The integration of artificial intelligence into pathology is transforming the assessment of histological and immunohistochemical (IHC) slides, offering opportunities to reduce variability and streamline diagnostics. In practical terms, most available tools and research models emulate the diagnostic capabilities of pathologists by detecting, grading, and classifying tumours and other diseases. More recent applications have moved beyond mimicry, aiming to predict established biomarkers, such as microsatellite instability or IHC-based markers, and to tackle even more ambitious tasks, such as directly predicting patient prognosis from H&E whole slide images. Remarkably, novel computational tools are now being designed as companion diagnostic assays, linking the automated evaluation of specific IHC biomarkers to the prediction of response to specific drugs, potentially marking a new chapter in the evolution of digital and computational pathology. The TROPION-PanTumor01 trial recently demonstrated the superiority of a supervised machine learning model (termed the quantitative continuous score [QCS] by the vendor) in assessing TROP2 IHC compared with human scoring, promising better stratification of patients with non-small cell lung cancer for treatment with datopotamab deruxtecan. The same approach has shown promise in refining HER2 (human epidermal growth factor receptor 2) and PD-L1 (programmed death-ligand 1) evaluations, revealing patient subgroups that may benefit from targeted therapies. Moreover, other similar approaches are progressively reaching the market, posing significant opportunities and challenges for clinicians involved in the care of patients with cancer. This Perspective is promoted by the European Society of Digital and Integrative Pathology (ESDIP, founded in 2016, and having long-standing experience in computational pathology, esdipath.org) and the European Interdisciplinary Society of Artificial Intelligence for Cancer Research (ESAC, a recently established initiative, founded in 2024, esac-network.eu), both bringing together clinicians, engineers and other professionals dedicated to the development and clinical translation of computational approaches aimed at improving patient care. It aims to provide an informed overview of novel computational pathology companion diagnostic tools, with a particular focus on the background that practicing pathologists and oncologists need to have with these tools, when transitioning from research to clinical practice, irrespective of their prior familiarity with computational approaches. © 2026 The Author(s). The Journal of Pathology published by John Wiley & Sons Ltd on behalf of The Pathological Society of Great Britain and Ireland.

Use of whole slide images for primary pathology evaluation of non-clinical toxicology studies in a GLP-validated environment.

Immune mechanisms in the pathogenesis of feline infectious peritonitis in renal tissue: Focus on lymphocytes and cytokines in effusive and non-effusive forms.

Hematopathology workflows are complex, since they include numerous data points necessary for guiding further testing, diagnosis, and patient management. The workflows start with complete blood cell counts, with subsequent morphologic evaluation of peripheral blood (PB) and bone marrow (BM). Digital pathology has the potential to revolutionize PB and BM assessment through the implementation of artificial intelligence for assisted and automated evaluation, but there remain major hurdles toward this ultimate goal, such as lack of regulatory oversight, data standardization, insufficient knowledge and training, and resistance to change, among others. This article reviews the current state of digitalization in the hematopathology practice, recent research using machine learning models for automated specimen analysis, outlines the advantages and barriers facing clinical implementation of artificial intelligence, and offers prospective artificial intelligence-driven clinical workflows for efficient and comprehensive clinical workup.

SparseXMIL: Leveraging sparse convolutions for context-aware and memory-efficient classification of whole slide images in digital pathology.

Colorectal cancer is the second leading cause of cancer-related mortality worldwide, posing a substantial burden on healthcare systems. Identifying DNA mismatch repair deficiency (dMMR) is critical for guiding treatment, yet conventional methods rely on labor-intensive DNA analysis. While deep-learning approaches have shown promise for predicting dMMR from histopathological images, most studies focus exclusively on tumor regions and single-scale representations. This study systematically evaluates the predictive value of tumor and non-tumor regions across multiple magnifications for dMMR prediction from whole-slide images (WSIs). A total of 24 different modeling approaches were evaluated, varying by tissue origin (tumor vs. non-tumor), magnification level (5x and 20x), and tile embedding strategy, including digital pathology foundation models. Tile embeddings were further trained with 1228 WSIs using multiple-instance learning (MIL) based approach. The best-performing configurations were selected for external evaluation. External testing was carried out on two independent cohorts consisting of 1010 and 457 WSIs, respectively. Non-tumorous regions demonstrated measurable predictive value, although performance remained lower than that obtained from tumor regions (F1 = 0.896, precision = 0.888, sensitivity = 0.594, specificity = 0.982). Among the nine models selected during internal validation, the top three models-one multi-scale approach and two models trained on 20x tumor regions-achieved F1 scores of 0.870-0.889 with precision of 0.885-0.920, sensitivity of 0.852, and specificity of 0.889-0.926. On external validation, the top three models, all based on foundation-model tile embeddings, achieved F1 scores of 0.916-0.919 on the first cohort and 0.928-0.934 on the second cohort. Across cohorts, specificity remained consistently high (0.964-0.992), while sensitivity ranged from 0.500 to 0.682. This study demonstrates that dMMR status in colorectal cancer can be effectively predicted from histopathological WSIs using MIL-based models, with moderate generalizability across independent cohorts. In addition to confirming the predictive value of tumor regions, the results reveal that non-tumorous tissue also contains detectable predictive signals, suggesting that microenvironmental features may contribute to dMMR-associated histological patterns. Furthermore, the use of foundation model-derived embeddings improved generalizability across datasets. Future work should explore integrating non-tumor tissue features and clinical data to further improve predictive performance.

Neoadjuvant systemic therapy has emerged as a strategy to improve outcomes in high-risk localized genitourinary malignancies. In bladder cancer, neoadjuvant cisplatin-based chemotherapy with or without immunotherapy is standard of care, with pathologic complete response (pCR) serving as a validated surrogate for survival after chemotherapy and a promising potential surrogate for survival after other neoadjuvant treatments like immunotherapy. Enfortumab vedotin and immune checkpoint inhibitors have expanded treatment options, with ongoing studies evaluating novel adjuvant approaches tailored to patients' postoperative circulating tumor DNA. In renal cell carcinoma (RCC), the field is nascent, no approved neoadjuvant regimens exist, and treatment outside clinical trials is not recommended. Early trials of tyrosine kinase inhibitor monotherapy showed limited pathologic responses, while immune checkpoint inhibitor-based combinations have demonstrated feasibility, safety, and the capacity to induce pathologic responses including pCR. Critical gaps remain in both diseases. In RCC, standardized pathologic response criteria are lacking, and surrogacy between pathologic end points and long-term outcomes has not been established. In bladder cancer, optimal post-pCR management and integration of novel agents require further study. Emerging technologies, particularly artificial intelligence (AI)-driven digital pathology, offer potential to enhance diagnosis, refine prognostic stratification, and predict treatment response, although prospective validation across diverse populations is needed. This chapter examines neoadjuvant therapy and pathologic response assessment in RCC and bladder cancer, explores pathologic biomarker development including AI applications, and highlights future directions to optimize therapeutic sequencing and outcomes.

Dissecting the prognostic role of metabolic markers in lung neuroendocrine Tumors: The MONET study.

Deep learning (DL)-based pathological image modelling and analysis approaches offer transformative potential for early cancer diagnostics, yet limited sample sizes and a lack of interpretability often hinder efficient clinical translation. Here, we present the interpretable Multi-Task Digital Pathology Model (iMDPath), an end-to-end, highly explainable multi-task deep learning framework that simultaneously addresses these challenges by integrating data augmentation, diagnostic prediction, and visualization of pathological image features. The iMDPath comprises three modules: Augmentation (iMDPath-Aug), Prediction (iMDPath-Pred), and Visualization (iMDPath-Vis). iMDPath-Aug incorporates a vector-quantized variational autoencoder (VQ-VAE) for enhanced data augmentation, capturing essential pathological features from limited datasets. A Swin Transformer-Based (Swin-B) predictor in the iMDPath-Pred module leverages the augmented data to achieve better performance than patch-level and foundation-model-based encoders such as InceptionV3 and Phikon across six diverse cancer pathology datasets, including gastric, breast, lung, and colorectal cancer. Finally, iMDPath-Vis, a novel visualization module combining the full gradient (FullGrad) and occlusion sensitivity analysis, provides pathologists with actionable insights by highlighting the specific tissue regions driving model predictions. Overall, iMDPath not only surpasses existing methods in diagnostic accuracy, sensitivity, and generalization across these datasets, but also offers a transparent and interpretable AI solution for precision oncology, paving the way for more reliable and efficient clinical decision-making.

IntroductionPrecisely grading renal cell carcinoma (RCC) through histopathology slides is a requisite to predict the cancer prognosis and select treatments, however, there is a considerable variability in the assessment between observers. While some AI systems mainly rely on transformer-based and nuclei-centric models, they are computationally very demanding so their clinical use is thus limited. Hence, there is a need for practical and easy-to-understand solutions that can be integrated into digital pathology workflows.MethodsWe built a hybrid framework that integrates U-Net-based tumor segmentation, convolutional feature extraction, nuclei-aware descriptors, stain normalization, and attention-based multiple instance learning for slide-level RCC grading. The framework was tested on 3 public datasets (TCGA-ccRCC, RCdpia, MMIST-ccRCC) with a cross-dataset validation approach. The metrics used for the performance evaluation were macro-F1 and quadratic weighted kappa (QWK).ResultsThe designed method yielded a macro-F1 of 0.94, QWK of 0.92, and accuracy of 0.95. Extracting tumor patches and performing aggregation based on attention resulted in the best improvements. The method was equally effective when tested with different datasets. Most of the errors made by the model were those within the clinical grading range variability. Average time for inference was around one minute and ten seconds per slide.ConclusionBy fine-tuning the convolutional pipeline, one can obtain a RCC grading capability that can be rivaled by very few, yet the model is efficient and interpretable, hence it will continue to be a strong candidate decision-support tool in digital pathology to be clinically deployed.

Identification of tissue-region-specific changes in glycosylation is crucial for understanding the pathogenesis of kidney diseases, yet it remains a great challenge. We developed a workflow that combines matrix-assisted laser desorption/ionization mass spectrometry imaging (MALDI-MSI) data with AI-based digital pathology annotations of kidney functional tissue units (FTUs) to profile N-glycan distributions within biopsy tissues. This approach can generate molecular-level data relevant to diverse pathological outcomes, thereby aiding in the elucidation of disease mechanisms. As a proof-of-concept, we demonstrate that this AI-based digital pathology approach to MALDI data segmentation enables the detection and differentiation of N-glycosylation within FTUs of healthy kidney tissue. We then elucidated differences in N-glycosylation between the diseased kidney tissue samples from patients with different diagnoses. Sialic acid N-glycans, which have been linked to various kidney diseases, displayed enrichment in the glomeruli and tubules of tissues from patients diagnosed with diabetic kidney disease (DKD), whereas they were enriched in the tubules and arteries from patients with acute kidney injury (AKI), in comparison to healthy tissue. Furthermore, we found that polylactosamine N-glycans were enriched only in the AKI samples, indicating their potential roles in tubular injury and inflammation. This workflow has the potential to bridge the gap between region-specific glycosylation and its implications on FTUs in diseases, paving the way for targeted molecular imaging studies in the kidney and other tissues.

Small tissue biopsies, including renal core biopsies, bone marrow trephines, gastrointestinal endoscopic samples, prostate needle cores, liver biopsies, and skin punch or shave specimens, are fundamental to contemporary diagnostic pathology. Their limited volume, focal sampling, and susceptibility to technical artifacts impose distinct interpretive challenges, often requiring semi-quantitative assessment within a restricted architectural context. Although artificial intelligence (AI) has rapidly expanded in digital pathology, most models have been developed using large surgical resection specimens, with comparatively limited attention to small biopsy material. This minireview examines current and emerging applications of AI in small biopsy diagnostics across renal, hepatic, gastrointestinal, hematopathological, dermatopathological, and urological pathology. Reported applications include glomerular segmentation and fibrosis quantification in renal biopsies; automated cellularity, blast detection, and fibrosis grading in bone marrow trephines; dysplasia and microorganism detection in gastrointestinal biopsies; quantitative steatosis and fibrosis assessment in liver samples; tumor detection and grading in prostate cores; and neoplastic pattern recognition in skin specimens. Despite encouraging performance in research settings, substantial barriers to routine clinical implementation remain, including limited dataset size, class imbalance, pre-analytical variability, inter-institutional heterogeneity, and insufficient external validation. We discuss methodological considerations relevant to diagnostic practice, including multi-institutional validation, stain normalization, multimodal integration with histochemistry and ancillary testing, explainability, and regulatory oversight. In the context of small biopsies, AI should be regarded as a quantitative adjunct to morphological interpretation rather than an autonomous diagnostic system. Careful integration within established histopathological workflows is essential to ensure reproducibility, safety, and clinical accountability.

Collaborative learning across medical institutions is essential for building robust and generalisable digital pathology models. Federated learning (FL) enables collaboration without centralising data, yet its adoption is limited by high communication costs, model heterogeneity, and privacy concerns. We propose Federated Deep Feature Prompting (FedDFP), an efficient FL framework tailored for heterogeneous clinical environments. FedDFP introduces lightweight, client-specific learnable prompts applied to patch-level embeddings from whole-slide images. By sharing only these compact prompts, FedDFP reduces communication overhead by over 99.9% compared to standard FL while improving classification accuracy. Extensive experiments on TCGA-IDH, CAMELYON16 and CAMELYON17 show that FedDFP consistently outperforms standard and personalised FL baselines, achieving mean AUC gains of 0.11-0.13 over local-only training and up to 0.10 over the strongest federated methods. FedDFP also converges 2-4× faster and remains effective across diverse feature extractors and multiple-instance learning architectures, demonstrating scalability, flexibility and privacy-aware collaboration.

Artificial intelligence (AI) and digital pathology have the potential to augment liver biopsy interpretation in MAFLD in clinical practice and trials assessment. However, attitudes and barriers to its implementation have not been systematically explored. A survey focusing on conventional liver histology, digital pathology and its AI applications in MAFLD/MASH was conducted among hepatologists and liver pathologists in the Asia Pacific region. AI-assisted digital pathology is perceived to be a valuable addition to existing histological reporting in MAFLD/MASH. Defined standards for application and validation of AI models are important priorities for their implementation. There is consensus among clinical experts in the Asia Pacific that AI-assisted histological assessment is useful in MAFLD/MASH interpretation. However, there remain important challenges to the adoption of these technologies into routine clinical workflows.

Tissue detection is a crucial first step in most digital pathology applications. By applying image segmentation algorithms, all tissue is delineated and background discarded from further analyses, improving both computational efficiency and analytical results. Details of the segmentation algorithm are rarely reported, and there is a lack of studies investigating the downstream effects of a poor segmentation algorithm. Disregarding tissue detection quality could jeopardize patient safety if diagnostically relevant parts of the specimen are excluded from analysis in clinical applications. This study aims to determine whether performance of downstream tasks is sensitive to the tissue detection method, and to compare the performance of a classical and an AI-based tissue detection approach. To this end, we trained an AI model for Gleason grading of prostate cancer in whole slide images (WSIs) using two different tissue detection algorithms: thresholding (classical) and UNet++ (AI). A total of 33,823 WSIs scanned on seven digital pathology scanners were used to train the segmentation AI model. The downstream Gleason grading algorithm was trained and tested using 70,524 WSIs from 13 clinical sites scanned on 13 different scanners. On the slides where tissue could be detected by both algorithms, no significant difference in overall Gleason grading performance was observed. There was a decrease from 118 (0.43%) to 24 (0.09%) fully undetected tissue samples when switching from thresholding-based tissue detection to AI-based, suggesting this AI model may be more reliable than the classical model for avoiding total failures on slides with unusual appearance. Moreover, tissue detection dependent clinically significant variations in AI grading were observed in 3.5% of malignant slides, highlighting the role of tissue detection for optimal clinical performance of diagnostic AI.

Accurate pathology diagnosis has remained the most essential factor in providing optimal breast cancer care across the globe. Core needle biopsies account for a significant portion of pathologists' workload, yet diagnostic variability remains especially for several difficult to diagnose entities such as borderline proliferative breast disease, atypical and low-grade breast lesions. Digital pathology and artificial intelligence (AI) are emerging tools that may help address these diagnostic challenges. To evaluate the diagnostic performance of the Pathology Artificial Intelligence Guidance Engine Breast (Paige PaBr), an AI tool, in analyzing digital histology images for breast cancer diagnosis, a retrospective study of 250 digitized breast biopsy images scanned at the University of Florida Jacksonville was conducted. AI-generated diagnoses were compared with reference diagnoses by experienced pathologists. Agreement, sensitivity, and specificity were calculated for suspicious for cancer and invasive lesions, respectively. For detection of suspicious lesions, AI showed 90.0% raw agreement with Cohen's kappa of 0.77 (95%CI: 0.69-0.86), sensitivity of 90.2%, and specificity of 89.5%. For invasive breast cancer specifically, Cohen's kappa was 0.82 (95%CI: 0.73-0.90). Of 76 invasive cases, 74 (97.4%) were correctly identified by AI. Among 98 cases of ductal carcinoma in situ (DCIS) and other proliferative lesions, 84 (85.7%) were classified as non-invasive, indicating high specificity. In summary, AI demonstrated substantial concordance with expert pathologists. These findings suggest that AI may serve as an adjunctive tool to support the evaluation of breast biopsies. Continued validation and prospective assessment are warranted to further define its potential role in routine pathology practice.

Histopathological grading remains the cornerstone of risk stratification in prostate cancer, yet conventional Gleason-based assessment is limited by interobserver variability and by the biological heterogeneity concealed within Gleason pattern 4. This review examines the evolution of prostate cancer grading from the original Gleason system to contemporary Grade Groups and to newer morphology-based frameworks that seek to refine prognostic stratification. Particular attention is given to the distinction between patterns 3 and 4, which remains clinically pivotal but diagnostically challenging, especially in the setting of poorly formed glands. By contrast, cribriform architecture has emerged as one of the most reproducible and prognostically adverse components of pattern 4. Intraductal carcinoma of the prostate (IDC-P), which often overlaps morphologically and biologically with cribriform carcinoma, is similarly associated with aggressive disease and is now addressed within a more unified diagnostic and grading framework following the recent joint GUPS/ISUP recommendations. Outcome-based morphometric studies further suggest that a diameter threshold of approximately 0.25mm can identify large cribriform glands with particularly adverse behavior, although standardization remains incomplete. These observations have contributed to the development of a risk-oriented taxonomy in which adverse architectural features may carry greater prognostic weight than numerical grade alone. Finally, we discuss how digital pathology and artificial intelligence are extending this conceptual shift by improving diagnostic reproducibility, enabling quantitative detection of cribriform morphology and supporting outcome-oriented histology-based risk prediction. Together, these developments suggest that prostate cancer grading is moving from a purely descriptive system toward a more integrated and biologically informed model of risk assessment.

Recent advances in deep learning have significantly improved the accuracy and efficiency of disease classification in digital pathology. Early diagnosis and precise classification of histopathological images are crucial for enabling timely treatment and improving therapeutic outcomes. Oral squamous cell carcinoma (OSCC) is one of the most common malignancies in the oral cavity, with manual histopathological examination serving as the gold standard for diagnosis-though it is time-consuming and subject to observer variability. This study investigates the performance of four deep learning convolutional neural network (CNN) models-ResNet50 (residual blocks), DenseNet201 (dense connectivity), EfficientNetB0 (compound scaling), and ConvNeXt_Tiny (transformer-based convolutions)-for binary classification (benign vs. carcinoma) of 10,000 histopathological images. Among the models, EfficientNetB0 achieved the highest accuracy of 97.6% and an ROC-AUC score of 0.9963, demonstrating superior generalization and discriminative power. ConvNeXt_Tiny followed with an accuracy of 95.92%, DenseNet201 with 86.08%, and ResNet50 with the lowest accuracy of 71.52%. The comparative analysis underscores the advantages of modern CNN architectures over traditional residual networks, supporting the integration of deep learning models into diagnostic frameworks for improved detection of oral squamous cell carcinoma.

Tubulointerstitial Diseases: An Updated Framework for Diverse and Emerging Entities.

Assess whether a single instance-segmentation model can operate robustly across multiple cytological stains, avoiding stain-specific pipelines without sacrificing accuracy. We consolidated three expert-annotated datasets (Papanicolaou, Feulgen, AgNOR), standardised their formats and trained Mask R-CNN, Mask2Former, YOLO11n and YOLO12n under two regimes: per-stain and unified multi-stain. We reported AP50 and AP75 on validation/test to decouple detection tolerance from boundary precision, using identical training and evaluation protocols across models. Feulgen specialist models yielded the highest scores in our benchmark, with YOLO11n reaching AP75 0.566 and AP50 0.643. Mask2Former led in geometric precision (AP75) in complex scenarios (AgNOR and Papanicolaou), achieving the best performance on the combined dataset. Crucially, unified training matched or exceeded stain-specific models, maintaining detection (AP50) while improving boundary definition (AP75). A single multi-stain model is feasible for cytology instance segmentation, preserving detection and improving boundary accuracy in diverse stains. Transformer-based Mask2Former showed the best AP75 on the combined dataset, supporting a unified multi-stain, scalable path to AI-assisted cytopathology and practical screening workflows. Can a single computer vision model handle the diversity of cytological stains found in clinical routine? This study validates a unified deep learning approach that effectively segments cells across Papanicolaou, Feulgen and AgNOR protocols. Results suggest that laboratories can deploy simplified, stain-invariant screening tools without compromising diagnostic precision, overcoming a major barrierin digital pathology interoperability.