This paper proposes a deep learning-based patch label denoising method (LossDiff) for improving the classification of whole-slide images of cancer using a convolutional neural network (CNN). Automated whole-slide image classification is often challenging, requiring a large amount of labeled data. Pathologists annotate the region of interest by marking malignant areas, which pose a high risk of introducing patch-based label noise by involving benign regions that are typically small in size within the malignant annotations, resulting in low classification accuracy with many Type-II errors. To overcome this critical problem, this paper presents a simple yet effective method for noisy patch classification. The proposed method, validated using stomach cancer images, provides a significant improvement compared to other existing methods in patch-based cancer classification, with accuracies of 98.81%, 97.30% and 89.47% for binary, ternary, and quaternary classes, respectively. Moreover, we conduct several experiments at different noise levels using a publicly available dataset to further demonstrate the robustness of the proposed method. Given the high cost of producing explicit annotations for whole-slide images and the unavoidable error-prone nature of the human annotation of medical images, the proposed method has practical implications for whole-slide image annotation and automated cancer diagnosis.

Background: Colorectal and gastric cancer are major causes of cancer-related deaths. In Korea, gastrointestinal (GI) endoscopic biopsy specimens account for a high percentage of histopathologic examinations. Lack of a sufficient pathologist workforce can cause an increase in human errors, threatening patient safety. Therefore, we developed a digital pathology total solution combining artificial intelligence (AI) classifier models and pathology laboratory information system for GI endoscopic biopsy specimens to establish a post-analytic daily fast quality control (QC) system, which was applied in clinical practice for a 3-month trial run by four pathologists. Methods and findings: Our whole slide image (WSI) classification framework comprised patch-generator, patch-level classifier, and WSI-level classifier. The classifiers were both based on DenseNet (Dense Convolutional Network). In laboratory tests, the WSI classifier achieved accuracy rates of 95.8% and 96.0% in classifying histopathological WSIs of colorectal and gastric endoscopic biopsy specimens, respectively, into three classes (Negative for dysplasia, Dysplasia, and Malignant). Classification by pathologic diagnosis and AI prediction were compared and daily reviews were conducted, focusing on discordant cases for early detection of potential human errors by the pathologists, allowing immediate correction, before the pathology report error is conveyed to the patients. During the 3-month AI-assisted daily QC trial run period, approximately 7–10 times the number of slides compared to that in the conventional monthly QC (33 months) were reviewed by pathologists; nearly 100% of GI endoscopy biopsy slides were double-checked by the AI models. Further, approximately 17–30 times the number of potential human errors were detected within an average of 1.2 days. Conclusions: The AI-assisted daily QC system that we developed and established demonstrated notable improvements in QC, in quantitative, qualitative, and time utility aspects. Ultimately, we developed an independent AI-assisted post-analytic daily fast QC system that was clinically applicable and influential, which could enhance patient safety.

Ko YS, et al. Korean J Pathol. 2013 Dec;47(6):592-595. http://dx.doi.org/10.4132/KoreanJPathol.2013.47.6.592

Molecular markers are helpful diagnostic tools, particularly for cytologically indeterminate thyroid nodules. Preoperative RET/PTC1 rearrangement analysis in BRAF and RAS wild-type indeterminate thyroid nodules would permit the formulation of an unambiguous surgical plan. Cycle threshold values according to the cell count for detection of the RET/PTC1 rearrangement by real-time reverse transcription-polymerase chain reaction (RT-PCR) using fresh and routine air-dried TPC1 cells were evaluated. The correlation of RET/PTC1 rearrangement between fine-needle aspiration (FNA) and paired formalin-fixed paraffin-embedded (FFPE) specimens was analyzed. RET/PTC1 rearrangements of 76 resected BRAF and RAS wild-type classical PTCs were also analyzed. Results of RT-PCR and the Nanostring were compared. When 100 fresh and air-dried TPC1 cells were used, expression of RET/PTC1 rearrangement was detectable after 35 and 33 PCR cycles, respectively. The results of RET/PTC1 rearrangement in 10 FNA and paired FFPE papillary thyroid carcinoma (PTC) specimens showed complete correlation. Twenty-nine (38.2%) of 76 BRAF and RAS wild-type classical PTCs had RET/PTC1 rearrangement. Comparison of RET/PTC1 rearrangement analysis between RT-PCR and the Nanostring showed moderate agreement with a κ value of 0.56 (p = 0.002). The RET/PTC1 rearrangement analysis by RT-PCR using routine air-dried FNA specimen was confirmed to be technically applicable. A significant proportion (38.2%) of the BRAF and RAS wild-type PTCs harbored RET/PTC1 rearrangements.