The 8th Banff Conference on Allograft Pathology was held in Edmonton, Canada, 15–21 July 2005. Major outcomes included the elimination of the non‐specific term ‘chronic allograft nephropathy’ (CAN) from the Banff classification for kidney allograft pathology, and the recognition of the entity of chronic antibody‐mediated rejection. Participation of B cells in allograft rejection and genomics markers of rejection were also major subjects addressed by the conference.
Accurate localization of tumor regions from hematoxylin and eosin-stained whole-slide images is fundamental for translational research including spatial analysis, molecular profiling, and tissue architecture investigation. However, deep learning-based tumor detection trained within specific cancers may exhibit reduced robustness when applied across different tumor types. We investigated whether balanced training across cancers at modest scale can achieve high performance and generalize to unseen tumor types. A multi-cancer tumor localization model (MuCTaL) was trained on 79,984 non-overlapping tiles from four cancers (melanoma, hepatocellular carcinoma, colorectal cancer, and non-small cell lung cancer) using transfer learning with DenseNet169. The model achieved a tile-level ROC-AUC of 0.97 in validation data from the four training cancers, and 0.71 on an independent pancreatic ductal adenocarcinoma cohort. A scalable inference workflow was built to generate spatial tumor probability heatmaps compatible with existing digital pathology tools. Code and models are publicly available at https://github.com/AivaraX-AI/MuCTaL.
Pathology foundation models (PFMs) have demonstrated strong representational capabilities through self-supervised pre-training on large-scale, unannotated histopathology image datasets. However, their diverse yet opaque pretraining contexts, shaped by both data-related and structural/training factors, introduce latent biases that hinder generalisability and transparency in downstream applications. In this paper, we propose AdaFusion, a novel prompt-guided inference framework that, to our knowledge, is among the very first to dynamically integrate complementary knowledge from multiple PFMs. Our method compresses and aligns tile-level features from diverse models and employs a lightweight attention mechanism to adaptively fuse them based on tissue phenotype context. We evaluate AdaFusion on three real-world benchmarks spanning treatment response prediction, tumour grading, and spatial gene expression inference. Our approach consistently surpasses individual PFMs across both classification and regression tasks, while offering interpretable insights into each model's biosemantic specialisation. These results highlight AdaFusion's ability to bridge heterogeneous PFMs, achieving both enhanced performance and interpretability of model-specific inductive biases.
Spatial transcriptomics (ST) is a novel technique that simultaneously captures pathological images and gene expression profiling with spatial coordinates. Since ST is closely related to pathological features such as disease subtypes, it may be valuable to augment image representation with pathological information. However, there are no attempts to leverage ST for image recognition ({\it i.e,} patch-level classification of subtypes of pathological image.). One of the big challenges is significant batch effects in spatial transcriptomics that make it difficult to extract pathological features of images from ST. In this paper, we propose a batch-agnostic contrastive learning framework that can extract consistent signals from gene expression of ST in multiple patients. To extract consistent signals from ST, we utilize the batch-agnostic gene encoder that is trained in a variational inference manner. Experiments demonstrated the effectiveness of our framework on a publicly available dataset. Code is publicly available at https://github.com/naivete5656/TPIRBAE
As recent text-conditioned diffusion models have enabled the generation of high-quality images, concerns over their potential misuse have also grown. This issue is critical in the medical domain, where text-conditioned generated medical images could enable insurance fraud or falsified records, highlighting the urgent need for reliable safeguards against unethical use. While watermarking techniques have emerged as a promising solution in general image domains, their direct application to medical imaging presents significant challenges. A key challenge is preserving fine-grained disease manifestations, as even minor distortions from a watermark may lead to clinical misinterpretation, which compromises diagnostic integrity. To overcome this gap, we present MedSign, a deep learning-based watermarking framework specifically designed for text-to-medical image synthesis, which preserves pathologically significant regions by adaptively adjusting watermark strength. Specifically, we generate a pathology localization map using cross-attention between medical text tokens and the diffusion denoising network, aggregating token-wise attention across layers, heads, and time steps. Leveraging this map, we optimize the LDM decoder to incorporate watermarking during image synthesis, ensuring cohesive integration while minimizing interference in diagnostically critical regions. Experimental results show that our MedSign preserves diagnostic integrity while ensuring watermark robustness, achieving state-of-the-art performance in image quality and detection accuracy on MIMIC-CXR and OIA-ODIR datasets.
Accurate and efficient access to laboratory protocols is essential in Anatomical Pathology (AP), where up to 70% of medical decisions depend on laboratory diagnoses. However, static documentation such as printed manuals or PDFs is often outdated, fragmented, and difficult to search, creating risks of workflow errors and diagnostic delays. This study proposes and evaluates a Retrieval-Augmented Generation (RAG) assistant tailored to AP laboratories, designed to provide technicians with context-grounded answers to protocol-related queries. We curated a novel corpus of 99 AP protocols from a Portuguese healthcare institution and constructed 323 question-answer pairs for systematic evaluation. Ten experiments were conducted, varying chunking strategies, retrieval methods, and embedding models. Performance was assessed using the RAGAS framework (faithfulness, answer relevance, context recall) alongside top-k retrieval metrics. Results show that recursive chunking and hybrid retrieval delivered the strongest baseline performance. Incorporating a biomedical-specific embedding model (MedEmbed) further improved answer relevance (0.74), faithfulness (0.70), and context recall (0.77), showing the importance of domain-specialised embeddings. Top-k analysis revealed that retrieving a single top-ranked chunk (k=1) maximized efficiency and accuracy, reflecting the modular structure of AP protocols. These findings highlight critical design considerations for deploying RAG systems in healthcare and demonstrate their potential to transform static documentation into dynamic, reliable knowledge assistants, thus improving laboratory workflow efficiency and supporting patient safety.
Despite the impressive performance across a wide range of applications, current computational pathology models face significant diagnostic efficiency challenges due to their reliance on high-magnification whole-slide image analysis. This limitation severely compromises their clinical utility, especially in time-sensitive diagnostic scenarios and situations requiring efficient data transfer. To address these issues, we present a novel computation- and communication-efficient framework called Magnification-Aligned Global-Local Transformer (MAG-GLTrans). Our approach significantly reduces computational time, file transfer requirements, and storage overhead by enabling effective analysis using low-magnification inputs rather than high-magnification ones. The key innovation lies in our proposed magnification alignment (MAG) mechanism, which employs self-supervised learning to bridge the information gap between low and high magnification levels by effectively aligning their feature representations. Through extensive evaluation across various fundamental CPath tasks, MAG-GLTrans demonstrates state-of-the-art classification performance while achieving remarkable efficiency gains: up to 10.7 times reduction in computational time and over 20 times reduction in file transfer and storage requirements. Furthermore, we highlight the versatility of our MAG framework through two significant extensions: (1) its applicability as a feature extractor to enhance the efficiency of any CPath architecture, and (2) its compatibility with existing foundation models and histopathology-specific encoders, enabling them to process low-magnification inputs with minimal information loss. These advancements position MAG-GLTrans as a particularly promising solution for time-sensitive applications, especially in the context of intraoperative frozen section diagnosis where both accuracy and efficiency are paramount.
Mecit Kantarcı, Volkan Kızılgöz, Ramazan Terzi
et al.
PURPOSE: This study aimed to evaluate the effectiveness of artificial intelligence (AI) in diagnosing focal nodular hyperplasia (FNH) of the liver using magnetic resonance imaging (MRI) and compare its performance with that of radiologists.
METHODS: In the first phase of the study, the MRIs of 60 patients (30 patients with FNH and 30 patients with no lesions or lesions other than FNH) were processed using a segmentation program and introduced to an AI model. After the learning process, the MRIs of 42 different patients that the AI model had no experience with were introduced to the system. In addition, a radiology resident and a radiology specialist evaluated patients with the same MR sequences. The sensitivity and specificity values were obtained from all three reviews.
RESULTS: The sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of the AI model were found to be 0.769, 0.966, 0.909, and 0.903, respectively. The sensitivity and specificity values were higher than those of the radiology resident and lower than those of the radiology specialist. The results of the specialist versus the AI model revealed a good agreement level, with a kappa (κ) value of 0.777.
CONCLUSION: For the diagnosis of FNH, the sensitivity, specificity, PPV, and NPV of the AI device were higher than those of the radiology resident and lower than those of the radiology specialist. With additional studies focused on different specific lesions of the liver, AI models are expected to be able to diagnose each liver lesion with high accuracy in the future.
CLINICAL SIGNIFICANCE: AI is studied to provide assisted or automated interpretation of radiological images with an accurate and reproducible imaging diagnosis.
Medical physics. Medical radiology. Nuclear medicine
Hazal Tunç Erdoğan, Özge Bülbül, Mustafa Fuat Açıkalın
et al.
Phosphaturic mesenchymal tumor (PMT) is a rare mesenchymal neoplasia usually located in the soft tissue and the bone. It is seen in older ages and is most commonly localized in the extremities. Here, we present a rare case of PMT located in the sinonasal region. A 56-year-old male patient was admitted with complaints of congestion in the right nasal cavity and limitation of upward gaze in the right eye. Computed tomography revealed a contrast-enhancing mass with heterogeneous density obliterating the bilateral frontal sinus, the frontoethmoidal recess, the right osteomeatal complex and the right sphenoid sinus, extending to the superior extraconal area in the right orbit. Since the tumor type cannot be determined precisely in the pathological evaluation of incisional biopsy, an excisional biopsy was performed with the preliminary diagnosis of malignancy. But histopathological examination revealed a PMT. PMT is a highly uncommon neoplasm that remains largely unfamiliar to clinicians, surgeons, and pathologists, particularly when arising in rare locations like the sinonasal region. Its histomorphological characteristics can overlap with various other entities, necessitating a broad differential diagnosis.
Abstract Objectives Cervical cancer is one of the most frequently diagnosed cancers and a leading cause of cancer-related deaths in women in low- and middle-income countries (LMICs), accounting for nearly 85% of the global cervical cancer burden. High-risk human papillomavirus (hrHPV) infection is the main cause of cervical cancer. Easy-to-use, rapid, scalable, high-throughput, and cost-effective HPV tests are urgently needed for low-resource settings. Atila Biosystems’ clinically validated ScreenFire HPV Risk Stratification (RS) assay identifies 13 hrHPV in 4 groups based on their oncogenic risk (i.e., HPV16, HPV18/45, HPV31/33/35/52/58, and HPV51/59/39/56/68). While the current standard format is subject to laboratory contamination Atila has developed an innovative, contamination-preventive Zebra BioDome format. Recently we published the analytical performance of ScreenFire RS Zebra BioDome on the BioRad CFX-96 real-time PCR instrument. This current study evaluated its analytical performance on three additional qPCR platforms: Atila Portable iAMP-PS96, Atila Powergene9600 Plus, and Thermo Fisher Quantstudio-7. Methods We tested 173 DNA samples from Nigerian women with cervical cancer. These samples were tested simultaneously using the ScreenFire HPV Zebra BioDome assay (M5FHPV-96) on four different real-time PCR machines (Atila portable iAMP-PS96, Atila Powergene9600 Plus, Thermo Fisher QuantStudio-7, and BioRad CFX-96). We used overall agreement rate and unweighted kappa values to compare different platforms. Results The overall agreement for detection of hrHPV using Atila portable iAMP-PS96 was 96.5% with kappa value 0.95 (95% confidence interval: 0.91–0.99) compared to Thermo Fisher QuantStudio-7, and 97.1% with kappa value 0.96 (95% confidence interval: 0.92–0.99) compared to BioRad CFX-96. For genotype HPV16 and risk stratification (RS) genotype groups (HPV18/45, HPV31/33/35/52/58, and HPV51/59/39/56/68) agreement rates were all > 98.3%. For Atila Powergene9600 Plus the overall agreement was 98.8% with a kappa value of 0.98 (95% confidence interval: 0.96–1.0) compared to Thermo Fisher QuantStudio-7, and 96.5% with a kappa value of 0.96 (95% confidence interval: 0.94–0.99) compared to BioRad CFX-96. The agreements for the HPV16 and RS genotype groups (HPV18/45, HPV31/33/35/52/58, and HPV39/51/56/59/68) were at least 98.3%. Conclusion The novel ScreenFire HPV Zebra BioDome format produced highly concordant hrHPV positivity and RS genotype results on all four qPCR platforms. The data suggests that this innovative technology has the potential to improve HPV testing uptake in low-resource settings without further investment in purchasing new equipment.
Neoplasms. Tumors. Oncology. Including cancer and carcinogens, Infectious and parasitic diseases
This paper addresses the problem of risk prediction on social media data, specifically focusing on the classification of Reddit users as having a pathological gambling disorder. To tackle this problem, this paper focuses on incorporating temporal and emotional features into the model. The preprocessing phase involves dealing with the time irregularity of posts by padding sequences. Two baseline architectures are used for preliminary evaluation: BERT classifier on concatenated posts per user and GRU with LSTM on sequential data. Experimental results demonstrate that the sequential models outperform the concatenation-based model. The results of the experiments conclude that the incorporation of a time decay layer (TD) and passing the emotion classification layer (EmoBERTa) through LSTM improves the performance significantly. Experiments concluded that the addition of a self-attention layer didn't significantly improve the performance of the model, however provided easily interpretable attention scores. The developed architecture with the inclusion of EmoBERTa and TD layers achieved a high F1 score, beating existing benchmarks on pathological gambling dataset. Future work may involve the early prediction of risk factors associated with pathological gambling disorder and testing models on other datasets. Overall, this research highlights the significance of the sequential processing of posts including temporal and emotional features to boost the predictive power, as well as adding an attention layer for interpretability.
Radiology reporting generative AI holds significant potential to alleviate clinical workloads and streamline medical care. However, achieving high clinical accuracy is challenging, as radiological images often feature subtle lesions and intricate structures. Existing systems often fall short, largely due to their reliance on fixed size, patch-level image features and insufficient incorporation of pathological information. This can result in the neglect of such subtle patterns and inconsistent descriptions of crucial pathologies. To address these challenges, we propose an innovative approach that leverages pathology-aware regional prompts to explicitly integrate anatomical and pathological information of various scales, significantly enhancing the precision and clinical relevance of generated reports. We develop an anatomical region detector that extracts features from distinct anatomical areas, coupled with a novel multi-label lesion detector that identifies global pathologies. Our approach emulates the diagnostic process of radiologists, producing clinically accurate reports with comprehensive diagnostic capabilities. Experimental results show that our model outperforms previous state-of-the-art methods on most natural language generation and clinical efficacy metrics, with formal expert evaluations affirming its potential to enhance radiology practice.
Pratibha Kumari, Daniel Reisenbüchler, Lucas Luttner
et al.
In recent years, there has been remarkable progress in the field of digital pathology, driven by the ability to model complex tissue patterns using advanced deep-learning algorithms. However, the robustness of these models is often severely compromised in the presence of data shifts (e.g., different stains, organs, centers, etc.). Alternatively, continual learning (CL) techniques aim to reduce the forgetting of past data when learning new data with distributional shift conditions. Specifically, rehearsal-based CL techniques, which store some past data in a buffer and then replay it with new data, have proven effective in medical image analysis tasks. However, privacy concerns arise as these approaches store past data, prompting the development of our novel Generative Latent Replay-based CL (GLRCL) approach. GLRCL captures the previous distribution through Gaussian Mixture Models instead of storing past samples, which are then utilized to generate features and perform latent replay with new data. We systematically evaluate our proposed framework under different shift conditions in histopathology data, including stain and organ shift. Our approach significantly outperforms popular buffer-free CL approaches and performs similarly to rehearsal-based CL approaches that require large buffers causing serious privacy violations.
Abstract Chronic rhinosinusitis with nasal polyps (CRSwNP) is a chronic inflammatory disease of the nasal mucosa, and epithelial–mesenchymal transition (EMT) is thought to be an essential process in the pathogenesis of CRSwNP. However, the mechanisms of epithelial and fibroblastic changes at the single-cell level are unclear. In this study, we investigated the epithelial cell, fibroblast, and key gene alterations in the development of CRSwNP. We revealed major cell types involved in CRSwNP and nasal mucosal inflammation formation, then mapped epithelial and fibroblast subpopulations. We showed that the apical and glandular epithelial cells and the ADGRB3+ and POSTN+ fibroblasts were the key cell subtypes in the progression of CRSwNP. Pseudotime and cell cycle analysis identified dynamic changes between epithelial cells and fibroblasts during its development. WFDC2 and CCL26 were identified as the key marker genes involved in the development of CRSwNP and were validated by IHC staining, which may provide a potential novel target for future CRSwNP therapy. ScRNA-seq data provided insights into the cellular landscape and the relationship between epithelial cells and fibroblasts in the progression of CRSwNP. WFDC2 and CCL26 were identified as the key genes involved in the development of CRSwNP and may be the potential markers for gene therapy.
Abstract Objective To investigate the sensitivity of laryngeal findings in predicting high‐grade dysplasia/carcinoma in situ (CIS) and squamous cell carcinoma (SCC) in patients with vocal fold leukoplakia. Methods A retrospective review of the medical records and video recordings of the laryngeal examination of patients with vocal fold leukoplakia who underwent un‐sedated office‐based laryngeal biopsy in a tertiary referral center between January 2022 and August 2023 was conducted. Laryngeal findings included the size, surface, projection, and edges of the lesion. Vocal fold leukoplakia was classified according to the WHO as benign, low‐grade dysplasia, high‐grade dysplasia/CIS, and squamous cell carcinoma. Results Seventy patients with 100 vocal fold leukoplakia were included. Size was found to have the highest sensitivity with an AUC of 0.730 (95% CI [0.618–0.842], p = 0.002) followed by surface and projection with AUCs of 0.672 (95% CI [0.548–0.795], p = 0.019) and 0.675 (95% CI [0.546–0.804], p = 0.017), respectively. Furthermore, the odds of diagnosing high‐risk lesions (high‐grade dysplasia/CIS and SCC) were the greatest when the lesion was large and rough (OR = 10.28; 95% CI [3.08–34.36]). Conclusion The morphological features of vocal fold leukoplakia may assist the physician in predicting the risk of malignancy. Large and rough lesions were more likely to harbor high‐grade dysplasia/CIS and SCC compared to small and smooth lesions.
The article aims to analyze the performance of ChatGPT, a large language model developed by OpenAI, in the context of cardiology and vascular pathologies. The study evaluated the accuracy of ChatGPT in answering challenging multiple-choice questions (QCM) using a dataset of 190 questions from the Siamois-QCM platform. The goal was to assess ChatGPT potential as a valuable tool in medical education compared to two well-ranked students of medicine. The results showed that ChatGPT outperformed the students, scoring 175 out of 190 correct answers with a percentage of 92.10\%, while the two students achieved scores of 163 and 159 with percentages of 85.78\% and 82.63\%, respectively. These results showcase how ChatGPT has the potential to be highly effective in the fields of cardiology and vascular pathologies by providing accurate answers to relevant questions.
The segment anything model (SAM) was released as a foundation model for image segmentation. The promptable segmentation model was trained by over 1 billion masks on 11M licensed and privacy-respecting images. The model supports zero-shot image segmentation with various segmentation prompts (e.g., points, boxes, masks). It makes the SAM attractive for medical image analysis, especially for digital pathology where the training data are rare. In this study, we evaluate the zero-shot segmentation performance of SAM model on representative segmentation tasks on whole slide imaging (WSI), including (1) tumor segmentation, (2) non-tumor tissue segmentation, (3) cell nuclei segmentation. Core Results: The results suggest that the zero-shot SAM model achieves remarkable segmentation performance for large connected objects. However, it does not consistently achieve satisfying performance for dense instance object segmentation, even with 20 prompts (clicks/boxes) on each image. We also summarized the identified limitations for digital pathology: (1) image resolution, (2) multiple scales, (3) prompt selection, and (4) model fine-tuning. In the future, the few-shot fine-tuning with images from downstream pathological segmentation tasks might help the model to achieve better performance in dense object segmentation.