Department of Pathology, Memorial Sloan Kettering Cancer Center, New York, New York, USA.
Department of Pathology, Memorial Sloan Kettering Cancer Center, New York, New York, USA,
Acta Cytol. 2021;65(4):324-329. doi: 10.1159/000512097. Epub 2020 Dec 16.
From cell phones to aerospace, artificial intelligence (AI) has wide-reaching influence in the modern age. In this review, we discuss the application of AI solutions to an equally ubiquitous problem in cytopathology - thyroid fine needle aspiration biopsy (FNAB). Thyroid nodules are common in the general population, and FNAB is the sampling modality of choice. The resulting prevalence in the practicing pathologist's daily workload makes thyroid FNAB an appealing target for the application of AI solutions.
This review summarizes all available literature on the application of AI to thyroid cytopathology. We follow the evolution from morphometric analysis to convolutional neural networks. We explore the application of AI technology to different questions in thyroid cytopathology, including distinguishing papillary carcinoma from benign, distinguishing follicular adenoma from carcinoma and identifying non-invasive follicular thyroid neoplasm with papillary-like nuclear features by key words and phrases. Key Messages: The current literature shows promise towards the application of AI technology to thyroid fine needle aspiration biopsy. Much work is needed to define how this powerful technology will be of best use to the future of cytopathology practice.
人工智能在现代社会的影响无处不在，从手机到航空航天领域都有涉及。在这篇综述中，我们讨论了人工智能解决方案在细胞病理学中同样普遍存在的问题-甲状腺细针抽吸活检（FNAB）中的应用。甲状腺结节在普通人群中很常见，FNAB 是首选的采样方式。由于病理医生日常工作量大，甲状腺 FNAB 成为人工智能解决方案应用的理想目标。
本文综述了所有关于人工智能在甲状腺细胞病理学中应用的文献。我们从形态计量分析到卷积神经网络的发展进行了探讨。我们还探讨了人工智能技术在甲状腺细胞学不同问题中的应用，包括从良性病变中区分甲状腺乳头状癌、从良性病变中区分滤泡性腺瘤和癌，以及通过关键词和短语识别具有甲状腺滤泡上皮肿瘤伴乳头状核特征的非浸润性滤泡性甲状腺肿瘤。
目前的文献表明，人工智能技术在甲状腺细针抽吸活检中的应用具有广阔的前景。需要进一步研究如何将这项强大的技术最有效地应用于细胞病理学实践的未来。
