Search Results (9,228)

Background/Objectives: Osteosarcoma (OS) is the most common malignant bone tumor in children and adolescents; the survival rate is as low as 24%. Accurate prediction of clinical outcomes remains a challenge due to tumor heterogeneity and the complexity of pediatric cases. This study aims to improve predictions of progressive disease, therapy response, relapse, and survival in pediatric OS using MRI-based radiomics and machine learning methods. Methods: Pre-treatment contrast-enhanced coronal T1-weighted MR scans were collected from 63 pediatric OS patients, with an additional nine external cases used for validation. Three strategies were considered for target region segmentation (whole-tumor, tumor sampling, and bone/soft tissue) and used for MRI-based radiomics. These were then combined with clinical features to predict OS clinical outcomes. Results: The mean age of OS patients was 11.8 ± 3.5 years. Most tumors were located in the femur (65%). Osteoblastic subtype was the most common histological classification (79%). The majority of OS patients (79%) did not have evidence of metastasis at diagnosis. Progressive disease occurred in 27% of patients, 59% of patients showed adequate therapy response, 25% experienced relapse after therapy, and 30% died from OS. Classification models based on bone/soft tissue segmentation generally performed the best, with certain clinical features improving performance, especially for therapy response and mortality. The top performing classifier in each outcome achieved 0.94–1.0 validation ROC AUC and 0.63–1.0 testing ROC AUC, while those without radiomic features (RFs) generally performed suboptimally. Conclusions: This study demonstrates the strong predictive capabilities of MRI-based radiomics and multi-region segmentations for predicting clinical outcomes in pediatric OS. Full article

If we do not urgently educate current and future medical professionals to critically evaluate and distinguish credible AI-assisted diagnostic tools from those whose performance is artificially inflated by data leakage or improper validation, we risk undermining clinician trust in all AI diagnostics and jeopardizing future advances in patient care. For instance, machine learning models have shown high accuracy in diagnosing Parkinson’s Disease when trained on clinical features that are themselves diagnostic, such as tremor and rigidity. This study systematically investigates the impact of data leakage and feature selection on the true clinical utility of machine learning models for early Parkinson’s Disease detection. We constructed two experimental pipelines: one excluding all overt motor symptoms to simulate a subclinical scenario and a control including these features. Nine machine learning algorithms were evaluated using a robust three-way data split and comprehensive metric analysis. Results reveal that, without overt features, all models exhibited superficially acceptable F1 scores but failed catastrophically in specificity, misclassifying most healthy controls as Parkinson’s Disease. The inclusion of overt features dramatically improved performance, confirming that high accuracy was due to data leakage rather than genuine predictive power. These findings underscore the necessity of rigorous experimental design, transparent reporting, and critical evaluation of machine learning models in clinically realistic settings. Our work highlights the risks of overestimating model utility due to data leakage and provides guidance for developing robust, clinically meaningful machine learning tools for early disease detection. Full article

Non-invasive ultrasound diagnosis, combined with deep learning, is frequently used for detecting thyroid diseases. However, real-time detection on portable devices faces limitations due to constrained computational resources, and existing models often lack sufficient capability for small object detection of thyroid nodules. To address this, this paper proposes an improved lightweight small object detection network framework called LISA-YOLO, which enhances the lightweight multi-scale collaborative fusion algorithm. The proposed framework exploits the inherent symmetrical characteristics of ultrasound images and the symmetrical architecture of the detection network to better capture and represent features of thyroid nodules. Specifically, an improved depthwise separable convolution algorithm replaces traditional convolution to construct a lightweight network (DG-FNet). Through symmetrical cross-scale fusion operations via FPN, detection accuracy is maintained while reducing computational overhead. Additionally, an improved bidirectional feature network (IMS F-NET) fully integrates the semantic and detailed information of high- and low-level features symmetrically, enhancing the representation capability for multi-scale features and improving the accuracy of small object detection. Finally, a collaborative attention mechanism (SAF-NET) uses a dual-channel and spatial attention mechanism to adaptively calibrate channel and spatial weights in a symmetric manner, effectively suppressing background noise and enabling the model to focus on small target areas in thyroid ultrasound images. Extensive experiments on two image datasets demonstrate that the proposed method achieves improvements of 2.3% in F1 score, 4.5% in mAP, and 9.0% in FPS, while maintaining only 2.6 M parameters and reducing GFLOPs from 6.1 to 5.8. The proposed framework provides significant advancements in lightweight real-time detection and demonstrates the important role of symmetry in enhancing the performance of ultrasound-based thyroid diagnosis. Full article

The Internet of Medical Things (IoMT) is revolutionizing healthcare by enabling continuous patient monitoring, early diagnosis, and personalized treatments. However, the het-erogeneity of IoMT devices and the lack of standardized protocols introduce serious security vulnerabilities. To address these challenges, we propose a hybrid BiLSTM-DNN intrusion detection system, named HBiLD-IDS, that combines Bidirectional Long Short-Term Memory (BiLSTM) networks with Deep Neural Networks (DNNs), leveraging both temporal dependencies in network traffic and hierarchical feature extraction. The model is trained and evaluated on the CICIoMT2024 dataset, which accurately reflects the diversity of devices and attack vectors encountered in connected healthcare environments. The dataset undergoes rigorous preprocessing, including data cleaning, feature selection through correlation analysis and recursive elimination, and feature normalization. Compared to existing IDS models, our approach significantly enhances detection accuracy and generalization capacity in the face of complex and evolving attack patterns. Experimental results show that the proposed IDS model achieves a classification accuracy of 98.81% across 19 attack types confirming its robustness and scalability. This approach represents a promising solution for strengthening the security posture of IoMT networks against emerging cyber threats. Full article

Objectives: Lermoyez syndrome (LS) is a rare variant of endolymphatic hydrops with a unique clinical presentation characterized by reversible sensorineural hearing loss preceding vertigo. This review aims to synthesize available literature on LS to clarify its clinical characteristics, diagnostic approach, management strategies, and outcomes, and to highlight the distinguishing features from Menière’s disease (MD). Methods: A systematic literature review according to PRISMA guidelines was conducted from 1919 to 2025. The extracted data included demographics, symptom profiles, audiovestibular testing, imaging findings, treatment approaches, and patient outcomes. Results: A total of 23 studies were identified, reporting 53 individual cases of LS. Patients ranged from 27 to 85 years of age, with a mean age of 50.34 years and a male predominance (64.1%). The hallmark of LS across cases was a reproducible clinical pattern of unilateral low-frequency hearing loss followed by vertigo and subsequent auditory recovery. Audiometry typically confirmed reversible sensorineural hearing loss, while vestibular tests and imaging were often unremarkable, primarily used to exclude alternative diagnoses. Treatment approaches varied and were often based on MD protocols, including dietary modifications, vasodilators, diuretics, and vestibular suppressants. Prognosis was generally favorable, with most patients experiencing both hearing recovery and symptom resolution. Conclusions: LS remains a clinically distinct but underrecognized inner ear disorder. Its defining feature—the paradoxical improvement in hearing after vertigo—distinguishes it from Menière’s disease and should prompt clinicians to consider LS in differential diagnosis. Due to the rarity of LS and the lack of standardized guidelines, diagnosis and treatment rely on careful clinical assessment and individualized management strategies. Full article

Despite advances in Convolutional Neural Networks (CNNs) for intelligent fault diagnosis in CNC machine tools, bearing fault diagnosis in CNC feed systems remains challenging, particularly in multi-scale feature extraction and generalization across operating conditions. This study introduces an enhanced multi-scale feature network (MSFN) that addresses these limitations through three integrated modules designed to extract critical fault features from vibration signals. First, a Soft-Scale Denoising (S2D) module forms the backbone of the MSFN, capturing multi-scale fault features from input signals. Second, a Multi-Scale Adaptive Feature Enhancement (MS-AFE) module based on long-range weighting mechanisms is developed to enhance the extraction of periodic fault features. Third, a Dynamic Sequence–Channel Attention (DSCA) module is incorporated to improve feature representation across channel and sequence dimensions. Experimental results on two datasets demonstrate that the proposed MSFN achieves high diagnostic accuracy and exhibits robust generalization across diverse operating conditions. Moreover, ablation studies validate the effectiveness and contributions of each module. Full article

Electroencephalogram (EEG) analysis is essential for diagnosing neurological disorders but typically requires expert interpretation and significant time. Purpose: This study aims to automate the classification of normal and abnormal EEG recordings to support clinical diagnosis and reduce manual workload. Automating the initial screening of EEGs can help clinicians quickly identify potential neurological abnormalities, enabling timely intervention and guiding further diagnostic and treatment strategies. Methodology: We utilized the Temple University Hospital EEG dataset to develop a DenseNet-based deep learning model. To enable a fair comparison of different EEG representations, we used three input types: signal images, spectrograms, and scalograms. To reduce dimensionality and simplify computation, we focused on two channels: T5 and O1. For interpretability, we applied Local Interpretable Model-agnostic Explanations (LIME) and Gradient-weighted Class Activation Mapping (Grad-CAM) to visualize the EEG regions influencing the model’s predictions. Key Findings: Among the input types, spectrogram-based representations achieved the highest classification accuracy, indicating that time-frequency features are especially effective for this task. The model demonstrated strong performance overall, and the integration of LIME and Grad-CAM provided transparent explanations of its decisions, enhancing interpretability. This approach offers a practical and interpretable solution for automated EEG screening, contributing to more efficient clinical workflows and better understanding of complex neurological conditions. Full article

Background/Objectives: Breast cancer was the leading cause of malignant tumors among women in 2022. About two-thirds of breast cancer cases are hormone-receptor-positive. In these patients, aromatase inhibitors are a mainstay of treatment, but associated musculoskeletal symptoms can negatively affect patient compliance. Aromatase-inhibitor-induced carpal tunnel syndrome represents one of the main causes of aromatase inhibitor discontinuation, with a non-compliance rate of up to 67%, potentially leading to increased cancer mortality. This study investigates estrogen receptor expression in aromatase-inhibitor-induced carpal tunnel syndrome tissues, in order to better define its etiopathogenesis and derive preventive or therapeutic measures that can improve aromatase inhibitor patient compliance. To our knowledge, there is no study on this subject in the literature. Methods: Between 2023 and 2024, we recruited 14 patients at the Jewish Hospital of Rome, including seven patients with aromatase-inhibitor-induced carpal tunnel syndrome (study group) and seven with postmenopausal idiopathic carpal tunnel syndrome (control group). Each patient was evaluated based on a clinical visit, a questionnaire, instrumental exams, and serum hormone dosages and were treated with open carpal tunnel release surgery, during which transverse carpal ligament and flexor tenosynovium samples were collected. For immunohistochemical experiments, sections were treated with anti-estrogen receptor α and anti-estrogen receptor β antibodies. Results: The immunohistochemical features in the study and control groups were similar, demonstrating that tissues affected by aromatase-inhibitor-induced carpal tunnel syndrome are targets of direct estrogen action and that estrogen deprivation is correlated with disease etiogenesis. Surgery was effective in patient treatment. Conclusions: Aromatase-inhibitor-induced carpal tunnel syndrome represents a newly defined form of the disease. This syndrome represents one of the main causes of aromatase inhibitor discontinuation, due to its negative impact on the patient’s quality of life. The identification by clinicians of aromatase inhibitor use as a possible risk factor for carpal tunnel syndrome development is of essential importance, as early diagnosis and prompt management can improve patient compliance and overall breast cancer treatment outcomes. Full article

Background: Breast cancer is the most prevalent malignancy among women globally. In Romania, it is the most frequent form of cancer affecting women, with approximately 12,000 new cases diagnosed annually, and the second most common cause of cancer-related mortality, second only to lung cancer. Methods: This study looked at 79 breast cancer patients from Oltenia, concentrating on epidemiology, histology, diagnostic features, and treatments. Patients were chosen based on inclusion criteria such as histopathologically verified diagnosis, availability of clinical and treatment data, and follow-up information. The analyzed biological material consisted of tissue samples taken from the breast parenchyma and axillary lymph nodes. Even though not the primary subject of this paper, all patients underwent immunohistochemical (IHC) evaluation both preoperatively and postoperatively. Results: We found invasive ductal carcinoma to be the predominant type, while ductal carcinoma in situ (DCIS) and mixed types were rare. We performed cross-tabulations of metastasis versus nodal status and age versus therapy type; none reached significance (all p > 0.05), suggesting observed differences were likely due to chance. A chi-square test comparing surgical interventions (breast-conserving vs. mastectomy) in patients who did or did not receive chemotherapy showed, χ² = 3.17, p = 0.367, indicating that chemotherapy did not significantly influence surgical choice. Importantly, adjuvant chemotherapy and radiotherapy were used at similar rates across age groups, whereas neoadjuvant hormonal (endocrine) therapy was more common in older patients (but without statistical significance). Conclusions: Finally, we discussed the consequences of individualized care and early detection. Romania’s shockingly low screening rate, which contributes to delayed diagnosis, emphasizes the importance of improved population medical examination and tailored treatment options. Also, the country has one of the lowest rates of mammography uptake in Europe and no systematic population screening program. Full article

Breast cancer is the leading cause of cancer-related mortality among women world-wide, and early screening is critical for improving patient survival. Medical imaging plays a central role in breast cancer screening, diagnosis, and treatment monitoring. However, conventional imaging modalities—including mammography, ultrasound, and magnetic resonance imaging—face limitations such as low diagnostic specificity, relatively slow imaging speed, ionizing radiation exposure, and dependence on exogenous contrast agents. Photoacoustic imaging (PAI), a novel hybrid imaging technique that combines optical contrast with ultrasonic spatial resolution, has shown great promise in addressing these challenges. By revealing anatomical, functional, and molecular features of the breast tumor microenvironment, PAI offers high spatial resolution, rapid imaging, and minimal operator dependence. This review outlines the fundamental principles of PAI and systematically examines recent advances in its application to breast cancer screening, diagnosis, and therapeutic evaluation. Furthermore, we discuss the translational potential of PAI as an emerging breast imaging modality, complementing existing clinical techniques. Full article

Cell confluence and number are critical indicators for assessing cellular growth status, contributing to disease diagnosis and the development of targeted therapies. Accurate and efficient cell segmentation is essential for quantifying these indicators. However, current segmentation methodologies still encounter significant challenges in addressing multi-scale heterogeneity, poorly delineated boundaries under limited annotation, and the inherent trade-off between computational efficiency and segmentation accuracy. We propose an innovative network architecture. First, a preprocessing pipeline combining contrast-limited adaptive histogram equalization (CLAHE) and Gaussian blur is introduced to balance noise suppression and local contrast enhancement. Second, a bidirectional feature pyramid network (BiFPN) is incorporated, leveraging cross-scale feature calibration to enhance multi-scale cell recognition. Third, adaptive kernel convolution (AKConv) is developed to capture the heterogeneous spatial distribution of glioma stem cells (GSCs) through dynamic kernel deformation, improving boundary segmentation while reducing model complexity. Finally, a probability density-guided non-maximum suppression (Soft-NMS) algorithm is proposed to alleviate cell under-detection. Experimental results demonstrate that the model achieves 95.7% mAP50 (box) and 95% mAP50 (mask) on the GSCs dataset with an inference speed of 38 frames per second. Moreover, it simultaneously supports dual-modality output for cell confluence assessment and precise counting, providing a reliable automated tool for tumor microenvironment research. Full article

Accurate detection of Alzheimer’s disease (AD) is critical yet challenging for early medical intervention. Deep learning methods, especially convolutional neural networks (CNNs), have shown promising potential for improving diagnostic accuracy using magnetic resonance imaging (MRI). This study aims to identify the most informative combination of MRI slice orientation and anatomical location for AD classification. We propose an automated framework that first selects the most relevant slices using a feature entropy-based method applied to activation maps from a pretrained CNN model. For classification, we employ a lightweight CNN architecture based on depthwise separable convolutions to efficiently analyze the selected 2D MRI slices extracted from preprocessed 3D brain scans. To further interpret model behavior, an attention mechanism is integrated to analyze which feature level contributes the most to the classification process. The model is evaluated on three binary tasks: AD vs. mild cognitive impairment (MCI), AD vs. cognitively normal (CN), and MCI vs. CN. The experimental results show the highest accuracy (97.4%) in distinguishing AD from CN when utilizing the selected slices from the ninth axial segment, followed by the tenth segment of coronal and sagittal orientations. These findings demonstrate the significance of slice location and orientation in MRI-based AD diagnosis and highlight the potential of lightweight CNNs for clinical use. Full article

Internet of Things (IoT)-based solar insecticidal lamps (SIL-IoTs) offer an eco-friendly alternative by merging solar energy harvesting with intelligent sensing, advancing sustainable smart agriculture. However, SIL-IoTs encounter practical challenges, e.g., hardware aging, electromagnetic interference, and abnormal data patterns. Therefore, developing an effective fault detection and diagnosis (FDD) system is essential. In this survey, we systematically identify and address the core challenges of implementing FDD of SIL-IoTs. Firstly, the fuzzy boundaries of sample features lead to complex feature interactions that increase the difficulty of accurate FDD. Secondly, the category imbalance in the fault samples limits the generalizability of the FDD models. Thirdly, models trained on single scenarios struggle to adapt to diverse and dynamic field conditions. To overcome these challenges, we propose a multi-level solution by discussing and merging existing FDD methods: (1) a data augmentation strategy can be adopted to improve model performance on small-sample datasets; (2) federated learning (FL) can be employed to enhance adaptability to heterogeneous environments, while transfer learning (TL) addresses data scarcity; and (3) deep learning techniques can be used to reduce dependence on labeled data; these methods provide a robust framework for intelligent and adaptive FDD of SIL-IoTs, supporting long-term reliability of IoT devices in smart agriculture. Full article

Tuberculosis (TB) is an ancient and re-emerging granulomatous infectious disease that continues to challenge public health. Early diagnosis and prompt effective treatment are crucial for preventing disease progression and reducing both morbidity and mortality. These steps play a vital role in infection control and in lowering death rates at both individual and population levels. Although diagnostic methods have improved sufficiently in recent decades, TB can still present with ambiguous laboratory and imaging features. This ambiguity can lead to diagnostic pitfalls and potentially disastrous outcomes due to delayed diagnosis. In this article, we present a case of TB that was difficult to diagnose. The disease had invaded the mediastinum, right atrium, right coronary artery, and inferior vena cava (IVC), resulting in Budd–Chiari syndrome. This rare presentation created clinical, laboratory, and radiological confusion, resulting in a diagnostic dilemma that ultimately led to open cardiac surgery. The patient initially presented with progressive shortness of breath on exertion and fatigue, which suggested possible heart disease. This suspicion was reinforced by computed tomography (CT) imaging, which showed infiltrative mass lesions predominantly in the right side of the heart, invading the right coronary artery and IVC, with imaging features mimicking angiosarcoma. Although laboratory findings revealed an exudative effusion with lymphocyte predominance and elevated adenosine deaminase (ADA), the Gram stain was negative for bacteria, and an acid-fast bacilli (AFB) smear was also negative. These findings contributed to diagnostic uncertainty and delayed the confirmation of TB. Open surgery with excisional biopsy and histopathological analysis ultimately confirmed TB. We conclude that TB should not be ruled out solely based on negative Mycobacterium bacteria in pericardial effusion or AFB smear. TB can mimic aggressive tumors such as angiosarcoma or lymphoma with invasion of the surrounding tissues and blood vessels. Awareness of the clinical presentation, imaging findings, and potential diagnostic pitfalls of TB is essential, especially in endemic regions. Full article

Gear wear degrades transmission performance, necessitating highly reliable fault diagnosis methods. To address the limitations of existing approaches—where dynamic models rely heavily on prior knowledge, while data-driven methods lack interpretability—this study proposes an integrated bidirectional verification framework combining dynamic modeling and deep learning for interpretable gear wear diagnosis. First, a dynamic gear wear model is established to quantitatively reveal wear-induced modulation effects on meshing stiffness and vibration responses. Then, a deep network incorporating Gradient-weighted Class Activation Mapping (Grad-CAM) enables visualized extraction of frequency-domain sensitive features. Bidirectional verification between the dynamic model and deep learning demonstrates enhanced meshing harmonics in wear faults, leading to a quantitative diagnostic index that achieves 0.9560 recognition accuracy for gear wear across four speed conditions, significantly outperforming comparative indicators. This research provides a novel approach for gear wear diagnosis that ensures both high accuracy and interpretability. Full article