Find the papers that actually matter

Developing Foundation Models for medical image analysis is essential to overcome the unique challenges of radiological tasks. The first challenges of this kind for 3D brain MRI, SSL3D and FOMO25, were held at MICCAI 2025. Our solution ranked first in tracks of both contests. It relies on a U-Net CNN architecture combined with strategies leveraging anatomical priors and neuroimaging domain knowledge. Notably, our models trained 1-2 orders of magnitude faster and were 10 times smaller than competing transformer-based approaches. Models are available here: https://github.com/jbanusco/BrainFM4Challenges.

Explainable Artificial Intelligence (XAI) techniques, such as Gradient-weighted Class Activation Mapping (Grad-CAM), have become indispensable for visualizing the reasoning process of deep neural networks in medical image analysis. Despite their popularity, the faithfulness and reliability of these heatmap-based explanations remain under scrutiny. This study critically investigates whether Grad-CAM truly represents the internal decision-making of deep models trained for lung cancer image classification. Using the publicly available IQ-OTH/NCCD dataset, we evaluate five representative architectures: ResNet-50, ResNet-101, DenseNet-161, EfficientNet-B0, and ViT-Base-Patch16-224, to explore model-dependent variations in Grad-CAM interpretability. We introduce a quantitative evaluation framework that combines localization accuracy, perturbation-based faithfulness, and explanation consistency to assess Grad-CAM reliability across architectures. Experimental findings reveal that while Grad-CAM effectively highlights salient tumor regions in most convolutional networks, its interpretive fidelity significantly degrades for Vision Transformer models due to non-local attention behavior. Furthermore, cross-model comparisons indicate substantial variability in saliency localization, implying that Grad-CAM explanations may not always correspond to the true diagnostic evidence used by the networks. This work exposes critical limitations of current saliency-based XAI approaches in medical imaging and emphasizes the need for model-aware interpretability methods that are both computationally sound and clinically meaningful. Our findings aim to inspire a more cautious and rigorous adoption of visual explanation tools in medical AI, urging the community to rethink what it truly means to "trust" a model's explanation.

Total-body PET/CT enables system-wide molecular imaging, but heterogeneous anatomical and metabolic signals, approximately 2 m axial coverage, and structured radiology semantics challenge existing medical AI models that assume single-modality inputs, localized fields of view, and coarse image-text alignment. We introduce SDF-HOLO (Systemic Dual-stream Fusion Holo Model), a multimodal foundation model for holistic total-body PET/CT, pre-trained on more than 10,000 patients. SDF-HOLO decouples CT and PET representation learning with dual-stream encoders and couples them through a cross-modal interaction module, allowing anatomical context to refine PET aggregation while metabolic saliency guides subtle morphological reasoning. To model long-range dependencies across the body, hierarchical context modeling combines efficient local windows with global attention. To bridge voxels and clinical language, we use anatomical segmentation masks as explicit semantic anchors and perform voxel-mask-text alignment during pre-training. Across tumor segmentation, low-dose lesion detection, and multilingual diagnostic report generation, SDF-HOLO outperforms strong task-specific and clinical-reference baselines while reducing localization errors and hallucinated findings. Beyond focal interpretation, the model enables system-wide metabolic profiling and reveals tumor-associated fingerprints of inter-organ metabolic network interactions, providing a scalable computational foundation for total-body PET/CT diagnostics and system-level precision oncology.

Efficient brain tumor diagnosis is crucial for early treatment; however, it is challenging because of lesion variability and image complexity. We evaluated convolutional neural networks (CNNs) in a federated learning (FL) setting, comparing models trained on original versus preprocessed MRI images (resizing, grayscale conversion, normalization, filtering, and histogram equalization). Preprocessing alone yielded negligible gains; combined with test-time augmentation (TTA), it delivered consistent, statistically significant improvements in federated MRI classification (p<0.001). In practice, TTA should be the default inference strategy in FL-based medical imaging; when the computational budget permits, pairing TTA with light preprocessing provides additional reliable gains.

Deep learning for cancer histopathology training conflicts with privacy constraints in clinical settings. Federated Learning (FL) mitigates this by keeping data local; however, its performance depends on hyperparameter choices under non-independent and identically distributed (non-IID) client datasets. This paper examined whether hyperparameters optimized on one cancer imaging dataset generalized across non-IID federated scenarios. We considered binary histopathology tasks for ovarian and colorectal cancers. We perform centralized Bayesian hyperparameter optimization and transfer dataset-specific optima to the non-IID FL setup. The main contribution of this study is the introduction of a simple cross-dataset aggregation heuristic by combining configurations by averaging the learning rates and considering the modal optimizers and batch sizes. This combined configuration achieves a competitive classification performance.

Procedural case logs are a core requirement in radiology training, yet they are time-consuming to complete and prone to inconsistency when authored manually. This study investigates whether large language models (LLMs) can automate procedural case log documentation directly from free-text radiology reports. We evaluate multiple local and commercial LLMs under instruction-based and chain-of-thought prompting to extract structured procedural information from 414 curated interventional radiology reports authored by nine residents between 2018 and 2024. Model performance is assessed using sensitivity, specificity, and F1-score, alongside inference latency and token efficiency to estimate operational cost. Results show that both local and commercial models achieve strong extraction performance, with best F1-scores approaching 0.87, while exhibiting different trade-offs between speed and cost. Automation using LLMs has the potential to substantially reduce clerical burden for trainees and improve consistency in case logging. These findings demonstrate the feasibility of AI-assisted documentation in medical education and highlight the need for further validation across institutions and clinical workflows.

Medical Vision-language models (VLMs) have shown remarkable performances in various medical imaging domains such as histo\-pathology by leveraging pre-trained, contrastive models that exploit visual and textual information. However, histopathology images may exhibit severe domain shifts, such as staining, contamination, blurring, and noise, which may severely degrade the VLM's downstream performance. In this work, we introduce Histopath-C, a new benchmark with realistic synthetic corruptions designed to mimic real-world distribution shifts observed in digital histopathology. Our framework dynamically applies corruptions to any available dataset and evaluates Test-Time Adaptation (TTA) mechanisms on the fly. We then propose LATTE, a transductive, low-rank adaptation strategy that exploits multiple text templates, mitigating the sensitivity of histopathology VLMs to diverse text inputs. Our approach outperforms state-of-the-art TTA methods originally designed for natural images across a breadth of histopathology datasets, demonstrating the effectiveness of our proposed design for robust adaptation in histopathology images. Code and data are available at https://github.com/Mehrdad-Noori/Histopath-C.

Triple-negative breast cancer (TNBC) is an aggressive disease with high mortality and limited treatment options, due to its lack of receptors that have targeted therapies available. The tumor microenvironment (TME) plays a critical role in TNBC progression and therapeutic resistance. In this work, we developed a novel mathematical model to describe key cellular interactions within the TNBC TME, informed by current literature and expert input. Our model consists of a system of ordinary differential equations representing five interacting cell populations: M2 macrophages, cancer-associated fibroblasts, TNBC tumor cells, cytotoxic T lymphocytes, and regulatory T cells. We performed global sensitivity analysis to determine which model parameters most strongly influence tumor burden over a clinically-relevant treatment timeframe. The pathways associated with the most-influential parameters correspond to biological mechanisms that are consistent with known and emerging therapeutic strategies in TNBC, including stromal-mediated tumor support. These results highlight key regulatory interactions within the TNBC TME and provide a quantitative framework for hypothesis generation and future investigation of combination treatment strategies.

Breast cancer is one of the most common cancers among women worldwide, and its accurate and timely diagnosis plays a critical role in improving treatment outcomes. This thesis presents an innovative framework for detecting malignant masses in mammographic images by integrating the Pyramid Adaptive Atrous Convolution (PAAC) and Transformer architectures. The proposed approach utilizes Multi-Scale Feature Fusion to enhance the extraction of features from benign and malignant tissues and combines Dice Loss and Focal Loss functions to improve the model's learning process, effectively reducing errors in binary breast cancer classification and achieving high accuracy and efficiency. In this study, a comprehensive dataset of breast cancer images from INbreast, MIAS, and DDSM was preprocessed through data augmentation and contrast enhancement and resized to 227x227 pixels for model training. Leveraging the Transformer's ability to manage long-range dependencies with Self-Attention mechanisms, the proposed model achieved high accuracy in detecting cancerous masses, outperforming foundational models such as BreastNet, DeepMammo, Multi-Scale CNN, Swin-Unet, and SegFormer. The final evaluation results for the proposed model include an accuracy of 98.5\%, sensitivity of 97.8\%, specificity of 96.3\%, F1-score of 98.2\%, and overall precision of 97.9\%. These metrics demonstrate a significant improvement over traditional methods and confirm the model's effectiveness in identifying cancerous masses in complex scenarios and large datasets. This model shows potential as a reliable and efficient tool for breast cancer diagnosis and can be effectively integrated into medical diagnostic systems.

Digital pathology plays a vital role across modern medicine, offering critical insights for disease diagnosis, prognosis, and treatment. However, histopathology images often contain artifacts introduced during slide preparation and digitization. Detecting and excluding them is essential to ensure reliable downstream analysis. Traditional supervised models typically require large annotated datasets, which is resource-intensive and not generalizable to novel artifact types. To address this, we propose DiffusionQC, which detects artifacts as outliers among clean images using a diffusion model. It requires only a set of clean images for training rather than pixel-level artifact annotations and predefined artifact types. Furthermore, we introduce a contrastive learning module to explicitly enlarge the distribution separation between artifact and clean images, yielding an enhanced version of our method. Empirical results demonstrate superior performance to state-of-the-art and offer cross-stain generalization capacity, with significantly less data and annotations.