Experienced Pathologist Research Articles

Compatible-domain Transfer Learning for Breast Cancer Classification with Limited Annotated Data

Microscopic analysis of breast cancer images is the primary task in diagnosing cancer malignancy. Recent attempts to automate this task have employed deep learning models whose success has depended on large volumes of data, while acquiring annotated data in biomedical domains is time-consuming and may not always be feasible. A typical strategy to address this is to apply transfer learning using pre-trained models on a large natural image database (e.g., ImageNet) instead of training a model from scratch. This approach, however, has not been effective in several previous studies due to fundamental differences between natural and medical images. In this study, for the first time we proposed the idea of using a compatible data set of histopathological images to classify breast cancer cytological biopsy specimens. Despite intrinsic differences between histopathological and cytological images, we demonstrate that the features learned by deep networks during the pre-training procedure are compatible with those obtained throughout fine-tuning process. To thoroughly investigate this assertion, we explore three different strategies for training as well as two different approaches for fine-tuning deep learning models. By comparing the obtained results with those of previous state-of-the-art research conducted on the same data set, we demonstrate that the proposed method boasts of improved classification accuracy by 6% to 17% compared to the studies which were based on traditional machine learning techniques, and also enhanced accuracy by roughly 7% compared to those who utilized deep learning methods, eventually achieving 98.73% validation accuracy and 94.55% test accuracy. Exploring different training scenarios also revealed that using a compatible dataset has helped to elevate the classification accuracy by 3.0% compared to the typical approach of using ImageNet. Experimental results show that our approach, despite using a very small number of training images, has achieved performance comparable to that of experienced pathologists and has the potential to be applied in clinical settings.

Protocol Biopsies in Pediatric Liver Transplantation Recipients Improve Graft Histology and Personalize Immunosuppression.

Protocol liver biopsies (PLBs) are part of the follow-up program at many pediatric liver transplant centers, but the impact on clinical decision-making and allograft histology following adjustments of immunosuppression (IS) after PLB has not been thoroughly analyzed. Following our previous single-center cohort study, we have now evaluated histological findings of 178 PLBs of 118 pediatric patients transplanted at our center between 1998 and 2017. In particular, we focused on the changes in allograft histology in the follow-up biopsy of a subgroup of 22 patients, in which the histologic findings led to an adjustment of immunosuppressive therapy. All biopsies of this sub-study group were reevaluated by an experienced pathologist. The overall frequency and severity of fibrosis increased over time after orthotopic liver transplantation. Patients with donor-specific antibodies (DSAs) had a higher prevalence of fibrosis than DSA-negative patients. Graft inflammation decreased significantly after intensifying IS, but renal function needs to be monitored. A significant increase in fibrosis was detected in children with reduced IS. The adjustment of IS following PLBs has a significant impact on allograft histology. Since chronic inflammatory changes may lead to graft failure, adjustment of IS seems to be of major importance for the long-term outcome.

Successful Utilization of Kidney Allografts with Diffuse Glomerular Fibrin Thrombi on the Preimplantation Biopsy after Circulatory Death: A Case Series

Background: Kidney allografts with the presence of diffuse glomerular fibrin thrombi are typically rejected by most centers due to concern for poor allograft outcomes in the recipients. The aim of this study was to report our single center experience in the use of such deceased donor allografts. Methods: Retrospective single-center cohort study of kidney transplant recipients who received deceased donor allografts with moderate-to-severe diffuse glomerular fibrin microthrombi on the pre-implantation biopsy. Results: Three adult recipients received deceased donor kidney transplantation from donation after circulatory death donors. One patient was pre-emptive to dialysis at the time of transplant. The donors had moderate-to-severe diffuse glomerular fibrin thrombi on preimplantation biopsies with no evidence of cortical necrosis. Mean follow-up period was 196 days. None of the recipients developed delayed allograft function. The mean 3-month and 6-month creatinine were 1.6 and 1.5 mg/dL, respectively, with corresponding mean eGFRs (estimated glomerular filtration rates) of 45.7 and 47.3 mL/min/1.73m2. Conclusions: After excluding significant cortical necrosis by experienced transplant renal pathologist, otherwise transplantable kidney allografts with diffuse fibrin thrombi may be successfully transplanted in renal transplant recipients with good renal outcomes.

Worst pattern of invasion and other histopathological features in oral cancer as determinants of prognosis and survival rate: A retrospective cohort analysis.

Oral cavity squamous cell carcinoma (OCSCC) is a well-recognized malignancy of the head and neck. Studies on patients with early-stage oral cancer have shown that they develop locally recurring and/or regional lymph node metastasis, which results in disease-associated mortality. Thus, early-stage oral cancer does not always present good prognoses. The present study aimed to determine the efficacy of using worst pattern of invasion (WPOI) and other histopathological features, such as prognostic factors in OCSCC, and analyze the impact of resection margin status and histopathological prognostic indicators on local recurrence (LR) and overall survival (OS) in patients with OCSCC. A retrospective cohort study was conducted by reviewing the charts of 63 patients with OCSCC treated with primary surgery at King Abdulaziz University Hospital between 2012 and 2019. An author and an experienced pathologist reviewed pathology slides. Associations of histopathological factors, including differentiation, stage, lymphovascular invasion, extracapsular extension, perineural invasion (PNI), WPOI and surgical margins, with LR or disease-free survival (DFS) were evaluated. Univariate analysis identified WPOI and PNI, and multivariate analysis identified the WPOI as predictive factors for LR and DFS. Kaplan-Meier analysis identified the WPOI and PNI as predictive factors for OS and WPOI as a predictive factor for DFS. Therefore, it may be concluded that WPOI and PNI are significant independent prognostic factors for local tumor control and DFS in patients with OCSCC.

Models for Predicting Velopharyngeal Competence Based on Speech and Resonance Errors and Velopharyngeal Area Estimation.

To develop tools for predicting velopharyngeal competence (VPC) based on auditory-perceptual assessment and its correlation with objective measures of velopharyngeal orifice area. Methodological study. Sixty-two patients with repaired cleft palate, aged 6 to 45 years, underwent aerodynamic evaluation by means of the pressure-flow technique and audiovisual recording of speech samples. Three experienced speech-language pathologists analysed the speech samples by rating the following resonance, visual, and speech variables: hypernasality, audible nasal air emission, nasal turbulence, weak pressure consonants, facial grimacing, active nonoral errors, and overall velopharyngeal competence. The correlation between the perceptual speech variables and velopharyngeal orifice area estimates was analysed with Spearman's correlation coefficient. Two statistical models (discriminant and exploratory) were used to predict VPC based on the orifice area estimates. Sensitivity and specificity analyses were performed to verify the clinical applicability of the models. There was a strong correlation between VPC (based on the orifice area estimates) and each speech variable. Both models showed 88.7% accuracy in predicting VPC. The sensitivity and specificity for the discriminant model were 92.3% and 97.2%, respectively, and 96.2% and 94.4% for the exploratory model. Two predictor models based on ratings of resonance, visual, and speech variables and a simple calculation of a composite variable, SOMA (Eng. "sum"), were developed and found to be efficient in predicting VPC defined by orifice estimates categories based on aerodynamic measurements. Both tools may contribute to the diagnosis of velopharyngeal dysfunction in clinical practice.

Democratizing Pathological Image Segmentation with Lay Annotators via Molecular-empowered Learning.

Multi-class cell segmentation in high-resolution Giga-pixel whole slide images (WSI) is critical for various clinical applications. Training such an AI model typically requires labor-intensive pixel-wise manual annotation from experienced domain experts (e.g., pathologists). Moreover, such annotation is error-prone when differentiating fine-grained cell types (e.g., podocyte and mesangial cells) via the naked human eye. In this study, we assess the feasibility of democratizing pathological AI deployment by only using lay annotators (annotators without medical domain knowledge). The contribution of this paper is threefold: (1) We proposed a molecular-empowered learning scheme for multi-class cell segmentation using partial labels from lay annotators; (2) The proposed method integrated Giga-pixel level molecular-morphology cross-modality registration, molecular-informed annotation, and molecular-oriented segmentation model, so as to achieve significantly superior performance via 3 lay annotators as compared with 2 experienced pathologists; (3) A deep corrective learning (learning with imperfect label) method is proposed to further improve the segmentation performance using partially annotated noisy data. From the experimental results, our learning method achieved F1 = 0.8496 using molecular-informed annotations from lay annotators, which is better than conventional morphology-based annotations (F1 = 0.7015) from experienced pathologists. Our method democratizes the development of a pathological segmentation deep model to the lay annotator level, which consequently scales up the learning process similar to a non-medical computer vision task. The official implementation and cell annotations are publicly available at https://github.com/hrlblab/MolecularEL.

The Diagnostic Approach towards Combined Hepatocellular-Cholangiocarcinoma-State of the Art and Future Perspectives.

Combined hepatocellular-cholangiocarcinoma (cHCC-CCA) is a rare primary liver cancer which displays clinicopathologic features of both hepatocellular (HCC) and cholangiocellular carcinoma (CCA). The similarity to HCC and CCA makes the diagnostic workup particularly challenging. Alpha-fetoprotein (AFP) and carbohydrate antigen 19-9 (CA 19-9) are blood tumour markers related with HCC and CCA, respectively. They can be used as diagnostic markers in cHCC-CCA as well, albeit with low sensitivity. The imaging features of cHCC-CCA overlap with those of HCC and CCA, dependent on the predominant histopathological component. Using the Liver Imaging and Reporting Data System (LI-RADS), as many as half of cHCC-CCAs may be falsely categorised as HCC. This is especially relevant since the diagnosis of HCC may be made without histopathological confirmation in certain cases. Thus, in instances of diagnostic uncertainty (e.g., simultaneous radiological HCC and CCA features, elevation of CA 19-9 and AFP, HCC imaging features and elevated CA 19-9, and vice versa) multiple image-guided core needle biopsies should be performed and analysed by an experienced pathologist. Recent advances in the molecular characterisation of cHCC-CCA, innovative diagnostic approaches (e.g., liquid biopsies) and methods to analyse multiple data points (e.g., clinical, radiological, laboratory, molecular, histopathological features) in an all-encompassing way (e.g., by using artificial intelligence) might help to address some of the existing diagnostic challenges.

Primary Mucinous Tumors of the Ovary: An Interobserver Reproducibility and Detailed Molecular Study Reveals Significant Overlap Between Diagnostic Categories

Primary ovarian mucinous tumors represent a heterogeneous group of neoplasms, and their diagnosis may be challenging. We analyzed 124 primary ovarian mucinous tumors originally diagnosed as mucinous borderline tumors (MBTs) or mucinous carcinomas (MCs), with an emphasis on interobserver diagnostic agreement and the potential for diagnostic support by molecular profiling using a next-generation sequencing targeted panel of 727 DNA and 147 RNA genes. Fourteen experienced pathologists independently assigned a diagnosis from preset options, based on a review of a single digitized slide from each tumor. After excluding 1 outlier participant, there was a moderate agreement in diagnosing the 124 cases when divided into 3 categories (κ = 0.524, for mucinous cystadenoma vs MBT vs MC). A perfect agreement for the distinction between mucinous cystadenoma/MBT as a combined category and MC was found in only 36.3% of the cases. Differentiating between MBTs and MCs with expansile invasion was particularly problematic. After a reclassification of the tumors into near-consensus diagnostic categories on the basis of the initial participant results, a comparison of molecular findings between the MBT and MC groups did not show major and unequivocal differences between MBTs and MCs or between MCs with expansile vs infiltrative pattern of invasion. In contrast, HER2 overexpression or amplification was found only in 5.3% of MBTs and in 35.3% of all MCs and in 45% of MCs with expansile invasion. Overall, HER2 alterations, including mutations, were found in 42.2% of MCs. KRAS mutations were found in 65.5% and PIK3CA mutations in 6% of MCs. In summary, although the diagnostic criteria are well-described, diagnostic agreement among our large group of experienced gynecologic pathologists was only moderate. Diagnostic categories showed a molecular overlap. Nonetheless, molecular profiling may prove to be therapeutically beneficial in advanced-stage, recurrent, or metastatic MCs.

Pathologists' perspective on the study design, analysis, and interpretation of proliferative lesions in lifetime and prenatal rodent carcinogenicity bioassays of aspartame.

Aspartame, an artificial sweetener commonly used as a sugar substitute, is currently authorized for use in more than 100 countries. Hundreds of studies, conducted in various countries dating back to the 1970s, have shown that aspartame is safe at real-world exposure levels. Furthermore, multiple human epidemiology studies have provided no indication that consumption of aspartame induces cancer. Given the continued controversy surrounding the Ramazzini Institute's (RI) studies suggesting that aspartame is a carcinogenic hazard in rodents and evaluation by the International Agency for Research on Cancer, this report aims to provide the perspective of experienced pathologists on publicly available pathology data regarding purported proliferative lesions in liver, lung, lymphoid organs, and mammary gland as well as their implications for human risk assessment as reported for three lifetime rodent carcinogenicity bioassays of aspartame conducted at the RI. In the authors' view, flaws in the design, methodology and reporting of the RI aspartame studies limit the utility of the data sets as evidence that this agent represents a carcinogenic hazard. Therefore, all three RI studies, and particularly the accuracy of their pathology diagnoses and interpretations, should be rigorously reviewed by qualified and experienced veterinary toxicologic pathologists in assessing aspartame's carcinogenic risk.

The correct methodological approach to the diagnosis of celiac disease: the point of view of the pathologist.

The diagnosis of celiac disease relies on the assessment of serological data and the presence of histological alterations in the duodenal mucosa. The duodenal biopsy is pivotal in adults, and in some circumstances in children, to confirm the clinical suspicion of celiac disease. The correct interpretation of duodenal biopsies is influenced by numerous variables. The aim of this overview is to describe the correct methodological approach including the procedures of biopsy sampling, orientation, processing, staining and histopathological classification in order to avoid or minimize the errors and the variability in duodenal biopsy interpretation. Multiple biopsies taken from different sites of the duodenum during endoscopy maximize the diagnostic yield of duodenal histological sampling. Proper orientation of the biopsy samples is of the utmost importance to assess histological features of pathological duodenal mucosa and to avoid artifacts that may lead even an experienced pathologist to a wrong histological interpretation with subsequent misdiagnosis of celiac disease. An immunohistochemical stain for CD3 can be invaluable to aid the pathologist in obtaining a more accurate intra-epithelial T lymphocytes count. A simplified histological classification facilitates the clinician's work and improves the communication between pathologist and clinician. An integrated clinical and pathological approach is required for a correct diagnosis of celiac disease since a relatively large number of conditions may cause duodenal damage with a histological appearance similar to that of celiac disease.

Exploring Hypoxia in Prostate Cancer With T2-weighted Magnetic Resonance Imaging Radiomics and Pimonidazole Scoring.

Radiomics involves high throughput extraction of mineable precise quantitative imaging features that serve as non-invasive prognostic or predictive biomarkers. High levels of hypoxia are associated with a poorer prognosis in prostate cancer and limit radiation therapy efficacy. Most patients with prostate cancer undergo magnetic resonance imaging (MRI) as a part of their diagnostics, and T2 imaging is the most utilised imaging method. The aim of this study was to determine whether hypoxia in prostate tumors could be identified using a radiomics model extracted from T2-weighted MR images. Eighty eight intermediate or high-risk prostate cancer patients were evaluated. Prior to radical prostatectomy, all patients received pimonidazole (PIMO). PIMO hypoxic scores were assigned in whole-mount sections from prostatectomy specimens by an experienced pathologist who was blinded to MRI. The region of interest used for radiomics analysis included the prostatic index tumor. Radiomics extraction yielded 165 features using a special evaluation version of RadiomiX [RadiomiX Research Toolbox version 20180831 (OncoRadiomics SA, Liège, Belgium)] for non-clinical use. Multivariable logistic regression with Elastic Net regularization was utilised using 10 times repeated 10-fold cross-validation to select the best model hyperparameters, optimizing for area under the receiver operating characteristic curve (AUC). The average (out of sample) performance based on the repeated cross validation using the ONESE model yielded an AUC of 0.60±0.2. Shape-based features were the most prominent in the model. The development of a radiomics hypoxia model using T2 weighted MR images, standard in the staging of prostate cancer, is possible.

Retrospective histological evaluation of orchiectomy specimens following testicular torsion reveals a 10% incidence of reversible injury. Is it time for a change of strategy?

According to standard medical practice, immediate orchiectomy is advised in the case of a non-viable testis following testicular torsion. Because of the lack of objective criteria in the evaluation of testicular viability, the fate of the testis solely depends on the discretion and experience of the surgeon. In this study, we retrospectively reassess the management of patients with testicular torsion in our center, and we ask the question, from a retrospective point of view, of whether the decision to perform orchiectomy has always been correct. We retrospectively reviewed all cases of patients with testicular torsion who underwent surgery in our center between 2001 and 2021. All orchiectomy specimens were reevaluated and graded by an experienced pathologist using the Mikuz grading system. Immediate orchiectomy was performed in 48 of the 136 patients (35%). Five (10.4%) of the 48 orchiectomy specimens were categorized as "grade 1," and 17 (35.4%) were categorized as "grade 2." The time between the onset of symptoms and surgical exploration exceeded 12h in three of the five patients with a "grade 1" testicular injury, and in one case, it even exceeded 24h. "Grade 1" testicular injury is potentially reversible, whereas the fate of a testis with a "grade 2" testicular injury remains unknown. Whether and to what extent partial testicular tissue salvage in a "grade 2" injury is still possible remain unclear. Our results indicate that at least 10% of the testicles removed in our center could, from a retrospective point of view, have been salvaged. Our study further demonstrated that the duration of symptoms is not an absolute indicator of testicular damage and that the decision of whether orchiectomy should be performed, based simply on the subjective macroscopic image of the affected testis, is not always correct.

Utility of CK8, CK10, CK13, and CK17 in Differential Diagnostics of Benign Lesions, Laryngeal Dysplasia, and Laryngeal Squamous Cell Carcinoma.

There are no reliable immunohistochemical markers for diagnosing laryngeal squamous cell carcinoma (SCC) or diagnosing and grading laryngeal dysplasia. We aimed to evaluate the diagnostic utility of CK8, CK10, CK13, and CK17 in benign laryngeal lesions, laryngeal dysplasia, and laryngeal SCC. This retrospective study included 151 patients diagnosed with laryngeal papilloma, laryngeal polyps, laryngeal dysplasia, and laryngeal SCC who underwent surgical treatment between 2010 and 2020. Immunohistochemistry (IHC) was carried out using specific monoclonal antibodies against CK8, CK10, CK13, and CK17. Two experienced pathologists performed semi-quantitative scoring of IHC positivity. The diagnostic significance of the markers was analyzed. CK13 showed a sensitivity of 100% and a specificity of 82.5% for distinguishing between laryngeal SCC and laryngeal dysplasia and benign lesions. CK17 showed a sensitivity of 78.3% and specificity of 57.1% for the detection of laryngeal SCC vs. laryngeal dysplasia. CK10 showed a sensitivity of 80.0% for discriminating between low-grade and high-grade dysplasia, and a specificity of 61.1%. Loss of CK13 expression is a reliable diagnostic tool for diagnosing laryngeal lesions with malignant potential and determining resection lines. In lesions with diminished CK13 expression, CK17 could be used as an auxiliary immunohistochemical marker in diagnosing laryngeal SCC. In CK13-negative and CK17-positive lesions, CK10 positivity could be used to determine low-grade dysplasia. CK8 is not a useful IHC marker in differentiating between benign laryngeal lesions, laryngeal dysplasia, and laryngeal SCC.

Adaptive fuzzy‐region growing fusion and improved CNN‐ANFIS‐based automated segmentation and classification of cervical cancer

SummaryThe cervical cancer screening mechanisms help to classify cervical lesions into cancerous, precancerous, and normal. But, some of the classification approaches provide optimal outcomes based on the precise spotting and segmenting of cervical cancer. Moreover, still, challenges persist in the precise spotting of cervical cancer. The conventional screening methods for cervical cancer require more experienced pathologists, but it may also lead to poor efficiency and low accuracy. To overcome these issues, medical image processing introduces both machine learning and deep learning methods into cervical cell classification. The major scope of this task is to plan a new cervical cancer diagnosis method using intelligent approaches. Here, the segmentation of abnormality is done by the novel adaptive fuzzy‐region growing fusion model, which is improved by the hybridized optimization algorithm named gray wolf‐tunicate swarm algorithm (GW‐TSA). Then, the new ensemble classification termed Improved convolutional neural network with artificial neural network fuzzy inference system, which depends on the proposed GW‐TSA for performance enhancement through optimizing the hyperparameters. Finally, the simulation findings show that the designed approach is achievable and provides the highest accuracy rate in diagnosing cervical cancer.

MRI characteristics for predicting histological subtypes in patients with uterine cervical adenocarcinoma

PurposeTo evaluate the magnetic resonance imaging (MRI) findings of uterine cervical adenocarcinoma for predicting different histological subtypes. Materials and methodsWe retrospectively analyzed MRI findings of 76 consecutive patients with histopathologically-confirmed uterine cervical adenocarcinoma undergoing preoperative MRI examination. An experienced pathologist classified the histological subtypes based on World Health Organization’s 2020 classification and into human papillomavirus (HPV)-associated adenocarcinomas (HPVAs, n = 54) (usual type and variants) and HPV-independent adenocarcinomas (HPVIs, n = 22) (gastric type adenocarcinoma (GAS), clear cell type, and other types). Different MRI variables were compared quantitatively and qualitatively between HPVA and HPVI and between GAS and non-GAS tumor types. ResultsThe maximum tumor diameter was significantly greater in HPVIs than HPVAs (41.9 ± 18.6 vs 32.7 ± 15.6 mm; p < 0.05). Heterogeneous enhancement on fat-suppressed gadolinium-enhanced T1-weighted images was more frequently seen in HPVIs than HPVAs (62 % vs 15 %; p < 0.01) and in GASs than non-GASs (78 % vs 16 %; p < 0.01). Also, infiltrative growth pattern (58 % vs 20 %; p < 0.05) and intratumoral cyst formation (83 % vs 47 %) (p < 0.05) were more frequent in GASs than non-GASs. ConclusionsCompared with HPVAs, HPVIs tend to have a larger tumor size with heterogeneous enhancement, of which GASs frequently show infiltrative growth patterns with intratumoral cyst formation and heterogeneous enhancement.

An international unified approach to reporting and grading invasive breast cancer. An overview of the International Collaboration on Cancer Reporting (ICCR) initiative.

Standardised reporting of breast cancer key pathology data has become the norm in some parts of the world, but are based on national or regional guidelines that differ in certain aspects, resulting in divergent reporting practices and a lack of comparability of data internationally. The International Collaboration on Cancer Reporting (ICCR), a global alliance of major (inter-)national pathology and cancer organizations, have recently produced a new international dataset for the pathology reporting of breast cancer, including resection specimens with invasive cancer and ductal carcinoma in situ (DCIS) of the breast. This initiative aims at providing an international unified approach to reporting cancer. The guidance was prepared by an international expert panel consisting of experienced breast pathologists, a surgeon, and an oncologist. The dataset includes core (essential) and noncore (optional) data items based on a critical review and discussion of current evidence. Commentary is provided for each data item to explain the rationale for selection, its clinical relevance, and to highlight potential areas of disagreement or lack of evidence, in which case a consensus position was formulated. The process concludes with international public consultation, before ratification and publication on the free open access ICCR website, with a synoptic reporting guide. The key aim is to promote high-quality, standardised pathology reporting that can be used worldwide. Histological grade, tumour size, and oestrogen receptor status are used in this article to illustrate this process and the detail provided to support its inclusion.

Automated Segmentation and Morphological Characterization of Hepatic Steatosis and Correlation with Histopathology

Prevalence of nonalcoholic fatty liver disease (NAFLD) has increased to 25% of the world population. Hepatic steatosis is a hallmark feature of NAFLD and is assessed histologically usingvisual and ordinal fat grading criteria (0-3) from the Nonalcoholic Steatohepatitis (NASH) Clinical Research Network (CRN) scoring system. The purpose of this study is to automatically segment and extract morphological characteristics and distributions of fat droplets (FDs) on liver histology images and find associations with severity of steatosis. A previously published human cohort of 68 NASH candidates was graded for steatosis by an experienced pathologist using the Fat CRN grading system. The automated segmentation algorithm quantified fat fraction (FF) and fat-affected hepatocyte ratio (FHR), extracted fat morphology by calculating radius and circularity of FDs, and examined FDs distribution and heterogeneity using nearest neighbor distance and regional isotropy. Regression analysis and Spearman correlation (ρ) yielded high correlations for radius (R2=0.86, ρ=0.72), nearest neighbor distance (R2=0.82, ρ=-0.82), regional isotropy (R2=0.84, ρ=0.74), and FHR (R2=0.90, ρ=0.85), and low correlation for circularity (R2=0.48, ρ=-0.32) with FF and pathologist grades, respectively. FHR showed a better distinction between pathologist Fat CRN grades compared to conventional FF measurements, making it a potential surrogate measure for Fat CRN scores. Our results showed variation in distribution of morphological features and steatosis heterogeneity within the same patient's biopsy sample as well as between patients of similar FF. The fat percentage measurements, specific morphological characteristics, and patterns of distribution quantified with the automated segmentation algorithm showed associations with steatosis severity; however, future studies are warranted to evaluate the clinical significance of these steatosis features in progression of NAFLD and NASH.

Prognostic Significance of p16 and Its Relationship with Human Papillomavirus Status in Patients with Penile Squamous Cell Carcinoma: Results of 5 Years Follow-Up.

Penile Squamous Cell Carcinoma (PSCC) is associated with high-risk human papillomavirus (HR-HPV). The immunohistochemical (IHC) test for p16INK4a (p16) is highly correlated with HR-HPV expression in other SCCs. To investigate whether the expression of p16 IHC or HR-HPV is associated with survival in PSCC, we conducted a single institution analysis of 143 patients with a diagnosis of PSCC and, available tissue were tested for p16 IHC staining patterns, histological subtype, tumor grade, and lymphovascular invasion (LVI) by an experienced pathologist. HR-HPV status using the Cobas PCR Assay or the RNAScope high-risk HPV in situ hybridization kit were also assessed. Patient characteristics were summarized using descriptive statistics of clinico-pathologic variables. Kaplan-Meier was used to estimate median overall survival (OS), cancer specific survival (CSS) and correlated with HPV, p16, and other study variables. Patients with p16+ tumors had a significantly longer median CSS in comparison to the p16- group (p = 0.004), with respective 5-year CSS probability of 88% (95% CI; 0.84, 1) versus 58% (95% CI; 0.55, 0.76; p = 0.004). HPV status did not predict survival outcomes. Multivariable analysis with respect to OS and CSS, showed that p16+ status was associated with a lower risk of death (HR = 0.36, 95%CI; 0.20-0.67, p = 0.001), and improved CSS (HR = 0.20, 95% CI; 0.07-0.54, p = 0.002) after adjusting for covariates. In conclusion, tumor p16 status via IHC was an easy to perform independent prognostic factor for OS and CSS that correlates with HR-HPV expression.

Convolutional Neural Networks for Classifying Cervical Cancer Types Using Histological Images.

Cervical cancer is the most common cancer among women worldwide. The diagnosis and classification of cancer are extremely important, as it influences the optimal treatment and length of survival. The objective was to develop and validate a diagnosis system based on convolutional neural networks (CNN) that identifies cervical malignancies and provides diagnostic interpretability. A total of 8496 labeled histology images were extracted from 229 cervical specimens (cervical squamous cell carcinoma, SCC, n = 37; cervical adenocarcinoma, AC, n = 8; nonmalignant cervical tissues, n = 184). AlexNet, VGG-19, Xception, and ResNet-50 with five-fold cross-validation were constructed to distinguish cervical cancer images from nonmalignant images. The performance of CNNs was quantified in terms of accuracy, precision, recall, and the area under the receiver operating curve (AUC). Six pathologists were recruited to make a comparison with the performance of CNNs. Guided Backpropagation and Gradient-weighted Class Activation Mapping (Grad-CAM) were deployed to highlight the area of high malignant probability. The Xception model had excellent performance in identifying cervical SCC and AC in test sets. For cervical SCC, AUC was 0.98 (internal validation) and 0.974 (external validation). For cervical AC, AUC was 0.966 (internal validation) and 0.958 (external validation). The performance of CNNs falls between experienced and inexperienced pathologists. Grad-CAM and Guided Gard-CAM ensured diagnoses interpretability by highlighting morphological features of malignant changes. CNN is efficient for histological image classification tasks of distinguishing cervical malignancies from benign tissues and could highlight the specific areas of concern. All these findings suggest that CNNs could serve as a diagnostic tool to aid pathologic diagnosis.

Retrospective Study of Oral Lichen Planus and Oral Lichenoid Lesions: Clinical Profile and Malignant Transformation.

Oral lichen planus (OLP) and other oral lichenoid lesions (OLL) are reported to have the potential of malignant transformation and dysplastic changes, turning into oral squamous cell carcinoma (SCC). While the world health organization (WHO) has classified OLP as a precancerous lesion of the oral cavity, there is still much debate among researchers about its risks and malignancy potential. The present study aimed to determine malignant transformation in OLP and OLL and understand related risk factors. This retrospective study was performed on 356 patients of the Oral Medicine Department of Dental School of Kerman Medical University from 1998 to 2020. All patients' records were gathered. In addition, patients were followed up routinely. Second biopsy was taken as needed. The samples, previously taken from the patients, were re-evaluated according to WHO histopathologic criteria for diagnosing OLP, OLL, dysplasia, and SCC by an experienced pathologist and compared with first reports. Dysplastic changes were observed in 6.2% of the patients. In more than half of the patients, dysplastic changes were present right from the start and 2.20% of the patients had experienced dysplastic changes averagely within 2.05 years of the onset of lesions. Multiple logistic regression showed that the risk of dysplasia increases with aging (p= 0.013), smoking (p= 0.0001), and thyroid disorders (p= 0.008). Given the rather high prevalence of oral lichen planus and lichenoid lesions, further research appears to be needed to determine the etiology of these lesions, malignant transformations, and the factors affecting this probability. Considering the findings, it is imperative to meticulously record the information of all patients with oral lichen planus and lichenoid lesions in the initial examinations as well as close follow-ups and employ diagnostic tools such as toluidine blue staining or even repeat biopsy when necessary.