Introduction The diagnosis of glomerular diseases is primarily based on visual assessment of histologic patterns. Semi-quantitative scoring of active and chronic lesions is often required to assess individual characteristics of the disease. Reproducibility of the visual scoring systems remains debatable, while digital and machine-learning technologies present opportunities to detect, classify and quantify glomerular lesions, also considering their inter- and intraglomerular heterogeneity. Materials and methods: We performed a cross-validated comparison of three modifications of a convolutional neural network (CNN)-based approach for recognition and intraglomerular quantification of nine main glomerular patterns of injury. Reference values provided by two nephropathologists were used for validation. For each glomerular image, visual attention heatmaps were generated with a probability of class attribution for further intraglomerular quantification. The quality of classifier-produced heatmaps was evaluated by intersection over union metrics (IoU) between predicted and ground truth localization heatmaps. Results: A proposed spatially guided modification of the CNN classifier achieved the highest glomerular pattern classification accuracies, with area under curve (AUC) values up to 0.981. With regards to heatmap overlap area and intraglomerular pattern quantification, the spatially guided classifier achieved a significantly higher generalized mean IoU value compared to single-multiclass and multiple-binary classifiers. Conclusions: We propose a spatially guided CNN classifier that in our experiments reveals the potential to achieve high accuracy for the localization of intraglomerular patterns.

The limited reproducibility of the grading of non-muscle invasive papillary urothelial carcinoma (NMIPUC) necessitates the search for more robust image-based predictive factors. In a cohort of 157 NMIPUC patients treated with Bacille Calmette–Guérin (BCG) immunotherapy, we explored the multiple instance learning (MIL)-based classification approach for the prediction of 2-year and 5-year relapse-free survival and the multiple instance survival learning (MISL) framework for survival regression. We used features extracted from image patches sampled from whole slide images of hematoxylin–eosin-stained transurethral resection (TUR) NPMIPUC specimens and tested several patch sampling and feature extraction network variations to optimize the model performance. We selected the model showing the best patient survival stratification for further testing in the context of clinical and pathological variables. MISL with the multiresolution patch sampling technique achieved the best patient risk stratification (concordance index = 0.574, p = 0.010), followed by a 2-year MIL classification. The best-selected model revealed an independent prognostic value in the context of other clinical and pathologic variables (tumor stage, grade, and presence of tumor on the repeated TUR) with statistically significant patient risk stratification. Our findings suggest that MISL-based predictions can improve NMIPUC patient risk stratification, while validation studies are needed to test the generalizability of our models.

Non-muscle-invasive papillary urothelial carcinoma (NMIPUC) of the urinary bladder is the most common type of bladder cancer. Intravesical Bacille Calmette–Guerin (BCG) immunotherapy is applied in patients with a high risk of recurrence and progression of NMIPUC to muscle-invasive disease. However, the tumor relapses in about 30% of patients despite the treatment, raising the need for better risk stratification. We explored the potential of spatial distributions of immune cell subtypes (CD20, CD11c, CD163, ICOS, and CD8) within the tumor microenvironment to predict NMIPUC recurrence following BCG immunotherapy. Based on analyses of digital whole-slide images, we assessed the densities of the immune cells in the epithelial–stromal interface zone compartments and their distribution, represented by an epithelial–stromal interface density ratio (IDR). While the densities of any cell type did not predict recurrence, a higher IDR of CD11c (HR: 0.0012, p-value = 0.0002), CD8 (HR: 0.0379, p-value = 0.005), and ICOS (HR: 0.0768, p-value = 0.0388) was associated with longer recurrence-free survival (RFS) based on the univariate Cox regression. The history of positive repeated TUR (re-TUR) (HR: 4.93, p-value = 0.0001) and T1 tumor stage (HR: 2.04, p-value = 0.0159) were associated with shorter RFS, while G3 tumor grade according to the 1973 WHO classification showed borderline significance (HR: 1.83, p-value = 0.0522). In a multivariate analysis, the two models with a concordance index exceeding 0.7 included the CD11c IDR in combination with either a history of positive re-TUR or tumor stage. We conclude that the CD11c IDR is the most informative predictor of NMIPUC recurrence after BCG immunotherapy. Our findings highlight the importance of assessment of the spatial distribution of immune cells in the tumor microenvironment.

Multidisciplinary teams (MDTs), the cornerstone of modern cancer care, are facing significant operational inefficiencies. These challenges include the laborious, manual synthesis of unstructured, multimodal patient data for case preparation, which is time-consuming and prone to information overload. Furthermore, heterogeneous operational processes across MDTs themselves compound these issues. Additionally, clinical decisions are often archived in static documents, preventing the systematic collection of decision rationales essential for continuous learning and research. We propose that artificial intelligence (AI), particularly natural language processing (NLP) and large language models (LLMs), can act as integrated partners to solve these problems. The capacity of AI to seamlessly integrate diverse datasets—including imaging, histopathology, genomics, and clinical data—may be instrumental in enhancing diagnostic accuracy, refining personalized treatment plans within a complex cancer management journey. This integration can be achieved through a tiered approach, utilizing models from small NLP for targeted information extraction to foundational generative NLP for complex evidence synthesis, while addressing key challenges in validation, ethical governance, and regulatory oversight. International initiatives are actively developing validated frameworks to facilitate the widespread and standardized adoption of these AI solutions, while taking into account heterogeneous operational processes. By improving data management, streamlining decision making, and establishing crucial feedback loops, AI integration promises to enhance patient outcomes and optimize resource utilization within cancer care.

Despite advances in diagnostic and treatment technologies, predicting outcomes of patients with hepatocellular carcinoma (HCC) remains a challenge. Prognostic models are further obscured by the variable impact of the tumor properties and the remaining liver parenchyma, often affected by cirrhosis or non-alcoholic fatty liver disease that tend to precede HCC. This study investigated the prognostic value of reticulin and collagen microarchitecture in liver resection samples. We analyzed 105 scanned tissue sections that were stained using a Gordon and Sweet’s silver impregnation protocol combined with Picric Acid–Sirius Red. A convolutional neural network was utilized to segment the red-staining collagen and black linear reticulin strands, generating a detailed map of the fiber structure within the HCC and adjacent liver tissue. Subsequent hexagonal grid subsampling coupled with automated epithelial edge detection and computational fiber morphometry provided the foundation for region-specific tissue analysis. Two penalized Cox regression models using LASSO achieved a concordance index (C-index) greater than 0.7. These models incorporated variables such as patient age, tumor multifocality, and fiber-derived features from the epithelial edge in both the tumor and liver compartments. The prognostic value at the tumor edge was derived from the reticulin structure, while collagen characteristics were significant at the epithelial edge of peritumoral liver. The prognostic performance of these models was superior to models solely reliant on conventional clinicopathologic parameters, highlighting the utility of AI-extracted microarchitectural features for the management of HCC.