深度学习卷积神经网络可识别胃病理学中的常见损伤模式。

A Deep Learning Convolutional Neural Network Can Recognize Common Patterns of Injury in Gastric Pathology.

机构信息

From the Departments of Pathology (Drs Martin, Hanson, Gullapalli, Sethi, and Clark, and Mr Schultz) and Chemical and Biological Engineering (Dr Gullapalli), University of New Mexico, Albuquerque.

出版信息

Arch Pathol Lab Med. 2020 Mar;144(3):370-378. doi: 10.5858/arpa.2019-0004-OA. Epub 2019 Jun 27.

DOI:10.5858/arpa.2019-0004-OA

PMID:31246112

Abstract

CONTEXT.—: Most deep learning (DL) studies have focused on neoplastic pathology, with the realm of inflammatory pathology remaining largely untouched.

OBJECTIVE.—: To investigate the use of DL for nonneoplastic gastric biopsies.

DESIGN.—: Gold standard diagnoses were blindly established by 2 gastrointestinal pathologists. For phase 1, 300 classic cases (100 normal, 100 , 100 reactive gastropathy) that best displayed the desired pathology were scanned and annotated for DL analysis. A total of 70% of the cases for each group were selected for the training set, and 30% were included in the test set. The software assigned colored labels to the test biopsies, which corresponded to the area of the tissue assigned a diagnosis by the DL algorithm, termed area distribution (AD). For Phase 2, an additional 106 consecutive nonclassical gastric biopsies from our archives were tested in the same fashion.

RESULTS.—: For Phase 1, receiver operating curves showed near perfect agreement with the gold standard diagnoses at an AD percentage cutoff of 50% for normal (area under the curve [AUC] = 99.7%) and (AUC = 100%), and 40% for reactive gastropathy (AUC = 99.9%). Sensitivity/specificity pairings were as follows: normal (96.7%, 86.7%), (100%, 98.3%), and reactive gastropathy (96.7%, 96.7%). For phase 2, receiver operating curves were slightly less discriminatory, with optimal AD cutoffs reduced to 40% across diagnostic groups. The AUCs were 91.9% for normal, 100% for , and 94.0% for reactive gastropathy. Sensitivity/specificity parings were as follows: normal (73.7%, 79.6%), (95.7%, 100%), reactive gastropathy (100%, 62.5%).

CONCLUSIONS.—: A convolutional neural network can serve as an effective screening tool/diagnostic aid for gastritis.

摘要

背景

大多数深度学习（DL）研究都集中在肿瘤病理学上，而炎症病理学领域基本上尚未涉及。

目的

研究使用深度学习（DL）对非肿瘤性胃活检进行分析。

设计

由 2 名胃肠病理学家盲法建立金标准诊断。在第 1 阶段，扫描并注释了 300 例最佳显示所需病理学的经典病例（100 例正常、100 例、100 例反应性胃炎），用于 DL 分析。每组的 70%的病例被选入训练集，30%的病例被纳入测试集。软件为测试活检分配了彩色标签，这些标签对应于 DL 算法分配诊断的组织区域，称为面积分布（AD）。在第 2 阶段，以相同的方式测试了来自我们档案的另外 106 例连续非典型胃活检。

结果

对于第 1 阶段，当 AD 百分比截断值为 50%时，对于正常（曲线下面积[AUC] = 99.7%）和（AUC = 100%），以及 40%对于反应性胃炎（AUC = 99.9%），接收者操作特征曲线与金标准诊断结果非常吻合。灵敏度/特异性组合如下：正常（96.7%，86.7%）、（100%，98.3%）和反应性胃炎（96.7%，96.7%）。对于第 2 阶段，接收者操作特征曲线的区分度稍低，最佳 AD 截断值降低到诊断组的 40%。AUC 分别为正常 91.9%、94.0%。正常（73.7%，79.6%）、（95.7%，100%）、反应性胃炎（100%，62.5%）。