Neoadjuvant therapy plays a pivotal role in breast cancer treatment, particularly for patients aiming to conserve their breast by reducing tumor size pre-surgery. The ultimate goal of this treatment is achieving a pathologic complete response (pCR), which signifies the complete eradication of cancer cells, thereby lowering the likelihood of recurrence. This study introduces a novel predictive approach to identify patients likely to achieve pCR using radiomic features extracted from MR images, enhanced by the InceptionV3 model and cutting-edge validation methodologies. In our study, we gathered data from 255 unique Patient IDs sourced from the -SPY 2 MRI database with the goal of classifying pCR (pathological complete response). Our research introduced two key areas of novelty.Firstly, we explored the extraction of advanced features from the dcom series such as Area, Perimeter, Entropy, Intensity of the places where the intensity is more than the average intensity of the image. These features provided deeper insights into the characteristics of the MRI data and enhanced the discriminative power of our classification model.Secondly, we applied these extracted features along with combine pixel array of the dcom series of each patient to the numerous deep learning model along with InceptionV3 (GoogleNet) model which provides the best accuracy. To optimize the model's performance, we experimented with different combinations of loss functions, optimizer functions, and activation functions. Lastly, our classification results were subjected to validation using accuracy, AUC, Sensitivity, Specificity and F1 Score. These evaluation metrics provided a robust assessment of the model's performance and ensured the reliability of our findings. The successful combination of advanced feature extraction, utilization of the InceptionV3 model with tailored hyperparameters, and thorough validation using cutting-edge techniques significantly enhanced the accuracy and reliability of our pCR classification study. By adopting a collaborative approach that involved both radiologists and the computer-aided system, we achieved superior predictive performance for pCR, as evidenced by the impressive values obtained for the area under the curve (AUC) at 0.91 having an accuracy of .92. Overall, the combination of advanced feature extraction, leveraging the InceptionV3 model with customized hyperparameters, and rigorous validation using state-of-the-art techniques contributed to the accuracy and credibility of our pCR classification study.

Radiation-induced xerostomia is a common sequela in patients who undergo head and neck radiation therapy. This study aims to develop a three-dimensional deep learning model to predict xerostomia by fusing data from the gross tumor volume primary (GTVp) channel and parotid glands (PGs) channel. Retrospective data were collected from 180 head and neck cancer patients. Xerostomia was defined as xerostomia of grade ≥ 2 occurring in the 6th month of radiation therapy. The dataset was split into 137 cases (58.4% xerostomia, 41.6% non-xerostomia) for training and 43 (55.8% xerostomia, 44.2% non-xerostomia) for testing. XeroNet was composed of GNet, PNet, and a Naive Bayes decision fusion layer. GNet processed data from the GTVp channel (CT, dose distributions corresponding and the GTVp contours). PNet processed data from the PGs channel (CT, dose distributions and the PGs contours). The Naive Bayes decision fusion layer was used to integrate the results from GNet and PNet. Model performance was evaluated using accuracy, F-score, sensitivity, specificity, and area under the receiver operator characteristic curve (AUC). The proposed model achieved promising prediction results. The accuracy, AUC, F-score, sensitivity and specificity were 0.779, 0.858, 0.797, 0.777, and 0.782, respectively. Features extracted from the CT and dose distributions in the GTVp and PGs regions were used to construct machine learning models. However, the performance of these models was inferior to our method. Compared with recent studies on xerostomia prediction, our method also showed better performance. The proposed model could effectively extract features from the GTVp and PGs channels, achieving good performance in xerostomia prediction.

Recent prospective studies have shown that AI may be integrated in double-reader settings to increase cancer detection. The ScreenTrustCAD study was conducted at the breast radiology department at the Capio S:t Göran Hospital where AI is now implemented in clinical practice. This study reports on how the hospital prepared by exploring risks from an enterprise risk management perspective, i.e., applying a holistic and proactive perspective, and developed risk mitigation actions. The study was conducted as an integral part of the preparations before implementing AI in a breast imaging department. Collaborative ideation sessions were conducted with personnel at the hospital, either directly or indirectly involved with AI, to identify risks. Two external experts with competencies in cybersecurity, machine learning, and the ethical aspects of AI, were interviewed as a complement. The risks identified were analyzed according to an Enterprise Risk Management framework, adopted for healthcare, that assumes risks to be emerging from eight different domains. Finally, appropriate risk mitigation actions were identified and discussed. Twenty-three risks were identified covering seven of eight risk domains, in turn generating 51 suggested risk mitigation actions. Not only does the study indicate the emergence of patient safety risks, but it also shows that there are operational, strategic, financial, human capital, legal, and technological risks. The risks with most suggested mitigation actions were ‘Radiographers unable to answer difficult questions from patients’, ‘Increased risk that patient-reported symptoms are missed by the single radiologist’, ‘Increased pressure on the single reader knowing they are the only radiologist to catch a mistake by AI’, and ‘The performance of the AI algorithm might deteriorate’. Before a clinical integration of AI, hospitals should expand, identify, and address risks beyond immediate patient safety by applying comprehensive and proactive risk management. The online version contains supplementary material available at 10.1186/s12913-025-13176-9.

Brain tumors (BT) can cause fatal outcomes by affecting body functions, making precise early detection via magnetic resonance imaging (MRI) examinations critical. The complex variations found in cells of BT may pose challenges in identifying the type of tumor and selecting the most suitable treatment strategy, potentially resulting in different assessments by doctors. As a result, in recent years, AI-powered diagnostic systems have been created to accurately and efficiently identify different types of BT using MRI images. Notably, state-of-the-art deep learning architectures, which have demonstrated efficacy in diverse domains, are now being employed effectively for classifying of brain MRI images. This research presents a hybrid model that integrates spatial attention mechanism (SAM) with ConvNeXt to classify three types of BT: meningioma, pituitary, and glioma. The hybrid model integrates ConvNeXt to enhance the receptive field, capturing information from a broader spatial context, crucial for recognizing tumor patterns spanning multiple pixels. SAM is applied after ConvNeXt, enabling the network to selectively focus on informative regions, thereby improving the model's ability to distinguish BT types and capture complex spatial relationships. Tested on BSF and Figshare datasets, the proposed model achieves a remarkable accuracy of 99.39% and 98.86%, respectively, outperforming the results of recent studies by achieving these results in fewer training periods. This hybrid model marks a major step forward in the automatic classification of BT, demonstrating superior performance in accuracy with efficient training.

Early-stage Brain tumor detection is critical for timely diagnosis and effective treatment. We propose a hybrid deep learning method, Convolutional Neural Network (CNN) integrated with YOLO (You Only Look once) and SAM (Segment Anything Model) for diagnosing tumors. A novel hybrid deep learning framework combining a CNN with YOLOv11 for real-time object detection and the SAM for precise segmentation. Enhancing the CNN backbone with deeper convolutional layers to enable robust feature extraction, while YOLOv11 localizes tumor regions, SAM is used to refine the tumor boundaries through detailed mask generation. A dataset of 896 MRI brain images is used for training, testing, and validating the model, including images of both tumors and healthy brains. Additionally, CNN-based YOLO+SAM methods were utilized successfully to segment and diagnose brain tumors. Our suggested model achieves good performance of Precision as 94.2%, Recall as 95.6% and mAP50(B) score as 96.5% demonstrating and highlighting the effectiveness of the proposed approach for early-stage brain tumor diagnosis Conclusion: The validation is demonstrated through a comprehensive ablation study. The robustness of the system makes it more suitable for clinical deployment.

Accurate differential diagnosis of pneumonia remains a challenging task, as different types of pneumonia require distinct treatment strategies. Early and precise diagnosis is crucial for minimizing the risk of misdiagnosis and for effectively guiding clinical decision-making and monitoring treatment response. This study proposes the WSDC-ViT network to enhance computer-aided pneumonia detection and alleviate the diagnostic workload for radiologists. Unlike existing models such as Swin Transformer or CoAtNet, which primarily improve attention mechanisms through hierarchical designs or convolutional embedding, WSDC-ViT introduces a novel architecture that simultaneously enhances global and local feature extraction through a scalable self-attention mechanism and convolutional refinement. Specifically, the network integrates a scalable self-attention mechanism that decouples the query, key, and value dimensions to reduce computational overhead and improve contextual learning, while an interactive window-based attention module further strengthens long-range dependency modeling. Additionally, a convolution-based module equipped with a dynamic ReLU activation function is embedded within the transformer encoder to capture fine-grained local details and adaptively enhance feature expression. Experimental results demonstrate that the proposed method achieves an average classification accuracy of 95.13% and an F1-score of 95.63% on a chest X-ray dataset, along with 99.36% accuracy and a 99.34% F1-score on a CT dataset. These results highlight the model's superior performance compared to existing automated pneumonia classification approaches, underscoring its potential clinical applicability.

The purpose of this study was to create and validate an ultrasound-based graph convolutional network (US-based GCN) model for the prediction of axillary lymph node metastasis (ALNM) in patients with breast cancer. A total of 820 eligible patients with breast cancer who underwent preoperative breast ultrasonography (US) between April 2016 and June 2022 were retrospectively enrolled. The training cohort consisted of 621 patients, whereas validation cohort 1 included 112 patients, and validation cohort 2 included 87 patients. A US-based GCN model was built using US deep learning features. In validation cohort 1, the US-based GCN model performed satisfactorily, with an AUC of 0.88 and an accuracy of 0.76. In validation cohort 2, the US-based GCN model performed satisfactorily, with an AUC of 0.84 and an accuracy of 0.75. This approach has the potential to help guide optimal ALNM management in breast cancer patients, particularly by preventing overtreatment. In conclusion, we developed a US-based GCN model to assess the ALN status of breast cancer patients prior to surgery. The US-based GCN model can provide a possible noninvasive method for detecting ALNM and aid in clinical decision-making. High-level evidence for clinical use in later studies is anticipated to be obtained through prospective studies.

This study proposes a deep learning-based framework for automated segmentation of brain regions and classification of amyloid positivity using positron emission tomography (PET) images alone, without the need for structural MRI or CT. A 3D U-Net architecture with four layers of depth was trained and validated on a dataset of 200 F18-florbetapir amyloid-PET scans, with an 130/20/50 train/validation/test split. Segmentation performance was evaluated using Dice similarity coefficients across 30 brain regions, with scores ranging from 0.45 to 0.88, demonstrating high anatomical accuracy, particularly in subcortical structures. Quantitative fidelity of PET uptake within clinically relevant regions. Precuneus, prefrontal cortex, gyrus rectus, and lateral temporal cortex was assessed using normalized root mean square error, achieving values as low as 0.0011. Furthermore, the model achieved a classification accuracy of 0.98 for amyloid positivity based on regional uptake quantification, with an area under the ROC curve (AUC) of 0.99. These results highlight the model's potential for integration into PET only diagnostic pipelines, particularly in settings where structural imaging is not available. This approach reduces dependence on coregistration and manual delineation, enabling scalable, reliable, and reproducible analysis in clinical and research applications. Future work will focus on clinical validation and extension to diverse PET tracers including C11 PiB and other F18 labeled compounds.

Safety-critical applications, such as autonomous driving and medical image analysis, require extensive multimodal data for rigorous testing. Synthetic data methods are gaining prominence due to the cost and complexity of gathering real-world data, but they demand a high degree of realism and controllability to be useful. This work introduces two novel methods for synthetic data generation in autonomous driving and medical image analysis, namely MObI and AnydoorMed, respectively. MObI is a first-of-its-kind framework for Multimodal Object Inpainting that leverages a diffusion model to produce realistic and controllable object inpaintings across perceptual modalities, demonstrated simultaneously for camera and lidar. Given a single reference RGB image, MObI enables seamless object insertion into existing multimodal scenes at a specified 3D location, guided by a bounding box, while maintaining semantic consistency and multimodal coherence. Unlike traditional inpainting methods that rely solely on edit masks, this approach uses 3D bounding box conditioning to ensure accurate spatial positioning and realistic scaling. AnydoorMed extends this paradigm to the medical imaging domain, focusing on reference-guided inpainting for mammography scans. It leverages a diffusion-based model to inpaint anomalies with impressive detail preservation, maintaining the reference anomaly's structural integrity while semantically blending it with the surrounding tissue. Together, these methods demonstrate that foundation models for reference-guided inpainting in natural images can be readily adapted to diverse perceptual modalities, paving the way for the next generation of systems capable of constructing highly realistic, controllable and multimodal counterfactual scenarios.

A fundamental challenge in neuroscience is to decode mental states from brain activity. While functional magnetic resonance imaging (fMRI) offers a non-invasive approach to capture brain-wide neural dynamics with high spatial precision, decoding from fMRI data -- particularly from task-evoked activity -- remains challenging due to its high dimensionality, low signal-to-noise ratio, and limited within-subject data. Here, we leverage recent advances in computer vision and propose STDA-SwiFT, a transformer-based model that learns transferable representations from large-scale fMRI datasets via spatial-temporal divided attention and self-supervised contrastive learning. Using pretrained voxel-wise representations from 995 subjects in the Human Connectome Project (HCP), we show that our model substantially improves downstream decoding performance of task-evoked activity across multiple sensory and cognitive domains, even with minimal data preprocessing. We demonstrate performance gains from larger receptor fields afforded by our memory-efficient attention mechanism, as well as the impact of functional relevance in pretraining data when fine-tuning on small samples. Our work showcases transfer learning as a viable approach to harness large-scale datasets to overcome challenges in decoding brain activity from fMRI data.