Senior Pathologist Research Articles

Lesion Localization and Pathological Diagnosis of Ovine Pulmonary Adenocarcinoma Based on MASK R-CNN.

Ovine pulmonary adenocarcinoma (OPA) is a contagious lung tumour caused by the Jaagsiekte Sheep Retrovirus (JSRV). Histopathological diagnosis is the gold standard for OPA diagnosis. However, interpretation of traditional pathology images is complex and operator dependent. The mask regional convolutional neural network (Mask R-CNN) has emerged as a valuable tool in pathological diagnosis. This study utilized 54 typical OPA whole slide images (WSI) to extract 7167 typical lesion images containing OPA to construct a Common Objects in Context (COCO) dataset for OPA pathological images. The dataset was categorized into training and test sets (8:2 ratio) for model training and validation. Mean average specificity (mASp) and average sensitivity (ASe) were used to evaluate model performance. Six WSI-level pathological images (three OPA and three non-OPA images), not included in the dataset, were used for anti-peeking model validation. A random selection of 500 images, not included in the dataset establishment, was used to compare the performance of the model with assessment by pathologists. Accuracy, sensitivity, specificity, and concordance rate were evaluated. The model achieved a mASp of 0.573 and an ASe of 0.745, demonstrating effective lesion detection and alignment with expert annotation. In Anti-Peeking verification, the model showed good performance in locating OPA lesions and distinguished OPA from non-OPA pathological images. In the random 500-image diagnosis, the model achieved 92.8% accuracy, 100% sensitivity, and 88% specificity. The agreement rates between junior and senior pathologists were 100% and 96.5%, respectively. In conclusion, the Mask R-CNN-based OPA diagnostic model developed for OPA facilitates rapid and accurate diagnosis in practical applications.

Clinical characteristics and surgical outcomes of vertebral lesions associated with tumor-induced osteomalacia: report of 16 patients and review of the literature

SummaryVertebral tumors in patients with tumor-induced osteomalacia (TIO) have a low diagnostic rate and poor postoperative outcomes. The application of 68 Ga-DOTATATE-PET/CT significantly increased the detection rate. Compared with tumor curettage, segmental resection was recommended as the preferred surgical type due to its high recovery rate.PurposeTumor-induced osteomalacia (TIO) is an acquired hypophosphatemic osteomalacia, and surgery is the first-line therapy. Most TIO tumors are found in the bones of the appendicular skeleton, cranium, and paranasal sinuses but rarely in the vertebrae. Tumor curettage and segmental resection are the two main surgical options for vertebral TIO patients. However, research on the clinical characteristics and surgical prognosis of vertebral TIO patients is rare. In the present study, for the first time, we investigated the clinical characteristics of 16 vertebral TIO patients and compared the surgical outcomes of patients who underwent surgery via two different surgical methods.MethodsThis was a retrospective cohort study. In this study, we included 16 adult TIO patients with lesions in vertebrae from Peking Union Medical College Hospital (PUMCH), all of whom underwent surgery. Baseline laboratory data were collected through medical records review. Technetium-99 m octreotide scintigraphy (99Tcm-OCT) and 68gallium-DOTA-TATE-positron emission tomography/computed tomography (68 Ga-DOTATATE-PET/CT) were conducted at the Department of Nuclear Medicine of PUMCH. The tumor histopathology was confirmed by a senior pathologist at our center.ResultsVertebral TIO patients had lower serum phosphorus and TmP/GFR and higher serum alkaline phosphatase (ALP), serum parathyroid hormone (PTH), and serum C-terminal cross-linked telopeptide of type I collagen (β-CTX) levels than the normal range. The sensitivity of 68 Ga‒DOTATATE PET/CT was 100%, significantly greater than that of 99Tcm-OCT (40%). After comparing the outcomes between the two surgical methods, we found that the recovery rate after segmental resection (62.5%) was greater than that after tumor curettage (12.5%). In the thoracic and sacral vertebrae, segmental resection surgery had a good prognosis.Conclusion68 Ga-DOTATATE PET/CT could serve as the first diagnostic tool in patients with vertebral TIO, and segmental resection could be used as the preferred surgery. This study would raise awareness of the clinical features and management of these rare vertebral TIO patients.

Development of a deep-learning model tailored for HER2 detection in breast cancer to aid pathologists in interpreting HER2-low cases.

Over 50% of breast cancer cases are "Human epidermal growth factor receptor 2 (HER2) low breast cancer (BC)", characterized by HER2 immunohistochemistry (IHC) scores of 1+ or 2+ alongside no amplification on fluorescence in situ hybridization (FISH) testing. The development of new anti-HER2 antibody-drug conjugates (ADCs) for treating HER2-low breast cancers illustrates the importance of accurately assessing HER2 status, particularly HER2-low breast cancer. In this study we evaluated the performance of a deep-learning (DL) model for the assessment of HER2, including an assessment of the causes of discordances of HER2-Null between a pathologist and the DL model. We specifically focussed on aligning the DL model rules with the ASCO/CAP guidelines, including stained cells' staining intensity and completeness of membrane staining. We trained a DL model on a multicentric cohort of breast cancer cases with HER2-IHC scores (n = 299). The model was validated on two independent multicentric validation cohorts (n = 369 and n = 92), with all cases reviewed by three senior breast pathologists. All cases underwent a thorough review by three senior breast pathologists, with the ground truth determined by a majority consensus on the final HER2 score among the pathologists. In total, 760 breast cancer cases were utilized throughout the training and validation phases of the study. The model's concordance with the ground truth (ICC = 0.77 [0.68-0.83]; Fisher P = 1.32e-10) is higher than the average agreement among the three senior pathologists (ICC = 0.45 [0.17-0.65]; Fisher P = 2e-3). In the two validation cohorts, the DL model identifies 95% [93% - 98%] and 97% [91% - 100%] of HER2-low and HER2-positive tumours, respectively. Discordant results were characterized by morphological features such as extended fibrosis, a high number of tumour-infiltrating lymphocytes, and necrosis, whilst some artefacts such as nonspecific background cytoplasmic stain in the cytoplasm of tumour cells also cause discrepancy. Deep learning can support pathologists' interpretation of difficult HER2-low cases. Morphological variables and some specific artefacts can cause discrepant HER2-scores between the pathologist and the DL model.

Reproducibility of c-Met Immunohistochemical Scoring (Clone SP44) for Non-Small Cell Lung Cancer Using Conventional Light Microscopy and Whole Slide Imaging.

Emerging therapies for non-small cell lung cancer targeting c-Met overexpression have recently demonstrated promising results. However, the evaluation of c-Met expression can be challenging. We aimed to study the inter and intraobserver reproducibility of c-Met expression evaluation. One hundred ten cases with non-small cell lung cancer (40 biopsies and 70 surgical specimens) were retrospectively selected in a single laboratory (LPCE) and evaluated for c-Met expression. Six pathologists (4 seniors and 2 juniors) evaluated the H-score and made a 3-tier classification of c-Met expression for all cases, using conventional light microscopy (CLM) and whole slide imaging (WSI). The interobserver reproducibility with CLM gave global Cohen Kappa coefficients (ƙ) ranging from 0.581 (95% CI: 0.364-0.771) to 0.763 (95% CI: 0.58-0.92) using the c-Met 3-tier classification and H-score, respectively. ƙ was higher for senior pathologists and biopsy samples. The interobserver reproducibility with WSI gave a global ƙ ranging from 0.543 (95% CI: 0.33-0.724) to 0.905 (95% CI: 0.618-1) using the c-Met H-score and 2-tier classification (≥25% 3+), respectively. ƙ for intraobserver reproducibility between CLM and WSI ranged from 0.713 to 0.898 for the c-Met H-score and from 0.600 to 0.779 for the c-Met 3-tier classification. We demonstrated a moderate to excellent interobserver agreement for c-Met expression with a substantial to excellent intraobserver agreement between CLM and WSI, thereby supporting the development of digital pathology. However, some factors (scoring method, type of tissue samples, and expertise level) affect reproducibility. Our findings highlight the importance of establishing a consensus definition and providing further training, particularly for inexperienced pathologists, for c-Met immunohistochemistry assessment in clinical practice.

New vision of HookEfficientNet deep neural network: Intelligent histopathological recognition system of non-small cell lung cancer

BackgroundEfficient and precise diagnosis of non-small cell lung cancer (NSCLC) is quite critical for subsequent targeted therapy and immunotherapy. Since the advent of whole slide images (WSIs), the transition from traditional histopathology to digital pathology has aroused the application of convolutional neural networks (CNNs) in histopathological recognition and diagnosis. HookNet can make full use of macroscopic and microscopic information for pathological diagnosis, but it cannot integrate other excellent CNN structures. The new version of HookEfficientNet is based on a combination of HookNet structure and EfficientNet that performs well in the recognition of general objects. Here, a high-precision artificial intelligence-guided histopathological recognition system was established by HookEfficientNet to provide a basis for the intelligent differential diagnosis of NSCLC. MethodsA total of 216 WSIs of lung adenocarcinoma (LUAD) and 192 WSIs of lung squamous cell carcinoma (LUSC) were recruited from the First Affiliated Hospital of Zhengzhou University. Deep learning methods based on HookEfficientNet, HookNet and EfficientNet B4–B6 were developed and compared with each other using area under the curve (AUC) and the Youden index. Temperature scaling was used to calibrate the heatmap and highlight the cancer region of interest. Four pathologists of different levels blindly reviewed 108 WSIs of LUAD and LUSC, and the diagnostic results were compared with the various deep learning models. ResultsThe HookEfficientNet model outperformed HookNet and EfficientNet B4–B6. After temperature scaling, the HookEfficientNet model achieved AUCs of 0.973, 0.980, and 0.989 and Youden index values of 0.863, 0.899, and 0.922 for LUAD, LUSC and normal lung tissue, respectively, in the testing set. The accuracy of the model was better than the average accuracy from experienced pathologists, and the model was superior to pathologists in the diagnosis of LUSC. ConclusionsHookEfficientNet can effectively recognize LUAD and LUSC with performance superior to that of senior pathologists, especially for LUSC. The model has great potential to facilitate the application of deep learning-assisted histopathological diagnosis for LUAD and LUSC in the future.

Development of an online teaching platform to improve access to postgraduate pathology training in sub-Saharan Africa.

Resource barriers to the provision of accessible training in cancer diagnosis in lower- and middle-income countries (LMICs) limit the potential of African health systems. Long-term provision via teaching visits from senior pathologists and trainee foreign placements is unsustainable due to the prohibitive costs of travel and subsistence. Emerging eLearning methods would allow pathologists to be trained by experts in a cheaper, more efficient, and more scalable way. This study aimed to develop an online teaching platform, starting with hematopathology, for trainee pathologists in sub-Saharan Africa, initially in Nairobi, Kenya, and Lusaka, Zambia. Course materials were prepared for both Canvas and the Zoom eLearning platforms using digitally scanned slides of lymph nodes and bone marrow trephines. Initial in-person visits were made to each site to establish trainee rapport and maximize engagement, evaluate different methods and course content, and obtain feedback to develop the project. The knowledge of trainees before and after course completion was used to measure initial effectiveness. Online teaching with the preferred platform is to be continued for 1 year before re-evaluation for long-term effectiveness. Canvas was selected as the preferred delivery platform as it is freely available and has good functionality to support all required tasks. Face-to-face teaching was considered optimal to establish the initial rapport necessary to maximize subsequent engagement with online teaching. Challenges have included sub-optimal internet speeds and connections and scheduling issues. Weekly online hematopathology teaching sessions using live image capture microscope sessions, Zoom, and Canvas have been delivered to students in Kenya and Zambia, with good attendance and interaction in case discussions. Our team has successfully designed and delivered an online training program in hematopathology to trainee pathologists in Kenya and Zambia, which has been ongoing for over a year. This project is now being scaled to other sub-Saharan countries and other sub-specialties.

Abstract PO4-26-09: The Advantage of Artificial-Intelligence in HER2 IHC 0 and 1+ Scoring in Breast Cancer

Abstract Background: Trastuzumab deruxtecan has substantially changed the treatment of HER2-low breast cancer, emphasizing the need for accurate differentiation between HER2 immunohistochemistry (IHC) scores of 0 and 1+. However, the current accuracy of HER2 IHC 0 and 1+ scoring in real-world is inadequate. Artificial intelligence (AI) has emerged as a potential solution to improve interpretation accuracy. We developed an AI algorithm based on whole slide images (WSI) to quantitatively assess HER2 expression and its role in interpreting HER2 IHC scores of 0 and 1+. Methods: Our three-phase AI analysis framework involved segmenting tumor areas (excluding ductal carcinoma in situ), detecting and classifying tumor cells based on membrane staining patterns, and grading HER2 IHC scores according to the 2018 ASCO/CAP HER2 guideline. The AI tool was trained using 6012 patches annotated by experienced pathologists. Performance evaluation was conducted using a test dataset comprising 265 slides. A total of 141 HER2 IHC slides (27 IHC 0 and 114 IHC 1+) from patients diagnosed with invasive breast cancer at Fudan University Shanghai Cancer Center in 2021 were included. Two pretrained expert pathologists independently rescored the HER2 slides, followed by analysis using the AI tool. All inconsistent cases were also reviewed by a third senior pathologist. Interobserver agreement between the pathologists and concordance between the pathologists and the AI interpretation results were assessed. We also explored potential ranges where HER2 interpretation by pathologists exhibited inconsistency or inaccuracy. Results: The HER2 AI algorithm showed high performance in interpreting all levels of HER2 expression, with an overall kappa value of 0.85. For HER2 IHC 0 and 1+ cases, the AI model demonstrated high sensitivities (0.839 and 0.914) and specificities (0.983 and 0.890) (Table 1). After the rescoring process, 51 cases were reclassified as IHC 0 and 90 cases as IHC 1+. The overall agreement rate between the historical and re-scoring results was 73.05% (103/141). The agreement rate between the AI and re-scoring results was 91.49% (129/141) (Table 2), indicating comparable interpretation capabilities between the AI model and well-trained pathologists. The interobserver agreement between the two pathologists was 83.69% (118/141), with agreement rates of 88.23% (45/51) for IHC 0 and 81.11% (73/90) for IHC 1+. Analyzing the inconsistent cases (n=12) between IHC 0 and IHC 1+, we examined the percentage of weak and incomplete expression provided by the AI. In 9 out of 12 cases, the AI provided the same results as the third senior pathologist, suggesting that AI could assist in interpretation within this range. We identified a range of 2.83% to 24.98% as a "grey-zone," where even well-trained pathologists provided inconsistent results. Conclusion: Our study demonstrates the excellent performance of an AI-based tool for scoring HER2 IHC 0 and 1+. The AI model exhibits comparable capabilities to well-trained pathologists in interpreting HER2 expression. Additionally, we identified a "grey-zone" among pathologists, highlighting the limitations of manual subjective interpretation. AI assistance within this zone may help mitigate subjective variations. Table 1. Correlation between pathologist read and AI read of HER2 immunohistochemistry. Table 2. Correlation between pathologist rescoring and AI read of HER2 immunohistochemistry. Citation Format: Ming Li, Hong Lv, Yizhi Zhao, Chenglu Zhu, Hansheng Li, Mingzhen Lin, Wen-Tao Yang. The Advantage of Artificial-Intelligence in HER2 IHC 0 and 1+ Scoring in Breast Cancer [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO4-26-09.

Development of an Interpretable Deep Learning Model for Pathological Tumor Response Assessment After Neoadjuvant Therapy.

Neoadjuvant therapy followed by surgery has become the standard of care for locally advanced esophageal squamous cell carcinoma (ESCC) and accurate pathological response assessment is critical to assess the therapeutic efficacy. However, it can be laborious and inconsistency between different observers may occur. Hence, we aim to develop an interpretable deep-learning model for efficient pathological response assessment following neoadjuvant therapy in ESCC. This retrospective study analyzed 337 ESCC resection specimens from 2020-2021 at the Pudong-Branch (Cohort 1) and 114 from 2021-2022 at the Puxi-Branch (External Cohort 2) of Fudan University Shanghai Cancer Center. Whole slide images (WSIs) from these two cohorts were generated using different scanning machines to test the ability of the model in handling color variations. Four pathologists independently assessed the pathological response. The senior pathologists annotated tumor beds and residual tumor percentages on WSIs to determine consensus labels. Furthermore, 1850 image patches were randomly extracted from Cohort 1 WSIs and binarily classified for tumor viability. A deep-learning model employing knowledge distillation was developed to automatically classify positive patches for each WSI and estimate the viable residual tumor percentages. Spatial heatmaps were output for model explanations and visualizations. The approach achieved high concordance with pathologist consensus, with an R^2 of 0.8437, a RAcc_0.1 of 0.7586, a RAcc_0.3 of 0.9885, which were comparable to two senior pathologists (R^2 of 0.9202/0.9619, RAcc_0.1 of 8506/0.9425, RAcc_0.3 of 1.000/1.000) and surpassing two junior pathologists (R^2 of 0.5592/0.5474, RAcc_0.1 of 0.5287/0.5287, RAcc_0.3 of 0.9080/0.9310). Visualizations enabled the localization of residual viable tumor to augment microscopic assessment. This work illustrates deep learning's potential for assisting pathological response assessment. Spatial heatmaps and patch examples provide intuitive explanations of model predictions, engendering clinical trust and adoption(Code and data will be available at https://github.com/WinnieLaugh/ESCC_Percentage once the paper has been conditionally accepted). Integrating interpretable computational pathology could help enhancethe efficiency and consistency of tumor response assessment and empower precise oncology treatment decisions.

An artificial intelligence model for detecting pathological lymph node metastasis in prostate cancer using whole slide images: a retrospective, multicentre, diagnostic study

The pathological examination of lymph node metastasis (LNM) is crucial for treating prostate cancer (PCa). However, the limitations with naked-eye detection and pathologist workload contribute to a high missed-diagnosis rate for nodal micrometastasis. We aimed to develop an artificial intelligence (AI)-based, time-efficient, and high-precision PCa LNM detector (ProCaLNMD) and evaluate its clinical application value. In this multicentre, retrospective, diagnostic study, consecutive patients with PCa who underwent radical prostatectomy and pelvic lymph node dissection at five centres between Sep 2, 2013 and Apr 28, 2023 were included, and histopathological slides of resected lymph nodes were collected and digitised as whole-slide images for model development and validation. ProCaLNMD was trained at a dataset from a single centre (the Sun Yat-sen Memorial Hospital of Sun Yat-sen University [SYSMH]), and externally validated in the other four centres. A bladder cancer dataset from SYSMH was used to further validate ProCaLNMD, and an additional validation (human-AI comparison and collaboration study) containing consecutive patients with PCa from SYSMH was implemented to evaluate the application value of integrating ProCaLNMD into the clinical workflow. The primary endpoint was the area under the receiver operating characteristic curve (AUROC) of ProCaLNMD. In addition, the performance measures for pathologists with ProCaLNMD assistance was also assessed. In total, 8225 slides from 1297 patients with PCa were collected and digitised. Overall, 8158 slides (18,761 lymph nodes) from 1297 patients with PCa (median age 68 years [interquartile range 64-73]; 331 [26%] with LNM) were used to train and validate ProCaLNMD. The AUROC of ProCaLNMD ranged from 0.975 (95% confidence interval 0.953-0.998) to 0.992 (0.982-1.000) in the training and validation datasets, with sensitivities > 0.955 and specificities > 0.921. ProCaLNMD also demonstrated an AUROC of 0.979 in the cross-cancer dataset. ProCaLNMD use triggered true reclassification in 43 (4.3%) slides in which micrometastatic tumour regions were initially missed by pathologists, thereby correcting 28 (8.5%) missed-diagnosed cases of previous routine pathological reports. In the human-AI comparison and collaboration study, the sensitivity of ProCaLNMD (0.983 [0.908-1.000]) surpassed that of two junior pathologists (0.862 [0.746-0.939], P=0.023; 0.879 [0.767-0.950], P=0.041) by 10-12% and showed no difference to that of two senior pathologists (both 0.983 [0.908-1.000], both P>0.99). Furthermore, ProCaLNMD significantly boosted the diagnostic sensitivity of two junior pathologists (both P=0.041) to the level of senior pathologists (both P>0.99), and substantially reduced the four pathologists' slide reviewing time (-31%, P<0.0001;-34%, P<0.0001;-29%, P<0.0001; and-27%, P=0.00031). ProCaLNMD demonstrated high diagnostic capabilities for identifying LNM in prostate cancer, reducing the likelihood of missed diagnoses by pathologists and decreasing the slide reviewing time, highlighting its potential for clinical application. National Natural Science Foundation of China, the Science and Technology Planning Project of Guangdong Province, the National Key Research and Development Programme of China, the Guangdong Provincial Clinical Research Centre for Urological Diseases, and the Science and Technology Projects in Guangzhou.

Factors influencing speech pathology practice in dysphagia after stroke: A qualitative focus group study.

Dysphagia affects over half of adults after stroke. Dysphagia rehabilitation aims to improve swallowing and reduce negative outcomes for these adults. However, significant variability exists in dysphagia rehabilitation. Research is needed to explore the underlying clinician-specific and workplace factors that contribute to variability in dysphagia rehabilitation after stroke. To explore factors influencing speech pathology practice in dysphagia rehabilitation after stroke. We used a phenomenological approach with an interpretivist perspective. Twenty speech pathologists working in dysphagia rehabilitation participated from different workplace settings around Australia. Five semi-structured focus groups were conducted online. Data were inductively analysed using thematic analysis with a coding reliability method. Four themes were discussed within focus groups: (1) relationship between experienced and less experienced clinicians: 'Following what other people have done', (2) need for collaborative learning: 'A safe space to share and train', (3) variation between settings impacts on continuity of care: 'There's a difference between community and acute', and (4) working effectively with multidisciplinary teams (MDT): 'An MDT which can listen to the voice of speech pathology'. Relationships between senior and junior speech pathologists, within speech pathology and MDT, and across inpatient and community settings influenced speech pathology practice. Flattened hierarchies in speech pathology, collaborative learning in workplaces, mutual respect within teams and connection across inpatient and community settings could improve the quality and consistency of dysphagia rehabilitation after stroke. What is already known on this subject Dysphagia rehabilitation can improve swallowing after a stroke. However, dysphagia rehabilitation is characterised by variability in clinical practice. Clinician-specific and workplace factors influence clinical practice and may contribute to variability in dysphagia rehabilitation. What this study adds Professional relationships influence speech pathologists' clinical practice, including relationships between senior and junior clinicians, between inpatient and community settings and with peers and multidisciplinary teams. Workplace norms and hierarchies, poor continuity of care between settings and competing priorities from other disciplines can hinder dysphagia rehabilitation. However, collaborative learning, positive workplace cultures and respectful transdisciplinary care can improve the quality and consistency of clinical practice. What are the clinical implications of this work? Flattening hierarchies in the workplace can foster a safe learning space. Further, questioning workplace norms and seeking out peer learning within and across settings can build clinical skills and confidence. Developing positive workplace cultures that support continuous development may be key for empowering speech pathologists to provide high-quality and consistent dysphagia rehabilitation.

Analysis of false reasons based on the artificial intelligence RRCART model to identify frozen sections of lymph nodes in breast cancer.

Breast cancer is the most common malignant tumor in the world. Intraoperative frozen section of sentinel lymph nodes is an important basis for determining whether axillary lymph node dissection is required for breast cancer surgery. We propose an RRCART model based on a deep-learning network to identify metastases in 2362 frozen sections and count the wrongly identified sections and the associated reasons. The purpose is to summarize the factors that affect the accuracy of the artificial intelligence model and propose corresponding solutions. We took the pathological diagnosis of senior pathologists as the gold standard and identified errors. The pathologists and artificial intelligence engineers jointly read the images and heatmaps to determine the locations of the identified errors on sections, and the pathologists found the reasons (false reasons) for the errors. Through NVivo 12 Plus, qualitative analysis of word frequency analysis and nodal analysis was performed on the error reasons, and the top-down error reason framework of "artificial intelligence RRCART model to identify frozen sections of breast cancer lymph nodes" was constructed based on the importance of false reasons. There were 101 incorrectly identified sections in 2362 slides, including 42 false negatives and 59 false positives. Through NVivo 12 Plus software, the error causes were node-coded, and finally, 2 parent nodes (high-frequency error, low-frequency error) and 5 child nodes (section quality, normal lymph node structure, secondary reaction of lymph nodes, micrometastasis, and special growth pattern of tumor) were obtained; among them, the error of highest frequency was that caused by normal lymph node structure, with a total of 45 cases (44.55%), followed by micrometastasis, which occurred in 30 cases (29.70%). The causes of identification errors in examination of sentinel lymph node frozen sections by artificial intelligence are, in descending order of influence, normal lymph node structure, micrometastases, section quality, special tumor growth patterns and secondary lymph node reactions. In this study, by constructing an artificial intelligence model to identify the error causes of frozen sections of lymph nodes in breast cancer and by analyzing the model in detail, we found that poor quality of slices was the preproblem of many identification errors, which can lead to other errors, such as unclear recognition of lymph node structure by computer. Therefore, we believe that the process of artificial intelligence pathological diagnosis should be optimized, and the quality control of the pathological sections included in the artificial intelligence reading should be carried out first to exclude the influence of poor section quality on the computer model. For cases of micrometastasis, we suggest that by differentiating slices into high- and low-confidence groups, low-confidence micrometastatic slices can be separated for manual identification. The normal lymph node structure can be improved by adding samples and training the model in a targeted manner.

Somatic PIK3CA Variants Are Associated With Eccrine Angiomatous Hamartomas

Objectives: Eccrine angiomatous hamartoma (EAH) is a rare vascular anomaly with mixed eccrine and vascular components, typically identified in children. While benign, EAH can cause significant morbidity and be difficult to treat. The aims of this case series were to identify all patients with EAH that have been seen at the Queensland Children’s Hospital and describe their phenotypic and somatic genotypic details, in an effort to contribute to the limiting understanding and literature surrounding this condition. Methods: Individuals with EAH were retrospectively identified through engagement in a multidisciplinary vascular anomaly clinic in a tertiary Australian children’s hospital. All individuals had a previous histological diagnosis of EAH. High-read-depth sequencing of a panel of 27 genes known to be associated with vascular anomalies was undertaken on affected tissue. Samples were rereviewed by a senior pathologist and geneticist for this study. Results: Five cases of EAH were identified. All were associated with 1 of 3 somatic PIK3CA variants (c.1633G&gt;A;p.Glu545Lys, c.1624G&gt;A;p.Glu542Lys, and c.3140A&gt;G;p.Histo1047Arg) in low allele fractions. These variants have previously been reported in a range of tumors and vascular anomalies, including PIK3CA-related overgrowth spectrum, but not in EAH. Conclusion: Occurrence of somatic PIK3CA variants in EAH provides evidence for a novel gene-disease association and is plausibly the cause of EAH in some individuals. This finding expands the phenotypic spectrum of PIK3CA, contributes to understanding of the pathophysiology of this rare condition, and may avail molecularly targeted therapy in the future.

Performance of clinical metagenomics in France: a prospective observational study

Metagenomic next-generation sequencing (mNGS) allows untargeted identification of a broad range of pathogens, including rare or novel microorganisms. Despite the recognition of mNGS as a valuable diagnostic tool for infections, the most relevant indications for this innovative strategy remain poorly defined. We aimed to assess the determinants of positivity and clinical utility of mNGS. In this observational study, we prospectively performed short-read shotgun metagenomics analysis as a second-line test (in cases of negative first-line test or when the symptoms were not fully explained by initial positive results) or as a first-line test in life-threatening situations requiring urgent non-targeted pathogen identification at the Necker-Enfants Malades Hospital (Paris, France). All sample types, clinical indications, and patient populations were included. Samples were accompanied by a mandatory form completed by the senior clinician or pathologist, on which the clinical level of suspected infection (defined as high or low) was indicated. We assessed the variables (gender, age, immune status, initial suspicion of infection, indication, and sample type) associated with mNGS pathogen detection using odds ratios (ORs) from multivariate logistic regression. Additional investigations were carried out using specific PCR or culture techniques, to confirm positive mNGS results, or when infectious suspicion was particularly high despite a negative mNGS result. Between Oct 29, 2019, and Nov 7, 2022, we analysed 742 samples collected from 523 patients. The initial suspicion of infection was either high (n=470, 63%) or low (n=272, 37%). Causative or possibly causative pathogens were detected in 117 (25%) samples from patients with high initial suspicion of infection, versus nine (3%) samples analysed to rule out infection (OR 9·1, 95% CI 4·6-20·4; p<0·0001). We showed that mNGS had higher odds of detecting a causative or possibly causative pathogenic virus on CNS biopsies than CSF samples (4·1, 1·7-10·7; p=0·0025) and in samples from immunodeficient compared with immunocompetent individuals (2·4, 1·4-4·1; p=0·0013). Concordance with conventional confirmatory tests results was 103 (97%) of 106, when mNGS detected causative or possibly causative pathogens. Altogether, among 231 samples investigated by both mNGS and subsequent specific tests, discordant results were found in 69 (30%) samples, of which 58 (84%) were mNGS positive and specific tests negative, and 11 (16%) mNGS negative and specific tests positive. Major determinants of pathogen detection by mNGS are immune status and initial level of suspicion of infection. These findings will contribute, along with future studies, to refining the positioning of mNGS in diagnostic and treatment decision-making algorithms. Necker-Enfants Malades Hospital and Institut Pasteur. For the French translation of the abstract see Supplementary Materials section.

A Retrospective Multicenter Italian Analysis of Epidemiological, Clinical and Histopathological Features in a Sample of Patients with Acinic Cell Carcinoma of the Parotid Gland.

The acinic cell carcinoma (AciCC) of the parotid gland is a rare tumor with an indolent behavior; however, a subgroup of this tumor presents an aggressive behavior with a tendency to recur. The aim of this multicenter study was to identify and stratify those patients with AciCC at high risk of tumor recurrence. A retrospective study was carried out involving 77 patients treated with surgery between January 2000 and September 2022, in different Italian referral centers. Data about tumor characteristics and its recurrence were collected. The histological specimens and slides were independently reviewed by a senior pathologist coordinator (L.C.) and the institution's local head and neck pathologist. The patients' age average was 53.6 years, with a female prevalence in the group. The mean follow-up was 67.4 months (1-258, SD 59.39). The five-year overall survival (OS) was 83.2%. The 5-year disease-free survival (DFS) was 60% (95% CI 58.2-61.7). A high incidence of necrosis, extraglandular spread, lymphovascular invasion (LVI), atypical mitosis, and cellular pleomorphism was observed in the high-risk tumors compared to the low-risk ones. AciCC generally had an indolent behavior, optimal OS, DFS with few cervical node metastases, and rare distant relapses. This multicenter retrospective case series provides evidence of the need for clinical-epidemiological-histological stratification for patients at risk of poor outcomes. Our results suggest that the correct definition of high-risk AciCC should include tumor size, the presence of necrosis, extraglandular spread, LVI, atypical mitosis, and cellular pleomorphism.

A quantitative comparison of bone resection margin distances in virtual surgical planning versus histopathology: a prospective study.

Positive bone margins have been shown to be associated with worse locoregional control and survival performance in oral oncology patients. With the application of computer-assisted surgery and patient-specific surgical guides, the authors can accurately execute the preoperative osteotomy plan. However, how well the authors can predict the margin distance in the final histopathology with a preoperative computed tomography (CT) scan, the factors associated with it, and how much leeway CT should spare when designing the osteotomy planes during virtual surgical planning (VSP) remain to be investigated. Patients from January 2021 to December 2022 with benign or malignant jaw tumors and with signs of bone marrow involvement in the preoperative CT scan in our center were prospectively recruited to the study. VSP and measurement of the closest margin distance in the CT scan were performed by the single team of surgeons. The resection specimen was processed, and the margin distances were measured by a dedicated senior pathologist with the knowledge of orientation of the osteotomy planes. A total of 35 patients were recruited, with 21 malignant and 14 benign cases. Sixty-eight bone margins were quantitatively analyzed. No significant difference in margin distances measured from the CT scan and final histopathology was detected ( P =0.19), and there was a strong correlation between the two (r s =0.74, P <0.01). A considerable amount of variance was detected in the level of discrepancy between margin distances measured in the CT scan and final histopathology (overall SD=6.26 mm, malignancy SD=7.44 mm, benign SD=4.40 mm). No significant correlation existed between the two margin distances when only maxilla tumor margins were assessed ( P =0.16). The bone margin distance in VSP is reliably correlated to the final pathological margin distance. A leeway distance of 15mm and 9mm should be considered when designing the osteotomy planes for malignancy and benign cases, respectively. Extra attention should be paid to maxilla cases when predetermining the osteotomy planes during VSP.

Novel role of fluorescent in situ hybridization technique (FISH) in recommended surgical margins of dermatofibrosarcoma protuberans: A preliminary study

Dermatofibrosarcoma protuberans (DFSP) is a rare skin tumor. There is no standard recommendation for its surgical management. The currently used histological analysis are HES (hematoxylin eosin saffron) staining and immunohistochemistry for CD34 expression in particular cases. Fluorescent in situ hybridization (FISH) technique is only used to qualify the DFSP as translocated or non-translocated and is not used as a diagnostic method. The aim of our study was to determine by FISH (as a diagnostic method) whether cancerous cells that could not be identified through HES staining±immunohistochemistry were present at the two-centimeter margins that were found to be tumor-free. Samples from patients who underwent surgery between 2010 and 2018 were collected. Intralesional and peripheral (at 2cm margins) paraffin slides were included. An average of 7.4 slides per specimen was analyzed. Firstly, the preselected slides were reread by a senior pathologist to confirm the absence of microscopic findings of DFSP at 2cm margins. Secondly a FISH analysis was used as a quantitative diagnostic approach, in order to find the t(17;22) translocation. Among the seven specimens that included 2cm margins, two samples presented one or more translocations, which were not visible in standard morphology assessments at two centimeters tumor-free margins. FISH analysis can have a new role in defining tumor-free margins. This would reduce the incidence of disease recurrence after resection and improve the post-operative complementary care.

Deep Learning–Enabled Diagnosis of Liver Adenocarcinoma

Diagnosis of adenocarcinoma in the liver is a frequent scenario in routine pathology and has a critical impact on clinical decision making. However, rendering a correct diagnosis can be challenging, and often requires the integration of clinical, radiologic, and immunohistochemical information. We present a deep learning model (HEPNET) to distinguish intrahepatic cholangiocarcinoma from colorectal liver metastasis, as the most frequent primary and secondary forms of liver adenocarcinoma, with clinical grade accuracy using H&E-stained whole-slide images. HEPNET was trained on 714,589 image tiles from 456 patients who were randomly selected in a stratified manner from a pool of 571 patients who underwent surgical resection or biopsy at Heidelberg University Hospital. Model performance was evaluated on a hold-out internal test set comprising 115 patients and externally validated on 159 patients recruited at Mainz University Hospital. On the hold-out internal test set, HEPNET achieved an area under the receiver operating characteristic curve of 0.994 (95% CI, 0.989-1.000) and an accuracy of 96.522% (95% CI, 94.521%-98.694%) at the patient level. Validation on the external test set yielded an area under the receiver operating characteristic curve of 0.997 (95% CI, 0.995-1.000), corresponding to an accuracy of 98.113% (95% CI, 96.907%-100.000%). HEPNET surpassed the performance of 6 pathology experts with different levels of experience in a reader study of 50 patients (P= .0005), boosted the performance of resident pathologists to the level of senior pathologists, and reduced potential downstream analyses. We provided a ready-to-use tool with clinical grade performance that may facilitate routine pathology by rendering a definitive diagnosis and guiding ancillary testing. The incorporation of HEPNET into pathology laboratories may optimize the diagnostic workflow, complemented by test-related labor and cost savings.

Assessment of endobronchial ultrasound-guided bronchoscopy (EBUS) intranodal forceps biopsy added to EBUS 19-gauge transbronchial needle aspiration: A blinded pathology panel analysis.

Endobronchial ultrasound-guided (EBUS) transbronchial needle aspiration (TBNA) has significantly improved the diagnostic workup for intrathoracic lymphadenopathies. More recently, EBUS intranodal forceps biopsy (IFB) has been developed in an attempt to maximize diagnostic yield by providing additional tissue. In this study, we aimed to assess the improvement of diagnostic yield with EBUS-TBNA combined with EBUS-IFB, compared to EBUS-TBNA alone. Consecutive patients who had 19-G EBUS-TBNA and EBUS-IFB from August 30, 2018, to September 28, 2021, were included. Four senior pathologists retrospectively analyzed, independently and blindly, first, only the EBUS-TBNA samples (cell block), then, at least 1 month later, both samples from EBUS-TBNA and from EBUS-IFB together. Fifty patients were included in the study and 52 lymph nodes were analyzed. Diagnostic yield was 77% (40/52) for EBUS-TBNA alone and 94% (49/52) when combined with EBUS-IFB (p = 0.023). Malignancy was diagnosed with EBUS-TBNA combined with EBUS-IFB in 25/26 cases (96%), versus 22/26 (85%) with EBUS-TBNA alone (p = 0.35); and 4/5 (80%) versus 2/5 (40%) for lymphoma specifically. Kappa interobserver agreement was 0.92 for EBUS-IFB and 0.87 for EBUS-TBNA alone. Nonmalignant condition was diagnosed with EBUS-TBNA combined with EBUS-IFB in 24/26 cases (92%), versus 18/26 (69%) for EBUS-TBNA alone (p = 0.07). The use of EBUS-IFB combined with 19-G EBUS-TBNA improves the mediastinal lymph node diagnostic yield However the benefit appears to be mainly restricted to nonmalignant histology.

Development and Validation of a Deep Learning-Based Histologic Diagnosis System for Diagnosing Colorectal Sessile Serrated Lesions.

The histopathologic diagnosis of colorectal sessile serrated lesions (SSLs) and hyperplastic polyps (HPs) is of low consistency among pathologists. This study aimed to develop and validate a deep learning (DL)-based logical anthropomorphic pathology diagnostic system (LA-SSLD) for the differential diagnosis of colorectal SSL and HP. The diagnosis framework of the LA-SSLD system was constructed according to the current guidelines and consisted of 4 DL models. Deep convolutional neural network (DCNN) 1 was the mucosal layer segmentation model, DCNN 2 was the muscularis mucosa segmentation model, DCNN 3 was the glandular lumen segmentation model, and DCNN 4 was the glandular lumen classification (aberrant or regular) model. A total of 175 HP and 127 SSL sections were collected from Renmin Hospital of Wuhan University during November 2016 to November 2022. The performance of the LA-SSLD system was compared to 11 pathologists with different qualifications through the human-machine contest. The Dice scores of DCNNs 1, 2, and 3 were 93.66%, 58.38%, and 74.04%, respectively. The accuracy of DCNN 4 was 92.72%. In the human-machine contest, the accuracy, sensitivity, and specificity of the LA-SSLD system were 85.71%, 86.36%, and 85.00%, respectively. In comparison with experts (pathologist D: accuracy 83.33%, sensitivity 90.91%, specificity 75.00%; pathologist E: accuracy 85.71%, sensitivity 90.91%, specificity 80.00%), LA-SSLD achieved expert-level accuracy and outperformed all the senior and junior pathologists. This study proposed a logical anthropomorphic diagnostic system for the differential diagnosis of colorectal SSL and HP. The diagnostic performance of the system is comparable to that of experts and has the potential to become a powerful diagnostic tool for SSL in the future. It is worth mentioning that a logical anthropomorphic system can achieve expert-level accuracy with fewer samples, providing potential ideas for the development of other artificial intelligence models.

The Outcome of Treatment for Thyroglossal Tract Anomalies.

Background Thyroglossal tract anomalies are the most common cause of midline neck swelling. Thyroglossal cysts present between the base of the tongue and cricoid cartilage as painless, midline swelling that moves on deglutition and protrusion of the tongue. If the thyroglossal cyst gets infected or is violated surgically, it can result in a thyroglossal sinus or fistula. Investigations in patients with suspected thyroglossal cysts include a thyroid function test, ultrasonography of the neck, and fine needle aspiration cytology (FNAC). Computed tomography (CT) or magnetic resonance imaging (MRI) can be done for larger cysts or suspected malignancies. The gold standard treatment is Sistrunk surgery. Recurrence rates with Sistrunk surgery are lower when the surgery is performed accurately. This study was conducted to document the clinical presentation and treatment outcome in patients treated for thyroglossal tract anomalies. Methods This is a retrospective analysis of 46 medical case records of patients operated on for thyroglossal tract anomalies at a rural tertiary care hospital from April 1995 to April 2021. Patients fulfilling the inclusion and exclusion criteria were evaluated with a detailed history, various clinical presentations, location, extent of anomalies, and thyroid function test results documented. Ultrasound images were reviewed, and the findings were documented. All patients have consent, and they underwent Sistrunk surgery. Patients in whom the normal thyroid gland was found to be absent were started on replacement thyroxine after surgery. The specimen was subjected to histopathological examination by a senior pathologist. The outcome regarding complications, recurrence, and further treatment were reviewed. The outcome of the thyroglossal fistula was compared with that of thyroglossal cysts, and the outcome of infrahyoid thyroglossal tract anomalies was compared with that of suprahyoid thyroglossal tract anomalies. Results In this study, among the 46 patients, 24 (52.2%) were female and 22 (47.8%) were male. The minimum age was three years, the maximum was 58, and the mean was 20.6 years. In this study, 71.7% of the patients were diagnosed with thyroglossal cysts, 10.9% had thyroglossal fistulas, and two had lingual thyroids. The most common location of the cyst was infrahyoid (73.9%). 44 patients underwent Sistrunk surgery, and two patients diagnosed with lingual thyroid underwent excision. Three patients had complications (two pharyngo-cutaneous fistulas, one wound dehiscence), and all were managed conservatively. There were no recurrences in our study. Conclusion Thyroglossal tract anomalies are the most common congenital cervical anomalies. A complete Sistrunk procedure includes the removal of the entire thyroglossal tract, inclusive of the body of the hyoid bone along with the cuff of base tongue tissue, and gives the best result for thyroglossal tract anomalies.