A deep learning diagnostic platform for diffuse large B-cell lymphoma with high accuracy across multiple hospitals.
A deep learning diagnostic platform for diffuse large B-cell lymphoma with high accuracy across multiple hospitals.
复制标题
跨多家医院的弥漫性大b细胞淋巴瘤高精度深度学习诊断平台。
DOI：
10.1038/s41467-020-19817-3
复制
发表时间：
2020-11-26
影响因子：
16.6
通讯作者：
Li S
中科院分区：
文献类型：
--
作者：
Li D;Bledsoe JR;Zeng Y;Liu W;Hu Y;Bi K;Liang A;Li S
诊断组织病理学是诊断血液系统恶性肿瘤的金标准。病理诊断需要劳动密集型阅读大量组织切片，具有等于或接近100%的高诊断准确率，以指导治疗方案，但这一要求很难满足。虽然人工智能(AI)有助于减少阅读病理切片的劳动强度，但诊断准确率尚未达到临床可用的水平。建立人工智能模型通常需要大数据集和处理样本准备和图像收集中的大变化的能力。在这里，我们建立了一个高精度的深度学习平台，由多个卷积神经网络组成，利用较小的数据集对病理图像进行分类。我们使用人工智能模型分别对三家医院的人弥漫性大B细胞淋巴瘤(DLBCL)和非DLBCL病理图像进行了分析，得到了接近100%的诊断率(A医院100%，B医院99.71%，C医院100%)。幻灯片准备和图像采集引入的技术可变性降低了AI模型在跨医院测试中的性能，但在消除它后，仍保持了100%的诊断准确率。现在，利用深度学习模型诊断DLBCL和最终诊断其他人类血液系统恶性肿瘤已经在临床上可行。用基于人工智能的工具取代诊断组织病理学需要大量的训练数据集和对样本变异性的稳健性。在这里，作者提出了一个深度学习平台，在多家医院的大型弥漫性B细胞淋巴瘤诊断中具有高精度，并在小数据集上进行了训练。
Diagnostic histopathology is a gold standard for diagnosing hematopoietic malignancies. Pathologic diagnosis requires labor-intensive reading of a large number of tissue slides with high diagnostic accuracy equal or close to 100 percent to guide treatment options, but this requirement is difficult to meet. Although artificial intelligence (AI) helps to reduce the labor of reading pathologic slides, diagnostic accuracy has not reached a clinically usable level. Establishment of an AI model often demands big datasets and an ability to handle large variations in sample preparation and image collection. Here, we establish a highly accurate deep learning platform, consisting of multiple convolutional neural networks, to classify pathologic images by using smaller datasets. We analyze human diffuse large B-cell lymphoma (DLBCL) and non-DLBCL pathologic images from three hospitals separately using AI models, and obtain a diagnostic rate of close to 100 percent (100% for hospital A, 99.71% for hospital B and 100% for hospital C). The technical variability introduced by slide preparation and image collection reduces AI model performance in cross-hospital tests, but the 100% diagnostic accuracy is maintained after its elimination. It is now clinically practical to utilize deep learning models for diagnosis of DLBCL and ultimately other human hematopoietic malignancies. Replacing diagnostic histopathology with AI-based tools requires large training datasets and robustness to sample variability. Here, the authors present a deep learning platform with high accuracy in large diffuse B-cell lymphoma diagnosis across multiple hospitals, trained on small datasets.
登录
查看更多内容
影响因子：
11.1
作者：
Khosravi P;Kazemi E;Imielinski M;Elemento O;Hajirasouliha I
通讯作者：
Hajirasouliha I
影响因子：
64.8
作者：
Esteva A;Kuprel B;Novoa RA;Ko J;Swetter SM;Blau HM;Thrun S
通讯作者：
Thrun S
影响因子：
82.9
作者：
Titano, Joseph J.;Badgeley, Marcus;Oermann, Eric K.
通讯作者：
Oermann, Eric K.
影响因子：
64.8
作者：
Mnih, Volodymyr;Kavukcuoglu, Koray;Hassabis, Demis
通讯作者：
Hassabis, Demis
影响因子：
64.8
作者：
McKinney, Scott Mayer;Sieniek, Marcin;Shetty, Shravya
通讯作者：
Shetty, Shravya