Although most odontogenic lesions in the jaw are benign, treatment varies widely depending on the nature of the lesion. This study was performed to assess the ability of a machine learning (ML) model using computed tomography (CT) and magnetic resonance imaging (MRI) radiomic features to classify odontogenic cysts and tumours. CT and MRI data from patients with odontogenic lesions including dentigerous cysts, odontogenic keratocysts, and ameloblastomas were analysed. Manual segmentation of the CT image and the apparent diffusion coefficient (ADC) map from diffusion-weighted MRI was performed to extract radiomic features. The extracted radiomic features were split into training (70%) and test (30%) sets. The random forest model was adjusted or optimized using 5-fold stratified cross-validation within the training set and assessed on a separate hold-out test set. Analysis of the CT-based ML model showed cross-validation accuracy of 0.59 and 0.60 for the training set and test set, respectively, with precision, recall, and F1 score all being 0.57. Analysis of the ADC-based ML model showed cross-validation accuracy of 0.90 and 0.94 for the training set and test set, respectively; the precision, recall, and F1 score were all 0.87. ML models, particularly when using MRI radiological features, can effectively classify odontogenic lesions.

Recent "segment anything" efforts show promise by learning from large-scale data, but adapting such models directly to medical images remains challenging due to the complexity of medical data, noisy annotations, and continual learning requirements across diverse modalities and anatomical structures. In this work, we propose SAMed-2, a new foundation model for medical image segmentation built upon the SAM-2 architecture. Specifically, we introduce a temporal adapter into the image encoder to capture image correlations and a confidence-driven memory mechanism to store high-certainty features for later retrieval. This memory-based strategy counters the pervasive noise in large-scale medical datasets and mitigates catastrophic forgetting when encountering new tasks or modalities. To train and evaluate SAMed-2, we curate MedBank-100k, a comprehensive dataset spanning seven imaging modalities and 21 medical segmentation tasks. Our experiments on both internal benchmarks and 10 external datasets demonstrate superior performance over state-of-the-art baselines in multi-task scenarios. The code is available at: https://github.com/ZhilingYan/Medical-SAM-Bench.

The clinical utility of deep learning models for medical image segmentation is severely constrained by their inability to generalize to unseen domains. This failure is often rooted in the models learning spurious correlations between anatomical content and domain-specific imaging styles. To overcome this fundamental challenge, we introduce Causal-SAM-LLM, a novel framework that elevates Large Language Models (LLMs) to the role of causal reasoners. Our framework, built upon a frozen Segment Anything Model (SAM) encoder, incorporates two synergistic innovations. First, Linguistic Adversarial Disentanglement (LAD) employs a Vision-Language Model to generate rich, textual descriptions of confounding image styles. By training the segmentation model's features to be contrastively dissimilar to these style descriptions, it learns a representation robustly purged of non-causal information. Second, Test-Time Causal Intervention (TCI) provides an interactive mechanism where an LLM interprets a clinician's natural language command to modulate the segmentation decoder's features in real-time, enabling targeted error correction. We conduct an extensive empirical evaluation on a composite benchmark from four public datasets (BTCV, CHAOS, AMOS, BraTS), assessing generalization under cross-scanner, cross-modality, and cross-anatomy settings. Causal-SAM-LLM establishes a new state of the art in out-of-distribution (OOD) robustness, improving the average Dice score by up to 6.2 points and reducing the Hausdorff Distance by 15.8 mm over the strongest baseline, all while using less than 9% of the full model's trainable parameters. Our work charts a new course for building robust, efficient, and interactively controllable medical AI systems.

The integration of artificial intelligence in medical image classification has significantly advanced disease detection. However, traditional deep learning models face persistent challenges, including poor generalizability, high false-positive rates, and difficulties in distinguishing overlapping anatomical features, limiting their clinical utility. To address these limitations, this study proposes a hybrid framework combining Vision Transformers (ViT) and Perceiver IO, designed to enhance multi-disease classification accuracy. Vision Transformers leverage self-attention mechanisms to capture global dependencies in medical images, while Perceiver IO optimizes feature extraction for computational efficiency and precision. The framework is evaluated across three critical clinical domains: neurological disorders, including Stroke (tested on the Brain Stroke Prediction CT Scan Image Dataset) and Alzheimer's (analyzed via the Best Alzheimer MRI Dataset); skin diseases, covering Tinea (trained on the Skin Diseases Dataset) and Melanoma (augmented with dermoscopic images from the HAM10000/HAM10k dataset); and lung diseases, focusing on Lung Cancer (using the Lung Cancer Image Dataset) and Pneumonia (evaluated with the Pneumonia Dataset containing bacterial, viral, and normal X-ray cases). For neurological disorders, the model achieved 0.99 accuracy, 0.99 precision, 1.00 recall, 0.99 F1-score, demonstrating robust detection of structural brain abnormalities. In skin disease classification, it attained 0.95 accuracy, 0.93 precision, 0.97 recall, 0.95 F1-score, highlighting its ability to differentiate fine-grained textural patterns in lesions. For lung diseases, the framework achieved 0.98 accuracy, 0.97 precision, 1.00 recall, 0.98 F1-score, confirming its efficacy in identifying respiratory conditions. To bridge research and clinical practice, an AI-powered chatbot was developed for real-time analysis, enabling users to upload MRI, X-ray, or skin images for automated diagnosis with confidence scores and interpretable insights. This work represents the first application of ViT and Perceiver IO for these disease categories, outperforming conventional architectures in accuracy, computational efficiency, and clinical interpretability. The framework holds significant potential for early disease detection in healthcare settings, reducing diagnostic errors, and improving treatment outcomes for clinicians, radiologists, and patients. By addressing critical limitations of traditional models, such as overlapping feature confusion and false positives, this research advances the deployment of reliable AI tools in neurology, dermatology, and pulmonology.

Image quality assessment (IQA) is crucial in the evaluation stage of novel algorithms operating on images, including traditional and machine learning based methods. Due to the lack of available quality-rated medical images, most commonly used IQA methods employing reference images (i.e. full-reference IQA) have been developed and tested for natural images. Reported application inconsistencies arising when employing such measures for medical images are not surprising, as they rely on different properties than natural images. In photoacoustic imaging (PAI), especially, standard benchmarking approaches for assessing the quality of image reconstructions are lacking. PAI is a multi-physics imaging modality, in which two inverse problems have to be solved, which makes the application of IQA measures uniquely challenging due to both, acoustic and optical, artifacts. To support the development and testing of full- and no-reference IQA measures we assembled PhotIQA, a data set consisting of 1134 reconstructed photoacoustic (PA) images that were rated by 2 experts across five quality properties (overall quality, edge visibility, homogeneity, inclusion and background intensity), where the detailed rating enables usage beyond PAI. To allow full-reference assessment, highly characterised imaging test objects were used, providing a ground truth. Our baseline experiments show that HaarPSI$_{med}$ significantly outperforms SSIM in correlating with the quality ratings (SRCC: 0.83 vs. 0.62). The dataset is publicly available at https://doi.org/10.5281/zenodo.13325196.

To evaluate the performance of deep learning models in classifying parotid gland tumors using T2-weighted, diffusion-weighted, and contrast-enhanced T1-weighted MR images, along with DCE data derived from time-intensity curves. In this retrospective, single-center study including a total of 164 participants, 124 patients with surgically confirmed parotid gland tumors and 40 individuals with normal parotid glands underwent multiparametric MRI, including DCE sequences. Data partitions were performed at the patient level (80% training, 10% validation, 10% testing). Two deep learning architectures (MobileNetV2 and EfficientNetB0), as well as a combined approach integrating predictions from both models, were fine-tuned using transfer learning to classify (i) normal versus tumor (Task 1), (ii) benign versus malignant tumors (Task 2), and (iii) benign subtypes (Warthin tumor vs. pleomorphic adenoma) (Task 3). For Tasks 2 and 3, DCE-derived metrics were integrated via a support vector machine. Classification performance was assessed using accuracy, precision, recall, and F1-score, with 95% confidence intervals derived via bootstrap resampling. In Task 1, EfficientNetB0 achieved the highest accuracy (85%). In Task 2, the combined approach reached an accuracy of 65%, while adding DCE data significantly improved performance, with MobileNetV2 achieving an accuracy of 96%. In Task 3, EfficientNetB0 demonstrated the highest accuracy without DCE data (75%), while including DCE data boosted the combined approach to an accuracy of 89%. Adding DCE-MRI data to deep learning models substantially enhances parotid gland tumor classification accuracy, highlighting the value of functional imaging biomarkers in improving noninvasive diagnostic workflows.

Early and accurate detection of oral cancer plays a pivotal role in improving patient outcomes. This research introduces a custom-designed, 19-layer convolutional neural network (CNN) for the automated diagnosis of oral cancer using clinical images of the lips and tongue. The methodology integrates advanced preprocessing steps, including min-max normalization and histogram-based contrast enhancement, to optimize image features critical for reliable classification. The model is extensively validated on the publicly available Oral Cancer (Lips and Tongue) Images (OCI) dataset, which is divided into 80% training and 20% testing subsets. Comprehensive performance evaluation employs established metrics-accuracy, sensitivity, specificity, precision, and F1-score. Our CNN architecture achieved an accuracy of 99.54%, sensitivity of 95.73%, specificity of 96.21%, precision of 96.34%, and F1-score of 96.03%, demonstrating substantial improvements over prominent transfer learning benchmarks, including SqueezeNet, AlexNet, Inception, VGG19, and ResNet50, all tested under identical experimental protocols. The model's robust performance, efficient computation, and high reliability underline its practicality for clinical application and support its superiority over existing approaches. This study provides a reproducible pipeline and a new reference point for deep learning-based oral cancer detection, facilitating translation into real-world healthcare environments and promising enhanced diagnostic confidence.

Magnetic Resonance Imaging (MRI) and Computed Tomography (CT) are essential clinical cross-sectional imaging techniques for diagnosing complex conditions. However, large 3D datasets with annotations for deep learning are scarce. While methods like DINOv2 are encouraging for 2D image analysis, these methods have not been applied to 3D medical images. Furthermore, deep learning models often lack explainability due to their "black-box" nature. This study aims to extend 2D self-supervised models, specifically DINOv2, to 3D medical imaging while evaluating their potential for explainable outcomes. We introduce the Medical Slice Transformer (MST) framework to adapt 2D self-supervised models for 3D medical image analysis. MST combines a Transformer architecture with a 2D feature extractor, i.e., DINOv2. We evaluate its diagnostic performance against a 3D convolutional neural network (3D ResNet) across three clinical datasets: breast MRI (651 patients), chest CT (722 patients), and knee MRI (1199 patients). Both methods were tested for diagnosing breast cancer, predicting lung nodule dignity, and detecting meniscus tears. Diagnostic performance was assessed by calculating the Area Under the Receiver Operating Characteristic Curve (AUC). Explainability was evaluated through a radiologist's qualitative comparison of saliency maps based on slice and lesion correctness. P-values were calculated using Delong's test. MST achieved higher AUC values compared to ResNet across all three datasets: breast (0.94 ± 0.01 vs. 0.91 ± 0.02, P = 0.02), chest (0.95 ± 0.01 vs. 0.92 ± 0.02, P = 0.13), and knee (0.85 ± 0.04 vs. 0.69 ± 0.05, P = 0.001). Saliency maps were consistently more precise and anatomically correct for MST than for ResNet. Self-supervised 2D models like DINOv2 can be effectively adapted for 3D medical imaging using MST, offering enhanced diagnostic accuracy and explainability compared to convolutional neural networks.

Skeletal Class II malocclusion is commonly treated using mandibular advancement appliances during growth. Evaluating the comparative effectiveness of different appliances can help optimize treatment outcomes. This study aimed to compare dental and skeletal outcomes of Class II malocclusion treatment using Herbst and PowerScope appliances in conjunction with fixed orthodontic therapy. This retrospective comparative study included 46 consecutively treated patients in two university clinics: 26 with PowerScope and 20 with Herbst MiniScope. CBCT scans were obtained before and after treatment. Skeletal and dental changes were analyzed using maxillary and mandibular voxel-based regional superimpositions and cranial base registrations, aided by AI-based landmark detection. Measurement bias was minimized through the use of a calibrated, blinded examiner. No patients were excluded from the analysis. Due to the study's retrospective nature, no prospective registration was performed; the institutional review board granted ethical approval. The Herbst group showed greater anterior displacement at B-point and Pogonion than PowerScope (2.4 mm and 2.6 mm, respectively). Both groups exhibited improved maxillomandibular relationships, with PowerScope's SNA angle reduced and Herbst's SNB increased. Vertical skeletal changes were observed at points A, B, and Pog in both groups. Herbst also resulted in less lower incisor proclination and more pronounced distal movement of upper incisors. Both appliances effectively corrected Class II malocclusion. Herbst promoted more pronounced skeletal advancement, while PowerScope induced greater dental compensation. These findings may be generalizable to similarly aged Class II patients in CVM stages 3-4.

Deep implicit functions (DIFs) effectively represent shapes by using a neural network to map 3D spatial coordinates to scalar values that encode the shape's geometry, but it is difficult to establish correspondences between shapes directly, limiting their use in medical image registration. The recently presented deformation field-based methods achieve implicit templates learning via template field learning with DIFs and deformation field learning, establishing shape correspondence through deformation fields. Although these approaches enable joint learning of shape representation and shape correspondence, the decoupled optimization for template field and deformation field, caused by the absence of deformation annotations lead to a relatively accurate template field but an underoptimized deformation field. In this paper, we propose a novel implicit template learning framework via a shared hybrid diffeomorphic flow (SHDF), which enables shared optimization for deformation and template, contributing to better deformations and shape representation. Specifically, we formulate the signed distance function (SDF, a type of DIFs) as a one-dimensional (1D) integral, unifying dimensions to match the form used in solving ordinary differential equation (ODE) for deformation field learning. Then, SDF in 1D integral form is integrated seamlessly into the deformation field learning. Using a recurrent learning strategy, we frame shape representations and deformations as solving different initial value problems of the same ODE. We also introduce a global smoothness regularization to handle local optima due to limited outside-of-shape data. Experiments on medical datasets show that SHDF outperforms state-of-the-art methods in shape representation and registration.