Department of Biomedical Engineering, Worcester Polytechnic Institute, Worcester, MA, 01609, USA.
Vascular Biology Program, Boston Children's Hospital, Boston, MA, 02115, USA.
Sci Rep. 2023 Aug 19;13(1):13525. doi: 10.1038/s41598-023-40652-1.
Fine needle aspiration (FNA) biopsy of thyroid nodules is a safe, cost-effective, and accurate diagnostic method for detecting thyroid cancer. However, about 10% of initial FNA biopsy samples from patients are non-diagnostic and require repeated FNA, which delays the diagnosis and appropriate care. On-site evaluation of the FNA sample can be performed to filter out non-diagnostic FNA samples. Unfortunately, it involves a time-consuming staining process, and a cytopathologist has to be present at the time of FNA. To bypass the staining process and expert interpretation of FNA specimens at the clinics, we developed a deep learning-based ensemble model termed FNA-Net that allows in situ screening of adequacy of unstained thyroid FNA samples smeared on a glass slide which can decrease the non-diagnostic rate in thyroid FNA. FNA-Net combines two deep learning models, a patch-based whole slide image classifier and Faster R-CNN, to detect follicular clusters with high precision. Then, FNA-Net classifies sample slides to be non-diagnostic if the total number of detected follicular clusters is less than a predetermined threshold. With bootstrapped sampling, FNA-Net achieved a 0.81 F1 score and 0.84 AUC in the precision-recall curve for detecting the non-diagnostic slides whose follicular clusters are less than six. We expect that FNA-Net can dramatically reduce the diagnostic cost associated with FNA biopsy and improve the quality of patient care.
甲状腺结节细针穿刺（FNA）活检是一种安全、经济有效的诊断甲状腺癌的方法。然而，约有 10%的初始 FNA 活检样本是非诊断性的，需要重复 FNA，这会延迟诊断和适当的治疗。可以对 FNA 样本进行现场评估，以筛选出非诊断性的 FNA 样本。不幸的是，这涉及到一个耗时的染色过程，而且在进行 FNA 时必须有细胞病理学家在场。为了绕过 FNA 标本的染色过程和专家解释在诊所，我们开发了一种基于深度学习的集成模型，称为 FNA-Net，可以在不染色的情况下对涂在载玻片上的甲状腺 FNA 样本进行原位筛选，从而降低甲状腺 FNA 的非诊断率。FNA-Net 结合了两种深度学习模型，即基于补丁的全幻灯片图像分类器和 Faster R-CNN，以高精度检测滤泡簇。然后，如果检测到的滤泡簇总数小于预定阈值，FNA-Net 将样本玻片分类为非诊断性。通过自举抽样，FNA-Net 在检测滤泡簇少于 6 个的非诊断性幻灯片的精度-召回曲线中达到了 0.81 的 F1 评分和 0.84 的 AUC。我们期望 FNA-Net 可以显著降低与 FNA 活检相关的诊断成本，并提高患者护理质量。
