Journal of Pathology Informatics最新文献

{"title":"Lessons learned from the validation of a machine learning-based colorectal carcinoma screening pipeline in sub-Saharan Africa","authors":"Alfred Githuka,&nbsp;Ulysses G.J. Balis,&nbsp;Ye Chan Kim,&nbsp;Eileen M. Weinheimer-Haus,&nbsp;Christopher L. Williams,&nbsp;Jerome Y. Cheng,&nbsp;Kelou Yao,&nbsp;John Blau,&nbsp;Jessica A. Baker,&nbsp;Priscilla Njenga,&nbsp;Winny Chepkemoi,&nbsp;Robert K. Parker,&nbsp;Akbar K. Waljee,&nbsp;Shahin Sayed,&nbsp;Mansoor N. Saleh","doi":"10.1016/j.jpi.2026.100691","DOIUrl":"10.1016/j.jpi.2026.100691","url":null,"abstract":"<div><h3>Background</h3><div>Artificial intelligence for digital pathology may improve cancer detection and workflow efficiency in low- and middle-income countries, but most models are trained and validated in high-income settings, creating uncertainty about generalizability and real-world deployability under differing pre-analytic and infrastructure conditions.</div></div><div><h3>Objective</h3><div>To validate a locally optimized colorectal carcinoma screening pipeline, retrained on data from a local Kenyan cohort, and to assess its feasibility as a sensitivity-forward assistive pre-screening workflow.</div></div><div><h3>Methods</h3><div>Formalin-fixed, paraffin-embedded hematoxylin and eosin-stained slides from 136 biopsy-proven colonic adenocarcinoma cases and 20 normal controls from 2 Kenyan institutions were digitized at 40× using a Grundium Ocus scanner. Whole-slide images were partitioned at the case level. A feature-enrichment and candidate tile-selection step identified adenocarcinoma-rich regions. Candidate 512 × 512 RGB tiles were then adjudicated by a gastrointestinal pathologist into adenocarcinoma-containing and benign/non-neoplastic tile libraries. These expert-curated tiles were used to train and validate a supervised convolutional neural network classifier, which generated tile-level malignancy probabilities. Tile scores were aggregated into case-level predictions using Top-K pooling and positive-tile burden rules to support sensitivity-forward screening.</div></div><div><h3>Results</h3><div>Among 14,452 tiles from 70 quality-controlled cases and 19 controls, the model achieved strong tile-level discrimination (sensitivity 0.9548, specificity 0.9926, F1 score 0.9686, and AUROC curve 0.966); case-level Top-K aggregation achieved an AUROC of 1.0.</div></div><div><h3>Conclusion</h3><div>A locally optimized computational pathology screening pipeline, based on data from a local population, demonstrated robust colorectal adenocarcinoma detection in a Kenyan cohort. These findings suggest that existing machine-learning models can be adapted to local populations through retraining with locally derived data.</div></div>","PeriodicalId":37769,"journal":{"name":"Journal of Pathology Informatics","volume":"22 ","pages":"Article 100691"},"PeriodicalIF":0.0,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148770082","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 and image analysis of different file formats on whole-slide images for colorectal and non-small cell lung carcinomas","authors":"Monika Lamba Saini,&nbsp;Darshan Kumar,&nbsp;Andreea Luchian,&nbsp;Hazel Smith,&nbsp;John D. Cochran","doi":"10.1016/j.jpi.2026.100693","DOIUrl":"10.1016/j.jpi.2026.100693","url":null,"abstract":"<div><div>Different scanners employ various image acquisition techniques and may store scanned images in proprietary formats or export them to a range of standard file formats. Understanding these technical limitations is essential for ensuring optimal scan quality and accurate image analysis. The present study compared image formats from two whole-slide imaging platforms—Leica's Aperio AT2 (.SVS) and 3DHistech's Panoramic P250 Flash III (.MRXS)—using artificial intelligence (AI) algorithms to quantify pixel-level, color-profile, and segmentation differences on 10 colorectal carcinoma (CRC) and 10 non-small cell lung carcinoma (NSCLC) cases. The same slides were scanned independently on both platforms at 20×, and native MRXS files were also converted to SVS using SlideMaster®, allowing us to isolate the effect of file-format translation (native MRXS vs. MRXS-to-SVS, same scanner) from that of scanner hardware (3DHistech-derived formats vs. native Aperio SVS). Native MRXS and MRXS-to-SVS files showed very high similarity in pixel values, color profile, and AI predictions, whereas native Aperio SVS files differed significantly across image size, pixel size, pixel number, and color profile—reflecting the combined effect of scanner hardware and file format rather than format alone. Given the limited sample size and absence of a pre-specified power analysis, these findings are exploratory and hypothesis-generating. We anticipate that an enterprise-level communication standard will emerge to harmonize image formats and support consistent, reproducible analysis by pathologists and AI software alike.</div></div>","PeriodicalId":37769,"journal":{"name":"Journal of Pathology Informatics","volume":"22 ","pages":"Article 100693"},"PeriodicalIF":0.0,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148770306","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":"Hierarchical vision transformers for Epstein-Barr virus status and histological subtype prediction in Hodgkin lymphoma whole-slide images","authors":"Zsolt Bedőházi,&nbsp;Zsófia Sztupinszki,&nbsp;Ragnar P. Kristjánsson,&nbsp;Mikkel Werling,&nbsp;Stephen Hamilton-Dutoit,&nbsp;Kristina L. Lauridsen,&nbsp;Lisa Ottander,&nbsp;Trine L. Plesner,&nbsp;Peter Hollander,&nbsp;Ingrid Glimelius,&nbsp;Lene Sjö,&nbsp;Estrid Høgdall,&nbsp;Carsten Utoft Niemann,&nbsp;Klaus Rostgaard,&nbsp;Péter Pollner,&nbsp;Henrik Hjalgrim,&nbsp;István Csabai","doi":"10.1016/j.jpi.2026.100689","DOIUrl":"10.1016/j.jpi.2026.100689","url":null,"abstract":"<div><div>Accurate stratification of Hodgkin lymphoma (HL) by immunologic/histological subtypes and Epstein-Barr virus (EBV) status is essential for epidemiological and translational research, yet large-scale testing is impractical and expensive. Digital pathology models that utilize routinely used hematoxylin and eosin (H&amp;E) whole-slide images (WSIs) could close this gap. We developed and validated a hierarchical Vision Transformer pipeline that aggregates cell-, patch-, and region-level context to predict EBV status and the three most prevalent immunological/histological HL subtypes: nodular sclerosis (NS), mixed cellularity (MC), and nodular lymphocyte-predominant HL (NLPHL)—from H&amp;E-stained WSIs, and additionally evaluated a standard attention-based multiple-instance learning (ABMIL) baseline for direct architectural comparison. The development pool comprised 1643 HL cases (1952 WSIs) from 18 Danish hospitals and was used for hospital-preserving 5-fold cross-validation; external validation was performed on an independent hold-out cohort of 458 cases (532 WSIs) from five hold-out hospitals. For subtype prediction, analyses were restricted to the 1560 cases belonging to NS, MC, or NLPHL. On the external EBV cohort (<span><math><mi>N</mi><mo>=</mo><mn>458</mn></math></span>) the hierarchical pipeline achieved an area under the receiver operating characteristic curve (ROC–AUC) of 0.73 (95% confidence interval (CI) 0.68–0.77), precision–recall (PR)–AUC 0.57 (95% CI 0.49–0.66) with recall (sensitivity) 0.74 (95% CI 0.67–0.80) and macro-F1 score 0.60 (95% CI 0.54, 0.65). For 3-class subtype prediction on 359 external cases, discrimination reached ROC–AUC 0.84 (95% CI 0.80–0.88) and PR–AUC 0.63 (95% CI 0.56–0.71) with a macro-F1 of 0.56 (95% CI 0.48, 0.64) and macro-recall 0.55 (95% CI 0.47, 0.63); residual errors were dominated by NS-MC confusions. An ABMIL baseline using the same patch embeddings achieved ROC–AUC 0.76 (95% CI 0.72–0.81) for EBV and 0.89 (95% CI 0.86–0.92) subtype prediction, outperforming the hierarchical model on both tasks. This multicenter study shows that both hierarchical and attention-based architectures can determine EBV status and major HL subtypes directly from routine H&amp;E slides with externally validated performance across hospitals, whereas the finding that the simpler baseline outperformed the hierarchical model suggests that strong foundation-model embeddings combined with attention-based pooling may reduce the need for explicit multi-scale modelling in cohorts of this size.</div></div>","PeriodicalId":37769,"journal":{"name":"Journal of Pathology Informatics","volume":"22 ","pages":"Article 100689"},"PeriodicalIF":0.0,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148640886","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":"Metric discordance under MIDOG++ domain shift: a leave-one-domain-out audit of F1, emitted-detection confidence calibration, and ROI count error","authors":"Cleverson de Souza","doi":"10.1016/j.jpi.2026.100705","DOIUrl":"10.1016/j.jpi.2026.100705","url":null,"abstract":"<div><h3>Background</h3><div>Mitotic-figure detectors are usually evaluated by detection accuracy, but use also depends on confidence and count reliability under domain shift.</div></div><div><h3>Methods</h3><div>We evaluated MIDOG++ with leave-one-domain-out testing across seven domains. Adaptive-bin detection expected calibration error (D-ECE1) measured the bin-mass-weighted mean absolute confidence-precision gap among post-non-maximum suppression detections retained at a 0.05 score floor; centroid-distance matching supplied the true positive/false positive labels. F1 and count error were evaluated at the fixed 0.5 operating threshold. Baseline and domain-adversarial RetinaNet arms were compared as a controlled perturbation, with exploratory domain-composition resampling and two secondary seeds.</div></div><div><h3>Results</h3><div>No directionally stable aggregate calibration advantage was established across three seed series. In the primary series, macro D-ECE1 was 0.0786 for baseline and 0.0741 for the domain-adversarial arm (paired difference, −0.0045), but two additional seed series produced paired differences of +0.0287 and −0.0068. A provenance audit confirmed the secondary series were generated under the locked protocol; they were reported as a sensitivity comparison and were not pooled with the primary point estimate. The fixed-domain case-bootstrap interval was below zero, whereas exploratory domain-composition resampling crossed zero (−0.0213 to +0.0112). The two domains with F1 gains had worse D-ECE1. Lymphoma/lymphosarcoma had the highest observed D-ECE1 in both arms and was systematically undercounted.</div></div><div><h3>Conclusions</h3><div>In this benchmark, F1, emitted-detection calibration, count behavior, and maximum observed-domain D-ECE1 were not interchangeable endpoints. The tested domain-adversarial configuration did not establish a directionally stable calibration benefit. Evaluations should report detection-specific calibration, retained-detection denominators, count consequences, and finite-benchmark observed-domain D-ECE1 alongside F1.</div></div>","PeriodicalId":37769,"journal":{"name":"Journal of Pathology Informatics","volume":"22 ","pages":"Article 100705"},"PeriodicalIF":0.0,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148770307","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":"Foundation artificial intelligence models enable high-accuracy diagnostic differentiation of hybrid neurofibroma/schwannoma using whole-slide images","authors":"Fabio Hellmann,&nbsp;Maxim Anokhin,&nbsp;Pascal Schimmler,&nbsp;Daniel Tippner,&nbsp;Stefan Plontke,&nbsp;Catena Kresbach,&nbsp;Martin Mensah,&nbsp;Elisabeth André,&nbsp;Anja Harder","doi":"10.1016/j.jpi.2026.100700","DOIUrl":"10.1016/j.jpi.2026.100700","url":null,"abstract":"<div><div>Peripheral nerve sheath tumors encompass a heterogeneous group including schwannoma, neurofibroma, and hybrid neurofibroma/schwannoma (HNS). Accurate differentiation is crucial due to distinct biological behavior, different management, and a possible assignment to a genetic disease. We investigated whether deep learning (DL)-based artificial intelligence (AI) reliably distinguished HNS from schwannomas and neurofibromas using whole-slide images (WSIs). Utilizing a dataset of H&amp;E-stained WSIs from 115 tumors, we applied state-of-the-art foundation models (UNI, CONCH, ResNet50) for feature extraction, combined with multiple instance learning algorithms CLAM and mMIL, and a patch-based classifier.</div><div>Our models achieved high validation performance, with macro-area under the receiver operating characteristic curve (AUC-ROC) values up to 1.0 for UNI-based MIL models. Attention maps correlated well with annotations. HNS were reliably distinguished from schwannoma and neurofibroma using a pretrained AI workflow that was easy to interpret. The choice of foundation model and stain normalization emerged as major performance drivers, with UNI-based mCLAM achieving the highest validation performance, whereas a ResNet50-based mCLAM model with stain normalization reached the best HNS accuracy (0.96) and F1 (0.98) on the test set. Patch-level classifiers with CONCH and UNI embeddings showed slightly lower three-class performance on the validation set (macro AUC-ROC approximately 0.8–0.87), but supported data-driven thresholds for the proportion of Sw-like and Nf-like tissue used for slide-level HNS classification, and a ResNet50-based patch model achieved an HNS accuracy of 0.79 and F1 of 0.88 on the test set.</div><div>We demonstrated that minimal preprocessing, combined with an interpretable output, quantified tumor components with very high accuracy. This study supports the feasibility of DL tools for the nuanced differentiation of schwannoma and neurofibroma from HNS. Our approach can therefore improve diagnostic accuracy. Limitations include the dataset, technical artifacts, and discrepancies between validation and test performance across encoders. Therefore, future work should include external validation across scanners, labs and centers, protocols, prospective assessment of workflow integration, and benchmarking against interobserver variability to confirm clinical utility and generalizability.</div></div>","PeriodicalId":37769,"journal":{"name":"Journal of Pathology Informatics","volume":"22 ","pages":"Article 100700"},"PeriodicalIF":0.0,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148770083","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":"Artificial intelligence for lung cancer classification in cytology specimens: A systematic review and diagnostic test accuracy meta-analysis of benign–malignant diagnosis and ADC/SCC/SCLC subtyping","authors":"Ayan Myssayev,&nbsp;Amin Tamadon,&nbsp;Nadiar M. Mussin,&nbsp;Reza Shirazi,&nbsp;Afshin Zare","doi":"10.1016/j.jpi.2026.100704","DOIUrl":"10.1016/j.jpi.2026.100704","url":null,"abstract":"<div><h3>Background</h3><div>Cytology specimens are essential for lung cancer diagnosis, yet distinguishing benign from malignant cells and subtyping adenocarcinoma, squamous cell carcinoma, and small-cell lung cancer remains challenging due to limited cellular material.</div></div><div><h3>Methods</h3><div>We systematically searched seven databases for studies evaluating artificial intelligence-based lung cancer classification in cytology. After screening, 32 studies were included, of which 20 contributed 33 analyzable two-by-two data rows for diagnostic test accuracy meta-analysis. Outcomes were pooled using bivariate logit random-effects models. Summary receiver operating characteristic curves, Fagan nomograms, Deeks' test for publication bias, and quality assessments using QUADAS-2 and artificial intelligence-specific tools were performed.</div></div><div><h3>Results</h3><div>For benign versus malignant diagnosis, 18 data rows encompassing 3464 observations yielded a pooled sensitivity of 89.3% (95% confidence interval, 84.1–92.9), a specificity of 88.5% (95% confidence interval, 83.2–92.3), and an area under the curve of 0.952. At a median prevalence of 58.2%, the post-test probability of malignancy after a positive artificial intelligence result was 91.2%, whereas a negative result yielded a post-test probability of 14.6%. For one-versus-rest subtyping among adenocarcinoma, squamous cell carcinoma, and small-cell lung cancer, 14 data rows comprising 2687 observations demonstrated a pooled sensitivity of 77.5%, specificity of 89.5%, and area under the curve of 0.901. Small-cell lung cancer showed the strongest performance (specificity 93.2%, area under the curve 0.949), whereas squamous cell carcinoma had lower sensitivity (67.0%). Substantial heterogeneity was observed across studies. Deeks' test suggested potential small-study effects, and the risk of bias was high in all included studies.</div></div><div><h3>Conclusion</h3><div>Artificial intelligence-based cytology classification achieves high diagnostic accuracy for benign-malignant discrimination and shows promising but less mature performance for histological subtyping. Prospective external validation, standardized reporting, and clinically integrated human–artificial intelligence studies are required before routine clinical deployment.</div></div>","PeriodicalId":37769,"journal":{"name":"Journal of Pathology Informatics","volume":"22 ","pages":"Article 100704"},"PeriodicalIF":0.0,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148854390","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":"How does AI perform compared to human expert panels in medical Delphi studies? A pilot study through the lens of pathology","authors":"Joshua Pantanowitz ,&nbsp;Christopher D. Manko ,&nbsp;John Majewski ,&nbsp;M. Alvaro Berbís ,&nbsp;Alton B. Farris III ,&nbsp;Aly Karsan ,&nbsp;Andrey Bychkov ,&nbsp;Antonio Luna ,&nbsp;Bethany Williams ,&nbsp;Brett Delahunt ,&nbsp;Catarina Eloy ,&nbsp;David S. McClintock ,&nbsp;Eva Hollemans ,&nbsp;J. Mark Tuthill ,&nbsp;Jeroen Van der Laak ,&nbsp;Jerome Y. Cheng ,&nbsp;Jochen K. Lennerz ,&nbsp;John H. Sinard ,&nbsp;Jose Aneiros-Fernandez ,&nbsp;Julien Calderaro ,&nbsp;Hooman Rashidi","doi":"10.1016/j.jpi.2026.100661","DOIUrl":"10.1016/j.jpi.2026.100661","url":null,"abstract":"<div><h3>Background</h3><div>Since their inception, Delphi studies have been a key part of medical literature. They consist of an expert panel tasked with coming to consensus on answers to various questions where obtaining objective results is difficult or impossible, with ranked responses based on a Likert scale. The ability of artificial intelligence (AI), particularly large language models (LLMs), to perform this role traditionally assigned to a panel of experts has been scarcely explored in medicine. This study accordingly aimed to explore the feasibility of an “AI-run” Delphi study applied to the practice of pathology.</div></div><div><h3>Methods</h3><div>A prior human-based Delphi study (PMID: <span><span>36603288</span><svg><path></path></svg></span>) employed to forecast the future role of AI in pathology was repeated, but this time with LLMs (Llama 3, ChatGPT-4, and ChatGPT-3.5 based on availability at the time of the study). This was done at various temperature settings (0, 0.7, and 1.0), a measurement of how much an LLM prioritizes determinism versus creativity. Low temperature caused the models to be more deterministic and focused, whereas high temperature increased creativity. “Delphi-GPT” was created to automate prompts that entailed 5 trials for 180 questions, leading to data that were compared to the original human expert panel.</div></div><div><h3>Findings</h3><div>All LLM and temperature combinations were able to reach consensus for a greater percentage of the 180 questions posed than human experts. Newer ChatGPT-4 and Llama 3 models performed better than ChatGPT-3.5. Whereas AI models and human experts did not always agree, the amount of agreement increased when the temperature setting was increased across all LLMs.</div></div><div><h3>Interpretation</h3><div>LLMs are shown here to successfully be able to simulate a Delphi study in medicine. The data show that generative AI models were consistently able to reach greater degrees of consensus than human experts in their responses to 180 prompts related to the future practice of pathology. This serves as a proof-of-concept that one day, pending further robust methodological validation, AI could even serve as a surrogate for de novo Delphi studies that ordinarily would have relied on feedback from a panel of experts. The reliability of consensus/concordance achieved will depend upon the combination of LLM and temperature setting selected.</div></div>","PeriodicalId":37769,"journal":{"name":"Journal of Pathology Informatics","volume":"21 ","pages":"Article 100661"},"PeriodicalIF":0.0,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147858393","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":"Defining hotspot for estimation of Ki-67 proliferation index of neuroendocrine tumors: QuPath algorithm vs. manual assessment","authors":"Ruben Oganesyan ,&nbsp;Osman Yilmaz ,&nbsp;Yevgen Chornenkyy ,&nbsp;Kanchan Kantekure ,&nbsp;Yuho Ono ,&nbsp;Buthania Hakamy ,&nbsp;Nandan Padmanabha ,&nbsp;Vikram Deshpande ,&nbsp;Raul S. Gonzalez ,&nbsp;Monika Vyas","doi":"10.1016/j.jpi.2026.100655","DOIUrl":"10.1016/j.jpi.2026.100655","url":null,"abstract":"<div><h3>Background</h3><div>Ki-67 proliferation index (PI) is essential for grading well-differentiated neuroendocrine tumors (WD-NETs). Pathologists traditionally assess Ki-67 PI by identifying hotspots and manually counting the positive cells among negative cells, which is expressed as a percentage. We developed an algorithm to objectively determine Ki-67 hotspots and calculate PI in WD-NETs, comparing its results with pathologists' selected hotspots to assess reliability.</div></div><div><h3>Methods</h3><div>Hotspots for gastroenteropancreatic WD-NETs (<em>n</em> = 20) were manually annotated on whole-slide images (WSIs) by six pathologists and compared with algorithm-selected areas. Ki-67 (DAKO, MIB-1 clone, and prediluted) scoring was performed using QuPath's custom object classification algorithm. Pathologists identified hotspots on WSI, captured images, and submitted them for PI determination using the same algorithm. Ki-67 PI was translated to grade per WHO classification (G1: &lt;3%, G2: 3–20%, G3: &gt;20%). A pathologist's consensus grade was determined based on majority pathologist grading (&gt;3/6) for each case. Fleiss's Kappa was used to assess inter-pathologist agreement, Cohen's Kappa was used to evaluate the agreement between pathologists and the algorithm, and Friedman test was used for hotspot area variability analysis.</div></div><div><h3>Results</h3><div>Pathologists showed moderate agreement (Fleiss's Kappa = 0.42, 80% agreement), whereas pathologist–algorithm agreement was fair (Cohen's Kappa = 0.32, 58.9% agreement). Among cases with pathologist consensus grade (<em>n</em> = 19), the algorithm assigned a higher grade in 8 cases (42%). In 60% of cases, hotspots overlapped between methods. There was significant hotspot area variability (Friedman statistic: 95.97, <em>p</em> &lt; 0.001).</div></div><div><h3>Conclusion</h3><div>Manual Ki-67 hotspot assessment is subjective, leading to grading variability. Algorithm-based assessment enhances reproducibility, though this occasionally leads to tumor upgrading, highlighting the need for standardization and further validation.</div></div>","PeriodicalId":37769,"journal":{"name":"Journal of Pathology Informatics","volume":"21 ","pages":"Article 100655"},"PeriodicalIF":0.0,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147750342","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":"PATHOS: Pathology attention framework for treatment response stratification in ovarian high-grade serous carcinomas following neoadjuvant chemotherapy on H&E images","authors":"Francesca Miccolis ,&nbsp;Marta Lovino ,&nbsp;Oskari Lehtonen ,&nbsp;Johanna Hynninen ,&nbsp;Sampsa Hautaniemi ,&nbsp;Anni Virtanen ,&nbsp;Elisa Ficarra","doi":"10.1016/j.jpi.2026.100545","DOIUrl":"10.1016/j.jpi.2026.100545","url":null,"abstract":"<div><div>Ovarian high-grade serous carcinoma (ovarian HGSC) is a clinically challenging disease with a poor prognosis, particularly for patients receiving neoadjuvant chemotherapy (NACT) before debulking surgery. In this study, we evaluate the progression-free interval (PFI) after NACT based on hematoxylin and eosin-stained whole-slide images (WSIs) of omental tumor tissue. Digital pathology tools are emerging, aiming at assisting pathologists in diagnosis and analysis; however, distinguishing features associated with response to NACT remain elusive. Multiple instance learning (MIL) coupled with attention mechanisms has shown promise in predicting treatment response from WSIs. Additionally, segmentation tools can identify and delineate regions in WSIs. Whereas some efforts have been made to develop explainable models for clinical outcome, there remains a need for genuinely interpretable models for pathologists. This article introduces the PATHOS framework, a novel approach to explaining crucial features of treatment response based on the PFI time in NACT treated patients from WSIs. PATHOS is composed of three blocks: (1) MIL block to identify informative regions, (2) panoptic segmentation and downstream analysis block for feature computation, and (3) classification block to predict the PFI. The results demonstrate that PATHOS enhances the interpretability of response to NACT in ovarian HGSC patients by highlighting pathologically significant features relevant to PFI prediction, such as tumor cell morphology, stromal abundance, and the spatial distribution of stromal regions. Furthermore, PATHOS identifies approximately 10% of the total WSI area as an informative region for clinical outcome.</div></div>","PeriodicalId":37769,"journal":{"name":"Journal of Pathology Informatics","volume":"21 ","pages":"Article 100545"},"PeriodicalIF":0.0,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147656619","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":"Pathologists vs. generative AI: A Turing Test for synthetic bone marrow cytology images","authors":"Justin Chang,&nbsp;Chelsea Kidd,&nbsp;Bryce Hatfield","doi":"10.1016/j.jpi.2026.100609","DOIUrl":"10.1016/j.jpi.2026.100609","url":null,"abstract":"","PeriodicalId":37769,"journal":{"name":"Journal of Pathology Informatics","volume":"21 ","pages":"Article 100609"},"PeriodicalIF":0.0,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148178238","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}