The COVID-19 pandemic has been the most catastrophic global health emergency of the [Formula: see text] century, resulting in hundreds of millions of reported cases and five million deaths. Chest X-ray (CXR) images are highly valuable for early detection of lung diseases in monitoring and investigating pulmonary disorders such as COVID-19, pneumonia, and tuberculosis. These CXR images offer crucial features about the lung's health condition and can assist in making accurate diagnoses. Manual interpretation of CXR images is challenging even for expert radiologists due to the overlapping radiological features. Therefore, Artificial Intelligence (AI) based image processing took over the charge in healthcare. But still it is uncertain to trust the prediction results by an AI model. However, this can be resolved by implementing explainable artificial intelligence (XAI) tools that transform a black-box AI into a glass-box model. In this research article, we have proposed a novel XAI-TRANS model with inception based transfer learning addressing the challenge of overlapping features in multiclass classification of CXR images. Also, we proposed an improved U-Net Lung segmentation dedicated to obtaining the radiological features for classification. The proposed approach achieved a maximum precision of 98% and accuracy of 97% in multiclass lung disease classification. By leveraging XAI techniques with the evident improvement of 4.75%, specifically LIME and Grad-CAM, to provide detailed and accurate explanations for the model's prediction.

In the evolving field of healthcare, centralized cloud-based medical image retrieval faces challenges related to security, availability, and adversarial threats. Existing deep learning-based solutions improve retrieval but remain vulnerable to adversarial attacks and quantum threats, necessitating a shift to more secure distributed cloud solutions. This article proposes SFMedIR, a secure and fault tolerant medical image retrieval framework that contains an adversarial attack-resistant federated learning for hashcode generation, utilizing a ConvNeXt-based model to improve accuracy and generalizability. The framework integrates quantum-chaos-based encryption for security, dynamic threshold-based shadow storage for fault tolerance, and a distributed cloud architecture to mitigate single points of failure. Unlike conventional methods, this approach significantly improves security and availability in cloud-based medical image retrieval systems, providing a resilient and efficient solution for healthcare applications. The framework is validated on Brain MRI and Kidney CT datasets, achieving a 60-70% improvement in retrieval accuracy for adversarial queries and an overall 90% retrieval accuracy, outperforming existing models by 5-10%. The results demonstrate superior performance in terms of both security and retrieval efficiency, making this framework a valuable contribution to the future of secure medical image management.

Deep learning has significantly advanced the field of computer vision; however, developing models that generalize effectively across diverse image domains remains a major research challenge. In this study, we introduce DeepFreqNet, a novel deep neural architecture specifically designed for high-performance multi-domain image classification. The innovative aspect of DeepFreqNet lies in its combination of three powerful components: multi-scale feature extraction for capturing patterns at different resolutions, depthwise separable convolutions for enhanced computational efficiency, and residual connections to maintain gradient flow and accelerate convergence. This hybrid design improves the architecture's ability to learn discriminative features and ensures scalability across domains with varying data complexities. Unlike traditional transfer learning models, DeepFreqNet adapts seamlessly to diverse datasets without requiring extensive reconfiguration. Experimental results from nine benchmark datasets, including MRI tumor classification, blood cell classification, and sign language recognition, demonstrate superior performance, achieving classification accuracies between 98.96% and 99.97%. These results highlight the effectiveness and versatility of DeepFreqNet, showcasing a significant improvement over existing state-of-the-art methods and establishing it as a robust solution for real-world image classification challenges.

Multimodal datasets offer valuable support for developing Clinical Decision Support Systems (CDSS), which leverage predictive models to enhance clinicians' decision-making. In this observational study, we present a dataset of suspected Coronary Artery Disease (CAD) patients - called MultiD4CAD - comprising imaging and clinical data. The imaging data obtained from Coronary Computed Tomography Angiography (CCTA) includes epicardial (EAT) and pericoronary (PAT) adipose tissue segmentations. These metabolically active fat tissues play a key role in cardiovascular diseases. In addition, clinical data include a set of biomarkers recognized as CAD risk factors. The validated EAT and PAT segmentations make the dataset suitable for training predictive models based on radiomics and deep learning architectures. The inclusion of CAD disease labels allows for its application in supervised learning algorithms to predict CAD outcomes. MultiD4CAD has revealed important correlations between imaging features, clinical biomarkers, and CAD status. The article concludes by discussing some challenges, such as classification, segmentation, radiomics, and deep training tasks, that can be investigated and validated using the proposed dataset.

Electrocardiogram (ECG) is crucial for synchronizing cardiovascular magnetic resonance imaging (CMRI) acquisition with the cardiac cycle and for continuous heart rate monitoring during prolonged scans. However, conventional electrode-based ECG systems in clinical MRI environments suffer from tedious setup, magnetohydrodynamic (MHD) waveform distortion, skin burn risks, and patient discomfort. This study proposes a contactless ECG measurement method in MRI to address these challenges. We integrated Beat Pilot Tone (BPT)-a contactless, high motion sensitivity, and easily integrable RF motion sensing modality-into CMRI to capture cardiac motion without direct patient contact. A deep neural network was trained to map the BPT-derived cardiac mechanical motion signals to corresponding ECG waveforms. The reconstructed ECG was evaluated against simultaneously acquired ground truth ECG through multiple metrics: Pearson correlation coefficient, relative root mean square error (RRMSE), cardiac trigger timing accuracy, and heart rate estimation error. Additionally, we performed MRI retrospective binning reconstruction using reconstructed ECG reference and evaluated image quality under both standard clinical conditions and challenging scenarios involving arrhythmias and subject motion. To examine scalability of our approach across field strength, the model pretrained on 1.5T data was applied to 3T BPT cardiac acquisitions. In optimal acquisition scenarios, the reconstructed ECG achieved a median Pearson correlation of 89% relative to the ground truth, while cardiac triggering accuracy reached 94%, and heart rate estimation error remained below 1 bpm. The quality of the reconstructed images was comparable to that of ground truth synchronization. The method exhibited a degree of adaptability to irregular heart rate patterns and subject motion, and scaled effectively across MRI systems operating at different field strengths. The proposed contactless ECG measurement method has the potential to streamline CMRI workflows, improve patient safety and comfort, mitigate MHD distortion challenges and find a robust clinical application.

With increasing imaging demands, radiologists face growing workload pressures, often resulting in delays and reduced diagnostic efficiency. Recent advances in artificial intelligence (AI) have introduced tools for automated report generation, particularly in simpler imaging modalities, such as X-rays. However, limited research has assessed AI performance in complex studies such as MRI and CT scans, where report accuracy and clinical interpretation are critical. To evaluate the performance of a semi-automated AI-based reporting platform in generating radiology reports for complex imaging studies, and to compare its accuracy, efficiency, and user confidence with the traditional dictation method. This study involved 100 imaging cases, including MRI knee (n=21), MRI lumbar spine (n=30), CT head (n=23), and CT Abdomen and Pelvis (n=26). Consultant musculoskeletal radiologists reported each case using both traditional dictation and the AI platform. The radiologist first identified and entered the key positive findings, based on which the AI system generated a full draft report. Reporting time was recorded, and both methods were evaluated on accuracy, user confidence, and overall reporting experience (rated on a scale of 1-5). Statistical analysis was conducted using two-tailed t-tests and 95% confidence intervals. AI-generated reports demonstrated significantly improved performance across all parameters. The mean reporting time reduced from 6.1 to 3.43 min (p<0.0001) with AI-assisted report generation. Accuracy improved from 3.81 to 4.65 (p<0.0001), confidence ratings increased from 3.91 to 4.67 (p<0.0001), and overall reporting experience favored using the AI platform for generating radiology reports (mean 4.7 vs. 3.69, p<0.0001). Minor formatting errors and occasional anatomical misinterpretations were observed in AI-generated reports, but could be easily corrected by the radiologist during review. The AI-assisted reporting platform significantly improved efficiency and radiologist confidence without compromising accuracy. Although the tool performs well when provided with key clinical findings, it still requires expert oversight, especially in anatomically complex reporting. These findings support the use of AI as a supportive tool in radiology practice, with a focus on data integrity, consistency, and human validation.

To develop a hybrid fusion model-deep learning radiomics nomograms (DLRN), integrating radiomics and transfer learning for assisting sonographers differentiate benign and malignant parotid gland tumors. This study retrospectively analyzed a total of 328 patients with pathologically confirmed parotid gland tumors from two centers. Radiomics features extracted from ultrasound images were input into eight machine learning classifiers to construct Radiomics (Rad) model. Additionally, images were also input into seven transfer learning networks to construct deep transfer learning (DTL) model. The prediction probabilities from these two models were combined through decision fusion to construct a DLR model. Clinical features were further integrated with the prediction probabilities of the DLR model to develop the DLRN model. The performance of these models was evaluated using receiver operating characteristic curve analysis, calibration curve, decision curve analysis and the Hosmer-Lemeshow test. In the internal and external validation cohorts, compared with Clinic (AUC = 0.891 and 0.734), Rad (AUC = 0.809 and 0.860), DTL (AUC = 0.905 and 0.782) and DLR (AUC = 0.932 and 0.828), the DLRN model demonstrated the greatest discriminative ability (AUC = 0.931 and 0.934), showing the best discriminative power. With the assistance of DLR, the diagnostic accuracy of resident, attending and chief physician increased by 6.6%, 6.5% and 1.2%, respectively. The hybrid fusion model DLRN significantly enhances the diagnostic performance for benign and malignant tumors of the parotid gland. It can effectively assist sonographers in making more accurate diagnoses.

Within the treatment process of modern stomatology, obtaining exquisite preoperative information is the key to accurate intraoperative planning with implementation and prognostic judgment. However, traditional single mode image has obvious shortcomings, such as "monotonous contents" and "unstable measurement accuracy", which could hardly meet the diversified needs of oral patients. Multimodal medical image fusion (MMIF) technique has been introduced into the studies of stomatology in the 1990s, aiming at realizing personalized patients' data analysis through multiple fusion algorithms, which combines the advantages of multimodal medical images while laying a stable foundation for new treatment technologies. Recently artificial intelligence (AI) has significantly increased the precision and efficiency of MMIF's registration: advanced algorithms and networks have confirmed the great compatibility between AI and MMIF. This article systematically reviews the development history of the multimodal image fusion technique and its current application in stomatology, while analyzing technological progresses within the domain combined with the background of AI's rapid development, in order to provide new ideas for achieving new advancements within the field of stomatology.

Anatomical understanding through deep learning is critical for automatic report generation, intra-operative navigation, and organ localization in medical imaging; however, its progress is constrained by the scarcity of expert-labeled data. A promising remedy is to leverage an annotated reference image to guide the interpretation of an unlabeled target. Although recent vision-language models (VLMs) exhibit non-trivial visual reasoning, their reference-based understanding and fine-grained localization remain limited. We introduce RAU, a framework for reference-based anatomical understanding with VLMs. We first show that a VLM learns to identify anatomical regions through relative spatial reasoning between reference and target images, trained on a moderately sized dataset. We validate this capability through visual question answering (VQA) and bounding box prediction. Next, we demonstrate that the VLM-derived spatial cues can be seamlessly integrated with the fine-grained segmentation capability of SAM2, enabling localization and pixel-level segmentation of small anatomical regions, such as vessel segments. Across two in-distribution and two out-of-distribution datasets, RAU consistently outperforms a SAM2 fine-tuning baseline using the same memory setup, yielding more accurate segmentations and more reliable localization. More importantly, its strong generalization ability makes it scalable to out-of-distribution datasets, a property crucial for medical image applications. To the best of our knowledge, RAU is the first to explore the capability of VLMs for reference-based identification, localization, and segmentation of anatomical structures in medical images. Its promising performance highlights the potential of VLM-driven approaches for anatomical understanding in automated clinical workflows.

We present a patch-based 3D nnUNet adaptation for MR to CT and CBCT to CT image translation using the multicenter SynthRAD2025 dataset, covering head and neck (HN), thorax (TH), and abdomen (AB) regions. Our approach leverages two main network configurations: a standard UNet and a residual UNet, both adapted from nnUNet for image synthesis. The Anatomical Feature-Prioritized (AFP) loss was introduced, which compares multilayer features extracted from a compact segmentation network trained on TotalSegmentator labels, enhancing reconstruction of clinically relevant structures. Input volumes were normalized per-case using zscore normalization for MRIs, and clipping plus dataset level zscore normalization for CBCT and CT. Training used 3D patches tailored to each anatomical region without additional data augmentation. Models were trained for 1000 and 1500 epochs, with AFP fine-tuning performed for 500 epochs using a combined L1+AFP objective. During inference, overlapping patches were aggregated via mean averaging with step size of 0.3, and postprocessing included reverse zscore normalization. Both network configurations were applied across all regions, allowing consistent model design while capturing local adaptations through residual learning and AFP loss. Qualitative and quantitative evaluation revealed that residual networks combined with AFP yielded sharper reconstructions and improved anatomical fidelity, particularly for bone structures in MR to CT and lesions in CBCT to CT, while L1only networks achieved slightly better intensity-based metrics. This methodology provides a stable solution for cross modality medical image synthesis, demonstrating the effectiveness of combining the automatic nnUNet pipeline with residual learning and anatomically guided feature losses.