Search Results (3,485)

Machine learning algorithms are commonly employed for classification and interpretation of high-dimensional data. The classification task is often broken down into two separate procedures, and different methods are applied to achieve accurate results and produce interpretable outcomes. First, an effective subset of high-dimensional features must be extracted and then the selected subset will be used to train a classifier. Gradient Boosted Trees (GBT) is an ensemble model and, particularly due to their robustness, ability to model complex nonlinear interactions, and feature interpretability, they are well suited for complex applications. XGBoost (eXtreme Gradient Boosting) is a high-performance implementation of GBT that incorporates regularization, parallel computation, and efficient tree pruning that makes it a suitable efficient, interpretable, and scalable classifier with potential applications to medical data analysis. In this study, a Monte Carlo Gradient Boosted Trees (MCGBT) model is proposed for both feature reduction and classification. The proposed MCGBT method was applied to a lung cancer dataset for feature identification and classification. The dataset contains 107 radiomics which are quantitative imaging biomarkers extracted from CT scans. A reduced set of 12 radiomics were identified, and patients were classified into different cancer stages. Cancer staging accuracy of 90.3% across 100 independent runs was achieved which was on par with that obtained using the full set of 107 radiomics, enabling lean and deployable classifiers. Full article

Objectives: Spinal diseases are commonly encountered health problems with a wide spectrum. In addition to degenerative changes, other common spinal pathologies include metastases and compression fractures. Benign tumors like hemangiomas and infections such as spondylodiscitis are also frequently observed. Although magnetic resonance imaging (MRI) is considered the gold standard in diagnostic imaging, the morphological similarities of lesions can pose significant challenges in differential diagnoses. In recent years, the use of artificial intelligence applications in medical imaging has become increasingly widespread. In this study, we aim to detect and classify vertebral body lesions using the YOLO-v8 (You Only Look Once, version 8) deep learning architecture. Materials and Methods: This study included MRI data from 235 patients with vertebral body lesions. The dataset comprised sagittal T1- and T2-weighted sequences. The diagnostic categories consisted of acute compression fractures, metastases, hemangiomas, atypical hemangiomas, and spondylodiscitis. For automated detection and classification of vertebral lesions, the YOLOv8 deep learning model was employed. Following image standardization and data augmentation, a total of 4179 images were generated. The dataset was randomly split into training (80%) and validation (20%) subsets. Additionally, an independent test set was constructed using MRI images from 54 patients who were not included in the training or validation phases to evaluate the model’s performance. Results: In the test, the YOLOv8 model achieved classification accuracies of 0.84 and 0.85 for T1- and T2-weighted MRI sequences, respectively. Among the diagnostic categories, spondylodiscitis had the highest accuracy in the T1 dataset (0.94), while acute compression fractures were most accurately detected in the T2 dataset (0.93). Hemangiomas exhibited the lowest classification accuracy in both modalities (0.73). The F1 scores were calculated as 0.83 for T1-weighted and 0.82 for T2-weighted sequences at optimal confidence thresholds. The model’s mean average precision (mAP) 0.5 values were 0.82 for T1 and 0.86 for T2 datasets, indicating high precision in lesion detection. Conclusions: The YOLO-v8 deep learning model we used demonstrates effective performance in distinguishing vertebral body metastases from different groups of benign pathologies. Full article

Background/Objectives: Despite medical advances, stigma remains a major challenge for people living with HIV (PLWH). This study examined clinical, sociodemographic, and psychological predictors of HIV-related stigma, and explored whether affective temperament moderates the impact of depression on stigma. Methods: This cross-sectional observational study included 97 PLWH attending a tertiary infectious disease unit in Rome, Italy. Participants completed a battery of validated psychometric instruments assessing depressive symptoms, anxiety, manic symptoms, mixed affective states, general psychopathology, impulsivity, and affective temperament. HIV-related stigma was evaluated using the Berger HIV Stigma Scale, which measures personalized stigma, disclosure concerns, negative self-image, and concerns with public attitudes. Descriptive statistics were used to characterize the sample. Univariate linear regressions were conducted to explore associations between clinical, psychometric, and sociodemographic variables and each stigma subdimension, as well as the total stigma score. Variables significant at p < 0.05 were included in five multivariate linear regression models. Moderation analyses were subsequently performed to assess whether affective temperaments moderated the relationship between significant psychopathological predictors and stigma. Bonferroni correction was applied where appropriate. Results: Higher depressive symptom scores are significantly associated with greater internalized stigma (B = 0.902, p = 0.006) and total stigma (B = 2.603, p = 0.008). Furthermore, moderation analyses showed that anxious temperament significantly intensified the relationship between depressive symptoms and both negative self-image (interaction term B = 0.125, p = 0.001) and total stigma (B = 0.336, p = 0.002). Conclusions: Depressive symptoms and anxious temperament are associated with HIV-related stigma. Integrating psychological screening and targeted interventions for mood and temperament vulnerabilities may help reduce stigma burden in PLWH and improve psychosocial outcomes. Full article

Background: Periodontitis is a multifactorial disease leading to the loss of clinical attachment and alveolar bone. The diagnosis of periodontitis involves a clinical examination and radiographic evaluation, including panoramic images. Panoramic radiographs are cost-effective methods widely used in periodontitis classification. The remaining bone height (RBH) is a parameter used to assess the alveolar bone level. Large language models are widely utilized in the medical sciences. ChatGPT, the leading conversational model, has recently been extended to process visual data. The aim of this study was to assess the effectiveness of the ChatGPT models 4.5, o1, o3 and o4-mini-high in RBH measurement and tooth counts in relation to dental professionals’ evaluations. Methods: The analysis was based on 10 panoramic images, from which 252, 251, 246 and 271 approximal sites were qualified for the RBH measurement (using the models 4.5, o1, o3 and o4-mini-high, respectively). Three examiners were asked to independently evaluate the RBH in approximal sites, while the tooth count was achieved by consensus. Subsequently, the results were compared with the ChatGPT outputs. Results: ChatGPT 4.5, ChatGPT o3 and ChatGPT o4-mini-high achieved substantial agreement with clinicians in the assessment of tooth counts (κ = 0.65, κ = 0.66, κ = 0.69, respectively), while ChatGPT o1 achieved moderate agreement (κ = 0.52). In the context of RBH values, the ChatGPT models consistently exhibited a positive mean bias compared with the clinicians. ChatGPT 4.5 was reported to provide the lowest bias (+12 percentage points (pp) for the distal surfaces, width of the 95% CI for limits of agreement (LoAs) ~60 pp; +11 pp for the mesial surfaces, LoA width ~54 pp). Conclusions: ChatGPT 4.5 and ChatGPT o3 show potential in the assessment of tooth counts on a panoramic radiograph; however, their present level of accuracy is insufficient for clinical use. In the current stage of development, the ChatGPT models substantially overestimated the RBH values; therefore, they are not applicable for classifying periodontal disease. Full article

Digital images have been widely applied in fields such as mobile devices, the Internet of Things, and medical imaging. Although significant progress has been made in image encryption technology, it still faces many challenges, such as attackers using powerful computing resources and advanced algorithms to crack encryption systems. To address these challenges, this paper proposes a novel image encryption algorithm based on one-dimensional sawtooth wave chaotic system (1D-SAW) and the three-strand structure of DNA. Firstly, a new 1D-SAW chaotic system was designed. By introducing nonlinear terms and periodic disturbances, this system is capable of generating chaotic sequences with high randomness and initial value sensitivity. Secondly, a new diffusion rule based on the three-strand structure of DNA is proposed. Compared with the traditional DNA encoding and XOR operation, this rule further enhances the complexity and anti-attack ability of the encryption process. Finally, the security and randomness of the 1D-SAW and image encryption algorithms were verified through various tests. Results show that this method exhibits better performance in resisting statistical attacks and differential attacks. Full article

This paper presents the PRONOBIS project, an ultrasound-only, robotically assisted, deep learning-based system for prostate scanning and biopsy treatment planning. The proposed system addresses the challenges of precise prostate segmentation, reconstruction and inter-operator variability by performing fully automated prostate scanning, real-time CNN-transformer-based image processing, 3D prostate reconstruction, and biopsy needle position planning. Fully automated prostate scanning is achieved by using a robotic arm equipped with an ultrasound system. Real-time ultrasound image processing utilizes state-of-the-art deep learning algorithms with intelligent post-processing techniques for precise prostate segmentation. To create a high-quality prostate segmentation dataset, this paper proposes a deep learning-based medical annotation platform, MedAP. For precise segmentation of the entire prostate sweep, DAF3D and MicroSegNet models are evaluated, and additional image post-processing methods are proposed. Three-dimensional visualization and prostate reconstruction are performed by utilizing the segmentation results and robotic positional data, enabling robust, user-friendly biopsy treatment planning. The real-time sweep scanning and segmentation operate at 30 Hz, which enable complete scan in 15 to 20 s, depending on the size of the prostate. The system is evaluated on prostate phantoms by reconstructing the sweep and by performing dimensional analysis, which indicates 92% and 98% volumetric accuracy on the tested phantoms. Three-dimansional prostate reconstruction takes approximately 3 s and enables fast and detailed insight for precise biopsy needle position planning. Full article

Mid-infrared passive spectroscopic imaging is a novel non-invasive and remote sensing method based on Planck’s law. It enables the acquisition of component-specific information from the human body by measuring naturally emitted thermal radiation in the mid-infrared region. Unlike active methods that require an external light source, our passive approach harnesses the body’s own emission, thereby enabling safe, long-term monitoring. In this study, we successfully demonstrated the simultaneous, non-invasive measurements of blood glucose and lactate levels of the human body using this method. The measurements, conducted over approximately 80 min, provided emittance data derived from mid-infrared passive spectroscopy that showed a temporal correlation with values obtained using conventional blood collection sensors. Furthermore, to evaluate localized metabolic changes, we performed k-means clustering analysis of the spectral data obtained from the upper arm. This enabled visualization of time-dependent lactate responses with spatial resolution. These results demonstrate the feasibility of multi-component monitoring without physical contact or biological sampling. The proposed technique holds promise for translation to medical diagnostics, continuous health monitoring, and sports medicine, in addition to facilitating the development of next-generation healthcare technologies. Full article

In the last two decades, there has been a considerable surge in the development of artificial intelligence. Imaging is most frequently employed for the diagnostic evaluation of patients, as it is regarded as one of the most precise methods for identifying the presence of a disease. However, a study indicates that approximately 800,000 individuals in the USA die or incur permanent disability because of misdiagnosis. The present study is based on the use of computer-aided diagnosis of colorectal cancer. The objective of this study is to develop a practical, low-cost, AI-based decision-support tool that integrates clinical test data (blood/stool) and, if needed, colonoscopy images to help reduce misdiagnosis and improve early detection of colorectal cancer for clinicians. Convolutional neural networks (CNNs) and artificial neural networks (ANNs) are utilized in conjunction with a graphical user interface (GUI), which caters to individuals lacking programming expertise. The performance of the artificial neural network (ANN) is measured using the mean squared error (MSE) metric, and the obtained performance is 7.38. For CNN, two distinct cases are under consideration: one with two outputs and one with three outputs. The precision of the models is 97.2% for RGB and 96.7% for grayscale, respectively, in the first instance, and 83% for RGB and 82% for grayscale in the second instance. However, using a pretrained network yielded superior performance with 99.5% for 2-output models and 93% for 3-output models. The GUI is composed of two panels, with the best ANN model and the best CNN model being utilized in each. The primary function of the tool is to assist medical personnel in reducing the time required to make decisions and the probability of misdiagnosis. Full article

Background/Objectives: Accurate segmentation of brain stroke lesions from MRI images is a critical task in medical image analysis that is essential for timely diagnosis and treatment planning. Methods: This paper presents a novel approach for segmenting brain stroke lesions using a deep learning model based on the U-Net neural network architecture. We enhanced the traditional U-Net by integrating a transformer-based backbone, specifically the Mix Vision Transformer (MiT), and compared its performance against other commonly used backbones such as ResNet and EfficientNet. Additionally, we implemented a 2.5D method, which leverages 2D networks to process three-dimensional data slices, effectively balancing the rich spatial context of 3D methods and the simplicity of 2D methods. The 2.5D approach captures inter-slice dependencies, leading to improved lesion delineation without the computational complexity of full 3D models. Utilizing the 2015 ISLES dataset, which includes MRI images and corresponding lesion masks for 20 patients, we conducted our experiments with 4-fold cross-validation to ensure robustness and reliability. To evaluate the effectiveness of our method, we conducted comparative experiments with several state-of-the-art (SOTA) segmentation models, including CNN-based UNet, nnU-Net, TransUNet, and SwinUNet. Results: Our proposed model outperformed all competing methods in terms of Dice Coefficient and Intersection over Union (IoU), demonstrating its robustness and superiority. Our extensive experiments demonstrate that the proposed U-Net with the MiT Backbone, combined with 2.5D data preparation, achieves superior performance metrics, specifically achieving DICE and IoU scores of 0.8153 ± 0.0101 and 0.7835 ± 0.0079, respectively, outperforming other backbone configurations. Conclusions: These results indicate that the integration of transformer-based backbones and 2.5D techniques offers a significant advancement in the accurate segmentation of brain stroke lesions, paving the way for more reliable and efficient diagnostic tools in clinical settings. Full article

Background/Objectives: Muscle dysmorphia (MD), a subtype of body dysmorphic disorder, is prevalent among males who engage in the non-medical use of anabolic–androgenic steroids (AASs) and performance-enhancing drugs (PEDs). These individuals often experience severe psychopathology, including mood instability, compulsivity, and a distorted body image. Despite its clinical severity, no randomized controlled trials (RCTs) have evaluated structured psychological treatments in this subgroup. This study aimed to assess the efficacy of a manualized cognitive behavioral therapy (CBT) protocol in reducing MD symptoms and associated psychological distress among male steroid users. Results: Participants in the CBT group showed significant reductions in MD symptoms from the baseline to post-treatment (MDDI: p < 0.001, d = 1.12), with gains sustained at follow-up. Large effect sizes were also observed in secondary outcomes including depressive symptoms (PHQ-9: d = 0.98), psychological distress (K10: d = 0.93), disordered eating (EDE-Q: d = 0.74), and exercise addiction (EAI: d = 1.07). No significant changes were observed in the control group. Significant group × time interactions were found for all outcomes (all p < 0.01), indicating CBT’s specific efficacy. Discussion: This study provides the first RCT evidence that CBT significantly reduces both core MD symptoms and steroid-related psychopathology in men engaged in AAS/PED misuse. Improvements extended to mood, body image perception, and compulsive exercise behaviors. These findings support CBT’s transdiagnostic applicability in addressing both the cognitive–behavioral and affective dimensions of MD. Materials and Methods: In this parallel-group, open-label RCT, 59 male gym-goers with DSM-5-TR diagnoses of MD and a history of AAS/PED use were randomized to either a 12-week CBT intervention (n = 30) or a waitlist control group (n = 29). CBT sessions were delivered weekly online and targeted distorted muscularity beliefs, compulsive behaviors, and emotional dysregulation. Primary and secondary outcomes—Muscle Dysmorphic Disorder Inventory (MDDI), PHQ-9, K10, EDE-Q, EAI, and BIG—were assessed at the baseline, post-treatment, and 3-month follow-up. A repeated-measures ANOVA and paired t-tests were used to analyze time × group interactions. Conclusions: CBT offers an effective, scalable intervention for individuals with muscle dysmorphia complicated by anabolic steroid use. It promotes broad psychological improvement and may serve as a first-line treatment option in high-risk male fitness populations. Future studies should examine long-term outcomes and investigate implementation in diverse clinical and cultural contexts. Full article

Oral cancer is typically diagnosed through histological examination; however, the primary issue with this type of procedure is tumor heterogeneity, where a subjective aspect of the examination may have a direct effect on the treatment plan for a patient. To reduce inter- and intra-observer variability, artificial intelligence algorithms are often used as computational aids in tumor classification and diagnosis. This research proposes a two-step approach for automatic multiclass grading using oral histopathology images (the first step) and Grad-CAM visualization (the second step) to assist clinicians in diagnosing oral squamous cell carcinoma. The Xception architecture achieved the highest classification values of 0.929 (±σ = 0.087) AUC_macro and 0.942 (±σ = 0.074) AUC_micro. Additionally, Grad-CAM provided visual explanations of the model’s predictions by highlighting the precise areas of histopathology images that influenced the model’s decision. These results emphasize the potential of integrated AI algorithms in medical diagnostics, offering a more precise, dependable, and effective method for disease analysis. Full article

Background: The central variant of posterior reversible encephalopathy syndrome (cvPRES) is an atypical subtype of PRES. Although no unifying definitions exists, it is most often characterized by vasogenic edema involving “central” structures, such as the brainstem, subcortical nuclei, and spinal cord, with relative sparing of the parieto-occipital lobes. Methods: This systematic review and meta-analysis followed the PRISMA guidelines and was pre-registered on PROSPERO [CRD42023483806]. Both the Joanna Briggs Institute and New-Castle Ottawa scale were used for case reports and cohort studies, respectively. The meta-analysis was completed using R-Studio and its associated “metafor” package. Results: A comprehensive search in four databases yielded 70 case reports/series (n = 100) and 12 cohort studies. The meta-analysis revealed a pooled incidence rate of 13% (95% CI: 9–18%) for cvPRES amongst included cohort studies on PRES. Significant heterogeneity was observed (I² = 71% and a τ² = 0.2046). The average age of affected individuals was 40.9 years, with a slightly higher prevalence in males (54%). The most common etiological factor was hypertension (72%). Fifty percent had an SBP >200 mmHg at presentation and a mean arterial pressure (MAP) of 217.6 ± 40.82. Imaging revealed an increased T₂ signal involving the brain stem (88%), most often in the pons (62/88; 70.45%), and 18/100 (18%) cases of PRES with spinal cord involvement (PRES-SCI). Management primarily involved blood pressure reduction, with adjunctive therapies for underlying causes such as anti-seizure medications or hemodialysis. The MAP between isolated PRES-SCI and cvPRES without spinal cord involvement did not show significant differences (p = 0.5205). Favorable outcomes were observed in most cases, with a mortality rate of only 2%. Conclusions: cvPRES is most often associated with higher blood pressure compared to prior studies with typical PRES. The pons is most often involved. Despite the severity of blood pressure and critical brain stem involvement, those with cvPRES have favorable functional outcomes and a lower mortality rate than typical PRES, likely attributable to reversible vasogenic edema without significant neuronal dysfunction. Full article

Background/Objectives: Discrepancies in diabetic retinopathy (DR) grading are well-documented, with retinal non-perfusion (RNP) quantification posing greater challenges. This study assessed intergrader agreement in DR evaluation, focusing on qualitative severity grading and quantitative RNP measurement. We aimed to improve agreement through structured consensus meetings. Methods: A retrospective analysis of 100 comparisons from 50 eyes (36 patients) was conducted. Two paired medical retina fellows graded ultra-widefield color fundus photographs (CFP) and fundus fluorescein angiography (FFA) images. CFP assessments included DR severity using the International Clinical Diabetic Retinopathy (ICDR) grading system, DR Severity Scale (DRSS), and predominantly peripheral lesions (PPL). FFA-based RNP was defined as capillary loss with grayscale matching the foveal avascular zone. Weekly adjudication by a senior specialist resolved discrepancies. Intergrader agreement was evaluated using Cohen’s kappa (qualitative DRSS) and intraclass correlation coefficients (ICC) (quantitative RNP). Bland–Altman analysis assessed bias and variability. Results: After eight consensus meetings, CFP grading agreement improved to excellent: kappa = 91% (ICDR DR severity), 89% (DRSS), and 89% (PPL). FFA-based PPL agreement reached 100%. For RNP, the non-perfusion index (NPI) showed moderate overall ICC (0.49), with regional ICCs ranging from 0.40 to 0.57 (highest in the nasal region, ICC = 0.57). Bland–Altman analysis revealed a mean NPI difference of 0.12 (limits: −0.11 to 0.35), indicating acceptable variability despite outliers. Conclusions: Structured consensus training achieved excellent intergrader agreement for DR severity and PPL grading, supporting the clinical reliability of ultra-widefield imaging. However, RNP measurement variability underscores the need for standardized protocols and automated tools to enhance reproducibility. This process is critical for developing robust AI-based screening systems. Full article

Over recent decades, significant advancements have been made in the treatment and imaging of gliomas. Conventional imaging techniques, such as MRI and CT, play critical roles in glioma diagnosis and treatment but often fail to distinguish between tumor pseudoprogression (Psp) and radiation necrosis (RN) versus true progression (TP). Emerging fields like radiomics and radiogenomics are addressing these challenges by extracting quantitative features from medical images and correlating them with genomic data, respectively. This article will discuss several studies that show how radiomic features (RFs) can aid in better patient stratification and prognosis. Radiogenomics, particularly in predicting biomarkers such as MGMT promoter methylation and 1p/19q codeletion, shows potential in non-invasive diagnostics. Radiomics also offers tools for predicting tumor recurrence (rBT), essential for treatment management. Further research is needed to standardize these methods and integrate them into clinical practice. This review underscores radiomics and radiogenomics’ potential to revolutionize glioma management, marking a significant shift towards precision neuro-oncology. Full article

Background: Optical Coherence Tomography has become a common imaging technique that enables a non-invasive and detailed visualization of the retina and allows for the identification of various diseases. Through the advancement of technology, the volume and complexity of OCT data have rendered manual analysis infeasible, creating the need for automated means of detection. Methods: This study investigates the ability of state-of-the-art object detection models, including the latest YOLO versions (from v8 to v12), YOLO-World, YOLOE, and RT-DETR, to accurately detect pathological biomarkers in two retinal OCT datasets. The AROI dataset focuses on fluid detection in Age-related Macular Degeneration, while the OCT5k dataset contains a wide range of retinal pathologies. Results: The experiments performed show that YOLOv12 offers the best balance between detection accuracy and computational efficiency, while YOLOE manages to consistently outperform all other models across both datasets and most classes, particularly in detecting pathologies that cover a smaller area. Conclusions: This work provides a comprehensive benchmark of the capabilities of state-of-the-art object detection for medical applications, specifically for identifying retinal pathologies from OCT scans, offering insights and a starting point for the development of future automated solutions for analysis in a clinical setting. Full article