Journal of Pathology Informatics最新文献

{"title":"PD-L1 expression assessment in Angiosarcoma improves with artificial intelligence support","authors":"F.H. Reith ,&nbsp;A. Jarosch ,&nbsp;J.P. Albrecht ,&nbsp;F. Ghoreschi ,&nbsp;A. Flörcken ,&nbsp;A. Dörr ,&nbsp;S. Roohani ,&nbsp;F.M. Schäfer ,&nbsp;R. Öllinger ,&nbsp;S. Märdian ,&nbsp;K. Tielking ,&nbsp;P. Bischoff ,&nbsp;N. Frühauf ,&nbsp;F. Brandes ,&nbsp;D. Horst ,&nbsp;C. Sers ,&nbsp;D. Kainmüller","doi":"10.1016/j.jpi.2025.100447","DOIUrl":"10.1016/j.jpi.2025.100447","url":null,"abstract":"<div><div>Tumoral PD-L1 expression is assessed to weigh immunotherapy options in the treatment of various types of cancer. To determine PD-L1 expression, each tumor cell needs to be assessed to calculate the percentage of PD-L1 positive tumor cells, called tumor proportion score (TPS). Pathologists cannot evaluate each cell individually due to time constraints and thus need to approximate TPS, which has been shown to result in low concordance rates.</div><div>Decision quality could be improved by an AI-based TPS prediction tool which serves as a “second opinion”. Establishing such a tool requires a certain amount of training data, which manifests a bottleneck for rare cancer types such as Angiosarcoma.</div><div>To address this challenge, we developed and open sourced a pipeline that leverages pre-trained and generalist models to achieve strong TPS prediction performance on limited data. Pathologists were asked to reassess patients for which their TPS strongly disagreed with the AI's prediction. In many of these cases, pathologists updated their TPS score, improving their assessment, thus demonstrating the technical feasibility and practical value of AI-based TPS scoring assistance for rare cancers.</div></div>","PeriodicalId":37769,"journal":{"name":"Journal of Pathology Informatics","volume":"18 ","pages":"Article 100447"},"PeriodicalIF":0.0,"publicationDate":"2025-05-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144166082","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Digital morphometry illustrates a relationship between percentage of ductal carcinoma in-situ in breast needle core biopsy and margin status at lumpectomy","authors":"Alexander R. Gross ,&nbsp;Gerald R. Hobbs Jr. ,&nbsp;Luis Samayoa ,&nbsp;Stell Santiago","doi":"10.1016/j.jpi.2025.100445","DOIUrl":"10.1016/j.jpi.2025.100445","url":null,"abstract":"<div><div>Candidates for breast conserving surgery are selected based on imprecise variables and there is uncertainty surrounding the risk of complicated margins. Earlier estimates made with light microscopy revealed the correlation between percentage of needle core biopsy ductal carcinoma in-situ and positive lumpectomy margin status. We now study this association utilizing the precision of digital pathology. One hundred and seventy-nine lumpectomy specimens of pure ductal neoplasia were identified and their pathological, clinical, and radiological parameters retrieved. Each lumpectomy had a corresponding needle core biopsy for histological review. Virtually all cases exhibited a luminal A phenotype. Eighty-three cases showed positive margins and ninety-six cases, negative margins. We used the 2019 American College of Breast Surgeons Lumpectomy Consensus Guidelines to define margin status. For each case, by analog microscopy, we selected a single needle core biopsy slide with the greatest absolute quantity of carcinoma in-situ; each selected slide was submitted for digital whole slide imaging. Digital images were manually annotated for carcinoma in-situ, invasive carcinoma, stroma, and fat strictly based on morphology. Morphometric variables were compiled and compared to the corresponding lumpectomy margin status. Increases in percent ductal carcinoma in-situ are associated with greater odds of positive lumpectomy margins (<em>P</em> &lt; 0.05). Above 10% carcinoma in-situ all but one case showed positive margins. This prediction was more precise compared to the association between pre-operative radiological studies and margin status, particularly in cases of pure ductal carcinoma in-situ. Our work suggests that needle core biopsy percentage of ductal carcinoma in-situ maybe clinically useful in assessing the risk of a positive lumpectomy margin in select patients. A larger, multi-institutional study can further elucidate if pathological reporting of needle core biopsies with pure ductal breast neoplasia should include a percentage needle core biopsy ductal carcinoma in-situ.</div></div>","PeriodicalId":37769,"journal":{"name":"Journal of Pathology Informatics","volume":"18 ","pages":"Article 100445"},"PeriodicalIF":0.0,"publicationDate":"2025-04-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144155024","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Technical note: Impact of tissue section thickness on accuracy of cell classification with a deep learning network","authors":"Ida Skovgaard Christiansen ,&nbsp;Rasmus Hartvig ,&nbsp;Thomas Hartvig Lindkær Jensen","doi":"10.1016/j.jpi.2025.100440","DOIUrl":"10.1016/j.jpi.2025.100440","url":null,"abstract":"<div><h3>Introduction</h3><div>We are currently developing a cell classification system intended for routine histopathology. During observation, cells of interest are added to a deep learning (DL) network, which after training classifies the remaining cells of interest with high and immediately validatable accuracy. In this study, we identify the optimal histological microsection thickness for this process and describe in high detail the morphological differences introduced by variation in microsection thickness.</div></div><div><h3>Method</h3><div>From HE-stained digitized sections of liver cut manually at 5 thicknesses and on an automated microtome (DS), hepatocytes and non-hepatocytes were manually annotated and loaded into a DL convolutional neural network (ResNet). The network was trained at different settings to identify the thickness with optimal relation between number of training cells and validation accuracy. To shed interpretable light on the impact of thickness, exhaustive morphological details of the annotated cells were quantified and the differences between hepatocytes and non-hepatocytes were analyzed with random forest.</div></div><div><h3>Results</h3><div>Classifying hepatocytes from DS sections clearly resulted in highest validation accuracy with least number of cells and for the remaining thicknesses a trend towards thin sections being more efficient was observed. Random forest analysis generally identified variations in nuclear granularity as the most important features in distinguishing cells. In DS and the thinner tissue sections, nuclear granularity features were more distinguished.</div></div><div><h3>Conclusion</h3><div>Microsections cut with DS in particular and thin sections in general are better suited for the intended cell classification system.</div></div>","PeriodicalId":37769,"journal":{"name":"Journal of Pathology Informatics","volume":"17 ","pages":"Article 100440"},"PeriodicalIF":0.0,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143874460","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Fully automatic HER2 tissue segmentation for interpretable HER2 scoring","authors":"Mathias Öttl ,&nbsp;Jana Steenpass ,&nbsp;Frauke Wilm ,&nbsp;Jingna Qiu ,&nbsp;Matthias Rübner ,&nbsp;Corinna Lang-Schwarz ,&nbsp;Cecilia Taverna ,&nbsp;Francesca Tava ,&nbsp;Arndt Hartmann ,&nbsp;Hanna Huebner ,&nbsp;Matthias W. Beckmann ,&nbsp;Peter A. Fasching ,&nbsp;Andreas Maier ,&nbsp;Ramona Erber ,&nbsp;Katharina Breininger","doi":"10.1016/j.jpi.2025.100435","DOIUrl":"10.1016/j.jpi.2025.100435","url":null,"abstract":"<div><div>Breast cancer is the most common cancer in women, with HER2 (human epidermal growth factor receptor 2) overexpression playing a critical role in regulating cell growth and division. HER2 status, assessed according to established scoring guidelines, offers important information for treatment selection. However, the complexity of the task leads to variability in human rater assessments. In this work, we propose a fully automated, interpretable HER2 scoring pipeline based on pixel-level semantic segmentations, designed to align with clinical guidelines. Using polygon annotations, our method balances annotation effort with the ability to capture fine-grained details and larger structures, such as non-invasive tumor tissue.</div><div>To enhance HER2 segmentation, we propose the use of a Wasserstein Dice loss to model class relationships, ensuring robust segmentation and HER2 scoring performance. Additionally, based on observations of pathologists' behavior in clinical practice, we propose a calibration step to the scoring rules, which positively impacts the accuracy and consistency of automated HER2 scoring. Our approach achieves an F1 score of 0.832 on HER2 scoring, demonstrating its effectiveness. This work establishes a potent segmentation pipeline that can be further leveraged to analyze HER2 expression in breast cancer tissue.</div></div>","PeriodicalId":37769,"journal":{"name":"Journal of Pathology Informatics","volume":"17 ","pages":"Article 100435"},"PeriodicalIF":0.0,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143739022","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Comparison of the efficiency of digital pathology with the conventional methodology for the diagnosis of biopsies in an anatomical pathology laboratory in Spain","authors":"J.I. Echeveste ,&nbsp;L. Alvarez-Gigli ,&nbsp;D. Carcedo ,&nbsp;Y. Soto-Serrano ,&nbsp;M.D. Lozano","doi":"10.1016/j.jpi.2025.100439","DOIUrl":"10.1016/j.jpi.2025.100439","url":null,"abstract":"<div><h3>Background/objective</h3><div>Digital pathology (DP) encompasses the digitization of processes related to the acquisition, storage, transmission, and analysis of pathological data, contrasting with conventional methodology (CM) using optical microscopes. This study evaluates the efficiency of DP versus CM in a Spanish pathology department.</div></div><div><h3>Methods</h3><div>Observational, retrospective, and non-interventional study comparing biopsy samples from 2021 (cases diagnosed using CM) and 2022 (using DP). Variables analyzed were the pathologist who made the diagnosis, the number of slides, and the case area. Outcome efficiency variables were the turnaround-time (TaT), pending cases (active cases each pathologist accumulates daily), and pathologist workload. A significance level of 5% was established, and an exploratory cost-analysis was also performed.</div></div><div><h3>Results</h3><div>11,922 cases were analyzed: 5,836 and 6,086 diagnosed with CM and DP methodologies, respectively. Mean TaT for CM-diagnosed cases was 10.58 (standard deviation [SD] 7.10) days, compared to 6.86 (SD 5.10) days for DP-diagnosed cases, reflecting a reduction of 3.72 days (<em>P</em> &lt; 0.001). With DP, the average reduction in pending cases over a year was around 25 cases, with peaks of 100 fewer pending cases during high workload months. Additionally, DP decreased the pathologist workload by 29.2% on average, with reductions exceeding 50% during peak months.</div></div><div><h3>Conclusion</h3><div>Our study is the first in Spain to compare the efficiency and costs of DP and CM. DP demonstrated significant efficiency improvements over CM, reducing TaT and pathologist workload. Despite higher initial costs, DP's operational benefits indicate its potential as a transformative diagnostic tool. Further studies are needed to evaluate its long-term cost-effectiveness.</div></div>","PeriodicalId":37769,"journal":{"name":"Journal of Pathology Informatics","volume":"17 ","pages":"Article 100439"},"PeriodicalIF":0.0,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143883255","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"ARMSprimer3: An open-source primer design Python program for amplification refractory mutation system PCR (ARMS-PCR)","authors":"Jingwen Guo ,&nbsp;Jeremy Grojean ,&nbsp;Huazhang Guo","doi":"10.1016/j.jpi.2025.100442","DOIUrl":"10.1016/j.jpi.2025.100442","url":null,"abstract":"<div><div>Single nucleotide polymorphisms (SNPs) are DNA sequence variations of a single base pair. They are the underlying mechanism of most human genetic variation and etiology of many heritable human diseases. SNPs can be reliably detected by amplification refractory mutation system PCR (ARMS-PCR). ARMS-PCR is based on allele-specific PCR primers that only amplify DNA samples with the target allele and do not amplify DNA samples without the target allele. In addition to the allele-specific mismatch at the 3′ end, ARMS-PCR introduces additional deliberate mismatches near the 3′ end of the allele-specific primers to further destabilize the non-specific binding and priming on non-target alleles. This modification increases the specificity for SNP detection, but also increases the complexity of PCR primer design. We developed ARMSprimer3, a Python program to automate the ARMS-PCR primer design process. The validity of ARMSprimer3 was confirmed by successfully using it to develop diagnostic tests in our clinical molecular diagnostic lab. ARMSprimer3 is open-source software and can be freely downloaded from <span><span>https://github.com/PCRPrimerDesign/ARMSprimer3</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":37769,"journal":{"name":"Journal of Pathology Informatics","volume":"17 ","pages":"Article 100442"},"PeriodicalIF":0.0,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143899418","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Erratum to “Artificial intelligence-based tumor segmentation in mouse models of lung adenocarcinoma” [Journal of Pathology Informatics Volume 13, 2022, 100007]","authors":"Alena Arlova ,&nbsp;Chengcheng Jin ,&nbsp;Abigail Wong-Rolle ,&nbsp;Eric S. Chen ,&nbsp;Curtis Lisle ,&nbsp;G. Thomas Brown ,&nbsp;Nathan Lay ,&nbsp;Peter L. Choyke ,&nbsp;Baris Turkbey ,&nbsp;Stephanie Harmon ,&nbsp;Chen Zhao","doi":"10.1016/j.jpi.2025.100425","DOIUrl":"10.1016/j.jpi.2025.100425","url":null,"abstract":"","PeriodicalId":37769,"journal":{"name":"Journal of Pathology Informatics","volume":"17 ","pages":"Article 100425"},"PeriodicalIF":0.0,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143874461","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Comparison of artificial intelligence image processing with manual leucocyte differential to score immune cell infiltration in a mouse infection model of cystic fibrosis","authors":"Madeline G. Williams,&nbsp;Zachary J. Faber,&nbsp;Thomas J. Kelley","doi":"10.1016/j.jpi.2025.100438","DOIUrl":"10.1016/j.jpi.2025.100438","url":null,"abstract":"<div><div>Immune cell differentials are most commonly performed manually or with the use of automated cell sorting devices. However, manual review by research personnel can be both subjective and time consuming, and cell sorting approaches consume samples and demand additional reagents to perform the differential. We have created an artificial intelligence (AI) image processing pipeline using the Biodock.ai platform to classify immune cell types from Giemsa-stained cytospins of mouse bronchoalveolar lavage fluid. Through multiple rounds of training and refinement, we have created a tool that is as accurate as manual review of slide images while removing the subjectivity and making the process mostly hands off, saving researcher time for other tasks and improving core turnaround for experiments.</div><div>This AI-based image processing is directly compatible with current workflows utilizing stained slides, in contrast to a change to a flow cytometry-based approach, which requires both specialized equipment, reagents, and expertise.</div></div>","PeriodicalId":37769,"journal":{"name":"Journal of Pathology Informatics","volume":"17 ","pages":"Article 100438"},"PeriodicalIF":0.0,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143838643","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Efficient merging and validation of deep learning-based nuclei segmentations in H&E slides from multiple models","authors":"Jagadheshwar Balan,&nbsp;Shannon K. McDonnell,&nbsp;Zachary Fogarty,&nbsp;Nicholas B. Larson","doi":"10.1016/j.jpi.2025.100443","DOIUrl":"10.1016/j.jpi.2025.100443","url":null,"abstract":"<div><div>Characterizing cellular composition in tissue samples offers fundamental insights into functional and biological processes. Understanding the abundance or lack of specific cell types, such as inflammatory cells in the context of microenvironments such as tumor can help guide disease progression and personalized medicine. Several clinical laboratory methods to characterize the cellular composition are limited by scalability and high-costs. Digitizing pathology slides and applying deep learning (DL) models have enabled efficient and cost-effective nuclei segmentation and cell type quantification; however, the DL-models are limited by their inability to segment specific cell types and specific models may be more effective than others at certain tasks. Consequently, there remains a need for methods that leverage the strengths of multiple models to efficiently integrate nuclei segmentation for various cell types. In this study, we propose a novel solution for integrating nuclei segmentation from multiple DL-methods on hematoxylin and eosin slides from 471 normal prostate samples and highlight the limitations of using a single DL-method. We validate the DL-derived cell type proportions, by comparing against estimates from a manual pathologist review and show that the integrated approach results in higher concordance over the individual models. We further validate the derived cell type proportions from the DL-methods by their ability to explain the variance of RNA gene expression. The integrated approach yields robust cell type proportions that explain the variance of the gene expression with 12% and 22% relative improvement than current state-of-the-art model and manual pathologist review, respectively. The subset of 403 genes with high explained variation (&gt;30%) by epithelial proportion were significantly enriched for relevant biological pathways. These findings indicate that ensemble approaches to nuclei segmentation and cell-type classification may provide more accurate representations of cellular composition from digitized slides.</div></div>","PeriodicalId":37769,"journal":{"name":"Journal of Pathology Informatics","volume":"17 ","pages":"Article 100443"},"PeriodicalIF":0.0,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143927403","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Erratum to “MiNuGAN: Dual segmentation of mitoses and nuclei using conditional GANs on multi-center breast H&E images” [Journal of Pathology Informatics Volume 13, 2022, 100002]","authors":"Salar Razavi ,&nbsp;Fariba D. Khameneh ,&nbsp;Hana Nouri ,&nbsp;Dimitrios Androutsos ,&nbsp;Susan J. Done ,&nbsp;April Khademi","doi":"10.1016/j.jpi.2025.100426","DOIUrl":"10.1016/j.jpi.2025.100426","url":null,"abstract":"","PeriodicalId":37769,"journal":{"name":"Journal of Pathology Informatics","volume":"17 ","pages":"Article 100426"},"PeriodicalIF":0.0,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143883256","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}