A predicted-loss based active learning approach for robust cancer pathology image analysis in the workplace.

BACKGROUND: Convolutional neural network-based image processing research is actively being conducted for pathology image analysis. As a convolutional neural network model requires a large amount of image data for training, active learning (AL) has been developed to produce efficient learning with a small amount of training data. However, existing studies have not specifically considered the characteristics of pathological data collected from the workplace. For various reasons, noisy patches can be selected instead of clean patches during AL, thereby reducing its efficiency. This study proposes an effective AL method for cancer pathology that works robustly on noisy datasets. METHODS: Our proposed method to develop a robust AL approach for noisy histopathology datasets consists of the following three steps: 1) training a loss prediction module, 2) collecting predicted loss values, and 3) sampling data for labeling. This proposed method calculates the amount of information in unlabeled data as predicted loss values and removes noisy data based on predicted loss values to reduce the rate at which noisy data are selected from the unlabeled dataset. We identified a suitable threshold for optimizing the efficiency of AL through sensitivity analysis. RESULTS: We compared the results obtained with the identified threshold with those of existing representative AL methods. In the final iteration, the proposed method achieved a performance of 91.7% on the noisy dataset and 92.4% on the clean dataset, resulting in a performance reduction of less than 1%. Concomitantly, the noise selection ratio averaged only 2.93% on each iteration. CONCLUSIONS: The proposed AL method showed robust performance on datasets containing noisy data by avoiding data selection in predictive loss intervals where noisy data are likely to be distributed. The proposed method contributes to medical image analysis by screening data and producing a robust and effective classification model tailored for cancer pathology image processing in the workplace.

Improving quality control in the routine practice for histopathological interpretation of gastrointestinal endoscopic biopsies using artificial intelligence.

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.

Overexpression of FRAT1 protein is closely related to triple-negative breast cancer.

Purpose: Triple-negative breast cancer (TNBC) is an aggressive subtype of breast cancer with a poor prognosis and a lack of targeted therapy. Overexpression of FRAT1 is thought to be associated with this aggressive subtype of cancer. Here, we performed a comprehensive analysis and assessed the association between overexpression of FRAT1 and TNBC. Methods: First, using different web-based bioinformatics platforms (TIMER 2.0, UALCAN, and GEPIA 2), the expression of FRAT1 was assessed. Then, the expression of the FRAT1 protein and hormone receptors and HER2 status were assessed by immunohistochemical analysis. For samples of tumors with equivocal immunoreactivity, we performed silver in situ hybridization of the HER2 gene to determine an accurate HER2 status. Next, we used the R package and bc-GenExMiner 4.8 to analyze the relationship between FRAT1 expression and clinicopathological parameters in breast cancer patients. Finally, we determined the relationship between FRAT1 overexpression and prognosis in patients. Results: The expression of FRAT1 in breast cancer tissues is significantly higher than in normal tissue. FRAT1 expression was significantly related to worse overall survival (P < 0.05) and was correlated with these clinicopathological features: T stage, N stage, age, high histologic grade, estrogen receptor status, progesterone receptor status, Her-2 status, TNBC status, basal-like status, CK5/6 status, and Ki67 status. Conclusion: FRAT1 was overexpressed in breast cancer compared to normal tissue, and it may be involved in the progression of breast cancer malignancy. This study provides suggestive evidence of the prognostic role of FRAT1 in breast cancer and the therapeutic target for TNBC.

MixPatch: A New Method for Training Histopathology Image Classifiers.

CNN-based image processing has been actively applied to histopathological analysis to detect and classify cancerous tumors automatically. However, CNN-based classifiers generally predict a label with overconfidence, which becomes a serious problem in the medical domain. The objective of this study is to propose a new training method, called MixPatch, designed to improve a CNN-based classifier by specifically addressing the prediction uncertainty problem and examine its effectiveness in improving diagnosis performance in the context of histopathological image analysis. MixPatch generates and uses a new sub-training dataset, which consists of mixed-patches and their predefined ground-truth labels, for every single mini-batch. Mixed-patches are generated using a small size of clean patches confirmed by pathologists while their ground-truth labels are defined using a proportion-based soft labeling method. Our results obtained using a large histopathological image dataset shows that the proposed method performs better and alleviates overconfidence more effectively than any other method examined in the study. More specifically, our model showed 97.06% accuracy, an increase of 1.6% to 12.18%, while achieving 0.76% of expected calibration error, a decrease of 0.6% to 6.3%, over the other models. By specifically considering the mixed-region variation characteristics of histopathology images, MixPatch augments the extant mixed image methods for medical image analysis in which prediction uncertainty is a crucial issue. The proposed method provides a new way to systematically alleviate the overconfidence problem of CNN-based classifiers and improve their prediction accuracy, contributing toward more calibrated and reliable histopathology image analysis.

A loss-based patch label denoising method for improving whole-slide image analysis using a convolutional neural network.

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.

Diagnostic Limitation of Fine-Needle Aspiration (FNA) on Indeterminate Thyroid Nodules Can Be Partially Overcome by Preoperative Molecular Analysis: Assessment of RET/PTC1 Rearrangement in BRAF and RAS Wild-Type Routine Air-Dried FNA Specimens.

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.

A Case of Multifocal Papillary Thyroid Carcinoma Consisting of One Encapsulated Follicular Variant with BRAF K601E Mutation and Three Conventional Types with BRAF V600E Mutation.

Multifocal papillary thyroid carcinoma (mPTC) comprises about 20-30% of PTC. In mPTC, individual tumor foci can be identical or frequently composed of different histological types including follicular, solid, tall-cell or conventional patterns. We report a case of mPTC consisting of one encapsulated follicular variant of papillary thyroid carcinoma (FVPTC) and three conventional PTCs in a 44-year-old woman. This case genetically demonstrates unique features including the simultaneous presence of the BRAF V600E (T1799A) mutation and the BRAF K601E (A1801G) mutation in conventional PTC and FVPTC, respectively.

Cytomorphological findings and histological correlation of low-grade cribriform cystadenocarcinoma of salivary gland in fine-needle aspiration: a case study.

Low-grade cribriform cystadenocarcinoma (LGCCC) of the salivary gland is a rare tumor. We report the cytologic features and histologic correlation of a patient with LGCCC. A 57-year-old man had a hardly palpable, nontender mass in the right cheek area followed over nine months. Radiologic analysis revealed a 1.2 cm multiseptated, cystic, solid nodule in an anterior superficial lobe of the right parotid gland. Fine-needle aspiration cytology revealed many irregular overlapping sheets or clusters of ductal epithelial cells forming solid, pseudopapillary, and cribriform architectures. Nuclei of the tumor cells revealed inconspicuous atypia with minimal size variation. On the basis of these findings, we confirmed a diagnosis of ductal epithelial proliferative lesion, favoring neoplasm, with uncertain malignant potential. Tumor excision was performed, revealing a tiny multicystic nodule (0.7 cm). Histopathologically, this tumor showed the characteristic morphology of LGCCC. This is the first report of cytomorphological findings of LGCCC in Korea.

Therapy-related acute leukemia with mixed phenotype and t(9;22)(q32;q11.2): a case report and review of the literature.

Therapy-related acute leukemia showing mixed phenotype is extremely rare. We report a 49-year-old woman who presented with palpable masses in her neck and back. She had received systemic chemotherapy (adriamycin and cisplatin) and radiotherapy for endometrial adenocarcinoma 7 years before. Her peripheral blood and bone marrow showed increased blasts, which coexpressed myeloid (CD13, CD33, and myeloperoxidase) and B-lymphoid antigens (CD19 and CD79a). Cytogenetic analysis showed a karyotype of 46,XX,dup(1)(q21q32),add(5)(q33),t(9;22)(q34;q11.2)[12]/47,idem,+der(22)t(9;22)[8], and BCR/ABL1 rearrangement was detected. Leukemic infiltration was also confirmed in her back mass. After induction chemotherapy with idarubicin, cytarabine, and imatinib, she achieved complete remission. Only 2 cases of therapy-related acute leukemia with mixed phenotype have been reported so far: one with hyperploidy and the other with t(1;21)(p36;q22). To the best of our knowledge, this is the first case of therapy-related acute leukemia with mixed phenotype and t(9;22) as well as extramedullary leukemic infiltrations.

Primary pure squamous cell carcinoma of the thyroid: report and histogenic consideration of a case involving a BRAF mutation.

Primary squamous cell carcinoma of the thyroid (SCC-T) is extremely rare. Its clinical presentation is similar to that of anaplastic carcinoma. Metastasis or extension from the head and neck area should be ruled out, as patients with SCC-T have a poorer prognosis than patients who have a thyroid extension from an adjacent tumor. An 87-year-old man presented with a longstanding painless mass in the right thyroid and had experienced 2 months of pain upon swallowing. A right lobectomy was performed with resection of thyroid cartilage, cricoid cartilage, a portion of the first to third tracheal ring and the right neck lymph node. A histological examination revealed pure SCC. The tumor cells showed diffuse immunoreactivity to CK5/6, CK19 and p63. Immunoreactivity to EMA and p53 was focally positive. TTF-1, galectin 3 and thyroglobulin immunoreactivity was restricted to the non-neoplastic thyroid tissue. Both tumor cells and non-neoplastic follicular cells were negative for CD5. The MIB-1 index was 36%. DNA extracted from the tumor identified a BRAF V600E mutation in exon 15 and a BRAF G468A mutation in exon 11, whereas DNA from non-tumorous cells did not contain a mutation. These molecular findings may suggest a direct transformation from papillary carcinoma to SCC-T.

Fluoroscopy and sonographic guided injection of obliquus capitis inferior muscle in an intractable occipital neuralgia.

Occipital neuralgia is a form of headache that involves the posterior occiput in the greater or lesser occipital nerve distribution. Pain can be severe and persistent with conservative treatment. We present a case of intractable occipital neuralgia that conventional therapeutic modalities failed to ameliorate. We speculate that, in this case, the cause of headache could be the greater occipital nerve entrapment by the obliquus capitis inferior muscle. After steroid and local anesthetic injection into obliquus capitis inferior muscles under fluoroscopic and sonographic guidance, the visual analogue scale was decreased from 9-10/10 to 1-2/10 for 2-3 weeks. The patient eventually got both greater occipital neurectomy and partial resection of obliquus capitis inferior muscles due to the short term effect of the injection. The successful steroid and local anesthetic injection for this occipital neuralgia shows that the refractory headache was caused by entrapment of greater occipital nerves by obliquus capitis inferior muscles.