向可解释的细胞图像表示和宫癌查的异常评分使用巴普查
Yu Ando1, Junghwan Cho2, Nora Jee-Young Park3,4
1Department of Biomedical Science, Kyungpook National University, Daegu 41566, Republic of Korea.
Bioengineering (Basel, Switzerland)
|June 27, 2024
概括
这项研究引入了一种新的深度学习方法,用于仅使用正常细胞样本进行宫癌查. 该方法有效地识别异常细胞,改善早期检测,而不需要异常训练数据.
科学领域:
- 医疗成像医学成像
- 计算病理学计算病理学
- 医疗保健中的人工智能
背景情况:
- 宫癌查对于早期检测至关重要,但需要大量的劳动力.
- 深度学习模型显示了自动化巴氏涂抹分析的前景.
- 阶级不平衡和医疗数据的高标签成本需要替代的培训策略.
研究的目的:
- 开发一种可解释的深度学习方法,用于使用一类分类来表示宫细胞.
- 为了在训练期间不使用异常样本,在巴氏涂片图像中检测异常.
- 有效地定位和解释检测到的细胞异常.
主要方法:
- 使用变异自编码器来对宫细胞学图像进行一类分类.
- 开发了一个基于学习表现的细胞异常的评分系统.
- 采用了聚合集群,并采用了新的交叉差度指标来定位异常.
主要成果:
- 实现了0.908±0.003的操作特征曲线 (AUC) 下的面积,以区分平细胞癌 (SCC) 和正常细胞.
- 达到0.920±0.002的AUC,可以从正常细胞中区分高度状内皮损伤 (HSIL).
- 与其他聚类方法相比,证明了V测量和同质性得分的改善,增强了异常区域的隔离.
结论:
- 拟议的一类分类方法有效地检测宫细胞异常,而不需要异常训练数据.
- 可解释的深度表示和新的本地化指标有助于解释查结果.
- 该模型在外部数据集上显示出强大的性能,表明其在宫癌查中的实际应用潜力.
相关概念视频
Imaging Studies for Cardiovascular System VI: Calcium -Scoring CT
Calcium-Scoring CT ScanA calcium-scoring CT scan, also known as coronary artery calcium (CAC) scan, detects calcium deposits in the coronary arteries. This test assesses the risk of coronary artery disease (CAD), which can lead to cardiovascular events such as angina, heart failure, and sudden cardiac arrest.A calcium-scoring CT scan is generally recommended for individuals at intermediate risk of CAD without symptoms. It includes:Men aged 40-75 and women aged 50-75: Especially those with a...
Cellular Adaptation IV: Dysplasia and Metaplasia
DysplasiaDysplasia refers to abnormal changes in the size, shape, and organization of mature cells, characterized by pleomorphism, nuclear abnormalities, and increased mitotic activity. It commonly affects epithelial tissues, including the cervix, gastrointestinal tract, respiratory mucosa, and endometrium. Although it may occur alongside hyperplasia, dysplasia is not a true adaptive response but a preneoplastic change with potential to progress to cancer.When confined above the basement...


