LungHist700:用于肺病理学深度学习的组织图像数据集
Jorge Diosdado1, Pere Gilabert2, Santi Seguí2
1Dept. de Matemàtiques i Informàtica, Universitat de Barcelona, Barcelona, Spain. diosdado100591@hotmail.com.
Scientific data
|October 5, 2024
概括
这项研究引入了一个新的肺组织病理学图像数据集,用于分类肺癌. 深度学习模型在区分瘤类型和等级方面取得了高准确度 (81-92%),有助于早期发现肺癌.
科学领域:
- 病理学 病理学 病理学
- 医疗成像医学成像
- 计算生物学 计算生物学
背景情况:
- 准确的肺癌检测对于患者的治疗结果至关重要.
- 传统的组织病理学是缓慢的,阻碍了及时的临床决策.
- 需要多样化,高分辨率的数据集来进行高级分析.
研究的目的:
- 创建和发布一个全面的肺组织病理学图像数据集.
- 为了评估深度学习和多重实例学习的肺癌分类.
- 建立一个对肺组织样本自动分析的基准.
主要方法:
- 编制了691张高分辨率肺部组织病理学图像 (1200x1600像素) 的数据集.
- 包括45名患者的腺癌,状细胞癌和正常组织.
- 以20x和40x放大度拍摄的图像,按差异化分类.
- 应用深度神经网络和多个实例学习,将其分类为七个类.
主要成果:
- 实现了从81%到92%的分类准确度.
- 性能因分类方法和图像放大 (20x,40x) 而有所不同.
- 证明了数据集在培训和评估AI模型中的有效性.
结论:
- 开发的数据集支持自动肺癌检测和分类.
- 深度学习方法在提高诊断效率方面显示出显著的前景.
- 该资源有助于进一步研究肺部疾病的计算病理学.
更多相关视频
07:53Author Spotlight: Advancing 3D Modeling for Enhanced Diagnosis and Treatment of Pulmonary Nodules in Early-Stage Lung Cancer
Published on: October 13, 2023
1.4K
07:17Author Spotlight: Studying Spatial Protein Expression Using Agarose Embedded Lung Tissue Sections
Published on: October 6, 2023
5.3K
相关概念视频
01:47Lung Capacity
44.5K
The air in the lungs is measured in volumes and capacities. Lung volume measures reflect the amount of air taken in, released, or left over after a lung function, like a single inhalation. Lung capacity measures are sums of two or more lung volume measures.
44.5K
01:17Gross Anatomy of the Lungs
6.4K
The lungs are a pair of vital organs connected to the trachea via the left and right bronchi. The base of these organs meets the dome-shaped muscle known as the diaphragm. Encased by the pleurae, the lungs contact the mediastinum. The right lung is shorter yet wider, and has a larger volume than the left lung. The left lung has an indentation known as the cardiac notch. The superior region of the lungs is referred to as the apex, whereas the base is the lower region near the diaphragm. The...
6.4K