Development and Validation of a Deep Learning Model to Quantify Glomerulosclerosis in Kidney Biopsy Specimens.
Development and Validation of a Deep Learning Model to Quantify Glomerulosclerosis in Kidney Biopsy Specimens.
复制标题
开发和验证深度学习模型以量化肾活检标本中的肾小球硬化。
DOI：
10.1001/jamanetworkopen.2020.30939
复制
发表时间：
2021-01-04
影响因子：
13.8
通讯作者：
Gaut JP
中科院分区：
文献类型：
--
作者：
Marsh JN;Liu TC;Wilson PC;Swamidass SJ;Gaut JP
深度神经网络能否通过精确量化苏木精-伊红染色活检标本的全载玻片图像上的肾小球硬化百分比来降低不必要的供体肾丢弃的可能性？在这项对83个供体肾脏的预后研究中，深度神经网络在整个载玻片图像中分割正常和全局硬化的肾小球，以量化全局肾小球硬化百分比，其性能高于病理学家。通过合并多个切片进一步提高了模型准确性，导致错误器官丢弃的可能性降低了37%。这项研究的结果表明，深度学习方法可能有助于防止错误的器官丢弃，因为它超出了病理学家在活检标本检查中的能力。供体肾脏的长期短缺与高废弃率相结合，而这一比率与活检标本评估直接相关，这表明病理学家的重现性较差。一种用于测量肾脏活检标本图像上的全球肾小球硬化百分比（一种重要的结果预测因子）的深度学习算法可以使病理学家能够更可重复和准确地量化全球肾小球硬化百分比，从而可能挽救可能被丢弃的器官。比较病理学家使用深度学习模型量化供体肾活检标本全载玻片图像中肾小球硬化百分比的表现，并确定深度学习模型对器官丢弃率的潜在益处。该预后研究使用了从83个肾脏中取出的98个苏木精-伊红染色的冷冻切片和51个永久性供体活检标本切片获得的全切片图像。由3名委员会认证的病理学家进行的系列注释作为模型训练和评价的基础事实。肾脏活检标本的图像来自华盛顿大学数据库（检索时间为2015年6月至2017年6月）。从超过1000例病例的数据库中随机选择病例，包括代表0%至5%、6%至10%、11%至15%、16%至20%和20%以上全球肾小球硬化症的公平分布的活检标本。当使用个体和合并切片结果时，计算关于注释的相关系数（r）和均方根误差（RMSE），用于交叉验证模型预测和随叫随到病理学家对总体肾小球硬化百分比的估计。数据分析时间为2018年3月至2020年8月。从83个供肾检索的切片图像的交叉验证模型结果显示与注释的相关性更高（r = 0.916; 95% CI，0.886-0.939）（r = 0.884; 95%CI，0.825-0.923），当合并来自多个水平的肾小球计数时（r = 0.933; 95%CI，0.898-0.956），其增强。单水平的模型预测误差（RMSE，5.631; 95% CI，4.735-6.517）比随叫随到的病理学家（RMSE，6.523; 95% CI，5.191-7.783）低14%，多水平模型预测误差提高到22%（RMSE，5.094; 95% CI，3.972-6.301）。与病理学家相比，该模型将不必要的器官丢弃的可能性降低了37%。这项预后研究的结果表明，这种深度学习模型提供了一种可扩展且强大的方法来量化供体肾脏全切片图像中的全球肾小球硬化百分比。通过分析切片的多个水平，模型性能得到改善，超过了病理学家在检查供体活检标本的时间敏感设置中的能力。结果表明，深度学习模型有可能防止错误的供体器官丢弃。这项预后研究比较了病理学家和深度学习模型之间的表现，以量化肾脏活检标本图像上的肾小球硬化百分比，并评估了该模型对供体肾脏丢弃率的潜在益处。
Can a deep neural network decrease likelihood of unnecessary donor kidney discard by precisely quantifying percent global glomerulosclerosis on whole-slide images of hematoxylin-eosin–stained biopsy specimens? In this prognostic study of 83 donor kidneys, a deep neural network segmented normal and globally sclerotic glomeruli in whole-slide images to quantify percent global glomerulosclerosis with higher performance than pathologists. Model accuracy further increased by pooling multiple sections, resulting in decreased likelihood of erroneous organ discard by 37%. This study’s findings suggest that deep learning methods may help prevent erroneous organ discard by performing beyond the capacity of pathologists in biopsy specimen examination. A chronic shortage of donor kidneys is compounded by a high discard rate, and this rate is directly associated with biopsy specimen evaluation, which shows poor reproducibility among pathologists. A deep learning algorithm for measuring percent global glomerulosclerosis (an important predictor of outcome) on images of kidney biopsy specimens could enable pathologists to more reproducibly and accurately quantify percent global glomerulosclerosis, potentially saving organs that would have been discarded. To compare the performances of pathologists with a deep learning model on quantification of percent global glomerulosclerosis in whole-slide images of donor kidney biopsy specimens, and to determine the potential benefit of a deep learning model on organ discard rates. This prognostic study used whole-slide images acquired from 98 hematoxylin-eosin–stained frozen and 51 permanent donor biopsy specimen sections retrieved from 83 kidneys. Serial annotation by 3 board-certified pathologists served as ground truth for model training and for evaluation. Images of kidney biopsy specimens were obtained from the Washington University database (retrieved between June 2015 and June 2017). Cases were selected randomly from a database of more than 1000 cases to include biopsy specimens representing an equitable distribution within 0% to 5%, 6% to 10%, 11% to 15%, 16% to 20%, and more than 20% global glomerulosclerosis. Correlation coefficient (r) and root-mean-square error (RMSE) with respect to annotations were computed for cross-validated model predictions and on-call pathologists’ estimates of percent global glomerulosclerosis when using individual and pooled slide results. Data were analyzed from March 2018 to August 2020. The cross-validated model results of section images retrieved from 83 donor kidneys showed higher correlation with annotations (r = 0.916; 95% CI, 0.886-0.939) than on-call pathologists (r = 0.884; 95% CI, 0.825-0.923) that was enhanced when pooling glomeruli counts from multiple levels (r = 0.933; 95% CI, 0.898-0.956). Model prediction error for single levels (RMSE, 5.631; 95% CI, 4.735-6.517) was 14% lower than on-call pathologists (RMSE, 6.523; 95% CI, 5.191-7.783), improving to 22% with multiple levels (RMSE, 5.094; 95% CI, 3.972-6.301). The model decreased the likelihood of unnecessary organ discard by 37% compared with pathologists. The findings of this prognostic study suggest that this deep learning model provided a scalable and robust method to quantify percent global glomerulosclerosis in whole-slide images of donor kidneys. The model performance improved by analyzing multiple levels of a section, surpassing the capacity of pathologists in the time-sensitive setting of examining donor biopsy specimens. The results indicate the potential of a deep learning model to prevent erroneous donor organ discard. This prognostic study compares performances between pathologists and a deep learning model to quantify percent global glomerulosclerosis on images of kidney biopsy specimens and assesses the potential benefit of the model on donor kidney discard rates.
登录
查看更多内容
影响因子：
2.1
作者：
Singh, Pooja;Farber, John L.;Frank, Adam M.
通讯作者：
Frank, Adam M.
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：
7.9
作者：
Djuric U;Zadeh G;Aldape K;Diamandis P
通讯作者：
Diamandis P
影响因子：
19.6
作者：
Azancot, M. Antonieta;Moreso, Francesc;Seron, Daniel
通讯作者：
Seron, Daniel
影响因子：
13.6
作者：
Ginley, Brandon;Lutnick, Brendon;Sarder, Pinaki
通讯作者：
Sarder, Pinaki