

This is a dataset of breast ultrasound images of patients collected from Ca Mau province general hospital - Vietnam. This data set includes 972 samples, and each sample has 03 parts: ultrasound image, description of signs and characteristics of breast tumors based on that image, and diagnosis results of doctors. All information is presented in Vietnamese. The data set is organized into 04 folders: - The "json" folder contains original information about the data such as: images in based64 format, descriptions, diagnoses, patient codes,... - The "image" folder contains images that have been decoded from base64 format in json file. - The "diagnoses" folder contains doctor's diagnoses corresponding to the images in the image folder. - "descriptions" folder: contains descriptions of signs and disease characteristics corresponding to the images in the image folder.本数据集收录了源自越南金瓯省综合医院的患者乳腺超声影像数据。本数据集共包含972份样本，每份样本涵盖三部分内容：超声影像、基于该影像的乳腺肿瘤体征与特征描述，以及医师诊断结果。所有信息均以越南语呈现。 本数据集分为四个文件夹： - "json"文件夹：存放数据集的原始信息，包括base64（Base64）格式的影像、描述文本、诊断结果、患者编号等。 - "image"文件夹：存放从"json"文件内的base64格式数据解码得到的影像文件。 - "diagnoses"文件夹：存放与"image"文件夹内影像对应的医师诊断结果。 - "descriptions"文件夹：存放与"image"文件夹内影像对应的体征与疾病特征描述文本。

搜集汇总

背景与挑战

背景概述

HiSBreast是一个包含972个样本的乳腺超声图像数据集，采集自越南Ca Mau省总医院，每个样本均包含超声图像、肿瘤特征描述和医生诊断结果，所有信息以越南语呈现。数据集采用多模态结构，分为图像、描述、诊断和原始JSON四个文件夹，适用于乳腺癌和医学超声相关研究。