人工智能提高了乳腺病变组织学分类的准确性。

Artificial Intelligence Improves the Accuracy in Histologic Classification of Breast Lesions.

机构信息

Department of Pathology, Ipatimup Diagnostics, Institute of Molecular Pathology and Immunology, University of Porto, Porto, Portugal.

I3S - Instituto de Investigação e Inovação em Saúde, University of Porto, Porto, Portugal.

出版信息

Am J Clin Pathol. 2021 Mar 15;155(4):527-536. doi: 10.1093/ajcp/aqaa151.

DOI:10.1093/ajcp/aqaa151

PMID:33118594

Abstract

OBJECTIVES

This study evaluated the usefulness of artificial intelligence (AI) algorithms as tools in improving the accuracy of histologic classification of breast tissue.

METHODS

Overall, 100 microscopic photographs (test A) and 152 regions of interest in whole-slide images (test B) of breast tissue were classified into 4 classes: normal, benign, carcinoma in situ (CIS), and invasive carcinoma. The accuracy of 4 pathologists and 3 pathology residents were evaluated without and with the assistance of algorithms.

RESULTS

In test A, algorithm A had accuracy of 0.87, with the lowest accuracy in the benign class (0.72). The observers had average accuracy of 0.80, and most clinically relevant discordances occurred in distinguishing benign from CIS (7.1% of classifications). With the assistance of algorithm A, the observers significantly increased their average accuracy to 0.88. In test B, algorithm B had accuracy of 0.49, with the lowest accuracy in the CIS class (0.06). The observers had average accuracy of 0.86, and most clinically relevant discordances occurred in distinguishing benign from CIS (6.3% of classifications). With the assistance of algorithm B, the observers maintained their average accuracy.

CONCLUSIONS

AI tools can increase the classification accuracy of pathologists in the setting of breast lesions.

摘要

目的

本研究评估人工智能（AI）算法作为提高乳腺组织组织学分类准确性的工具的有用性。

方法

总共对 100 张乳腺组织的显微镜照片（测试 A）和 152 个全切片图像的感兴趣区域（测试 B）进行分类，分为 4 类：正常、良性、原位癌（CIS）和浸润性癌。评估了 4 名病理学家和 3 名病理住院医师在没有和有算法辅助的情况下的准确性。

结果

在测试 A 中，算法 A 的准确率为 0.87，在良性组的准确率最低（0.72）。观察者的平均准确率为 0.80，最相关的临床差异发生在区分良性和 CIS（7.1%的分类）。通过算法 A 的辅助，观察者的平均准确率显著提高到 0.88。在测试 B 中，算法 B 的准确率为 0.49，在 CIS 组的准确率最低（0.06）。观察者的平均准确率为 0.86，最相关的临床差异发生在区分良性和 CIS（6.3%的分类）。通过算法 B 的辅助，观察者保持了他们的平均准确率。