INTRODUCTION/BACKGROUND:Imaging surveillance after Breast-Conserving Surgery (BCS) is essential for identifying potentially curable local or ipsilateral breast tumor recurrence. Mammography (MG) remains the standard surveillance modality, but postoperative scarring, radiotherapy-related distortion, and breast density may reduce its interpretability. Ultrasonography (US) may provide complementary assessment of non-calcified soft-tissue abnormalities. MATERIALS AND METHODS:A systematic search of PubMed, Embase, Web of Science, the Cochrane Library, and Ovid MEDLINE was performed from database inception to August 2025. Eligible studies enrolled patients after BCS or breast-conserving treatment, evaluated MG and/or US for postoperative local or ipsilateral recurrence, and reported extractable diagnostic accuracy data. Study selection, data extraction, and QUADAS-2 assessment were conducted independently by two reviewers. Owing to marked threshold and methodological heterogeneity, modality-specific performance was summarized using paired forest plots, Hierarchical Summary Receiver Operating Characteristic (HSROC) curves, Area Under the Curve (AUC), and leave-one-out sensitivity analyses rather than simple pooled sensitivity/specificity estimates. RESULTS:Nine studies were included. Six studies contributed MG-specific data. MG sensitivity ranged from 0.25 to 1.00 and specificity from 0.51 to 1.00; the HSROC AUC was 0.752, and leave-one-out AUCs ranged from 0.732 to 0.839. Three studies contributed US-specific data. US sensitivity ranged from 0.50 to 0.95 and specificity was consistently high (approximately 0.98 in the included modality-specific studies); the HSROC AUC was 0.985, and leave-one-out AUCs ranged from 0.966 to 0.985. Because direct comparative evidence was limited and heterogeneous, no formal pooled superiority estimate was retained. DISCUSSION:The available evidence suggests that US may be a valuable complementary modality after BCS, particularly for soft-tissue or non-calcified recurrence, whereas MG remains important for detecting microcalcifications and architectural changes. The evidence base is limited by small numbers of studies, variable reference standards, older imaging technology in some reports, and incomplete reporting of recurrence morphology and surveillance timing. CONCLUSION:US appears to have complementary value in postoperative surveillance after BCS, but the current evidence does not justify replacing MG or making an unconditional claim that US is superior. Larger prospective, head-to-head studies using standardized imaging protocols and reference standards are warranted.
OBJECTIVE:This study was designed to explore the diagnostic value of contrast-enhanced multi-slice spiral computed tomography (MSCT) in portal hypertensive gastrointestinal disease (PHGD) among patients with liver cirrhosis. METHODS:A retrospective analysis was performed on cirrhotic patients admitted to our hospital from January 2020 to December 2024. Using endoscopic findings as the reference standard, these patients were divided into the PHGD group and the control group (non-PHGD group). Imaging parameters and serum markers were measured, and differences between the two groups were compared. RESULTS:A total of 69 PHGD patients and 71 non-PHGD patients were enrolled. In the PHGD group, platelet count (PLT), hemoglobin (HB), and total protein (TP) were significantly lower than those in the non-PHGD group (P < 0.05). Regarding imaging indicators, portal vein diameter (PVD), superior mesenteric vein diameter (SMVD), splenic vein diameter (SVD), stomach wall thickness (SD), and stomach wall enhancement thickness (ST) in the PHGD group were significantly greater than those in the non-PHGD group (P < 0.05). The diagnostic efficacy of individual imaging parameters was moderate to good, with AUCs ranging from 0.742 to 0.880. DISCUSSION:Contrast-enhanced MSCT parameters, particularly SMVD (AUC 0.880), PVD (AUC 0.852), and ST (AUC 0.849), demonstrated good diagnostic performance for identifying PHGD in cirrhotic patients. Imaging findings such as venous dilation, splenomegaly, gastrointestinal wall thickening, and ascites were significantly associated with PHGD, suggesting that MSCT may serve as a valuable non-invasive adjunctive tool when endoscopy is unavailable or inconclusive. CONCLUSION:PHGD is a common and significant complication of portal hypertension, but there is a lack of standardized clinical diagnostic criteria. Contrastenhanced MSCT shows considerable potential in assisting the diagnosis of PHGD.
Introduction/ Objective: MRI-based Alzheimer's staging is often modeled as separate tasks, although clinical assessment progresses from screening to more refined staging. This study developed a continual-learning framework for retaining earlier diagnostic knowledge. METHODS:A nested multi-head framework was implemented for sequential learning of related binary classification tasks from structural MRI slices. Because verified subject identifiers were unavailable in the Mendeley dataset, the primary experiments were kept at the level of slice-based methodological analysis. Subject-level external validation was additionally carried out on OASIS-VBM with subject-wise three-fold cross-validation. RESULTS:In the Mendeley slice-level evaluation, the proposed EWC + replay model retained high post-sequence discrimination, with AUC values of 0.976, 0.985, and 1.000 for T1, T2, and T3. The ECE values were 0.034, 0.057, and 0.187. When the model was tested at the subject level on OASIS, AUC values decreased to 0.753, 0.748, and 0.749. This difference shows that the slice-level results should be interpreted cautiously. DISCUSSION:The findings support the use of the proposed framework for studying sequential retention and forgetting in hierarchical AD staging. However, Mendeley results are methodological findings at the slice level and should not be interpreted as patient-level clinical performance. CONCLUSION:This study presents a nested continual-learning framework for hierarchical Alzheimer's disease staging. The framework was tested with ablation, calibration, baseline, and subject-level validation analyses. The results show that it can be used to study retention across related AD staging tasks. Future studies using subject-wise longitudinal cohorts, 3D models, and multimodal data may further improve its clinical relevance.
INTRODUCTION/BACKGROUND:Osteoporosis is a systemic skeletal disorder characterised by low bone mass. The classification typically includes three stages: normal, osteopenia, and osteoporosis. Early detection is essential for effective treatment and reducing fracture risk. This detection is possible through X-ray imaging and clinical assessments, such as measuring T-score and SoS, etc. This study aims to enhance the accuracy of osteoporosis assessment by combining both imaging and clinical data using advanced AI techniques. MATERIALS AND METHODS:A new multimodal approach is proposed for osteoporosis assessment based on imaging and clinical data. The dataset is collected from Begum Shafia Nazir Husain Trust Hospital in Chennai, as well as from various rural clinic camps organized across different regions. For X-ray image classification, a new DL model, the Dynamic Topological Convolutional Transformer (DTCT), is proposed. For clinical data processing, a hybrid model combining Self-Organising Maps (SOM) with Graph Neural Networks (GNN) is applied. The approach is evaluated on a dataset collected from the Begum Shafia Nazir Husain Trust Hospital. RESULTS:The DTCT-based classification achieves the highest accuracy of 96% for multi-class classification. The SOM-combined GNN model achieves a maximum accuracy of 95% in classification. It achieves a robust framework for integrating different data modalities in medical diagnostics. DISCUSSION:The proposed method provides high classification accuracy, improved interpretability, robust performance against noise and variations, and a scalable architecture for efficient adaptation to diverse datasets. CONCLUSION:The proposed approach combines imaging and clinical data for an accurate osteoporosis assessment with superior performance over existing techniques.
BACKGROUND/OBJECTIVE:Fractal Analysis (FA) is a mathematical method used to quantify complex self-similar structures and provides a numerical representation of trabecular bone microarchitecture on radiographic images. This study aimed to evaluate region-specific sex differences in Fractal Dimension (FD) and assess the forensic applicability of FA for sex estimation. MATERIALS AND METHODS:This prospective study analyzed 130 digital panoramic radiographs from a systemically healthy cohort (65 males, 65 females; aged 20-40 years). Six bilateral regions of interest (50 × 50 pixels), representing the condylar, angular, and interdental trabecular regions, were selected, yielding 780 ROIs. FD values were calculated using ImageJ with a standardized box-counting method. Sex-related differences were analyzed using appropriate statistical tests and binary logistic regression. RESULTS:Intra-observer reliability was good to excellent (ICC: 0.767-0.879). Bilateral analysis showed significant sex-related differences in the left condylar (p < 0.001) and left interdental regions (p = 0.048). After bilateral averaging, only the condylar region remained significant. Females showed lower mean condylar FD values than males (1.381 ± 0.094 vs. 1.423 ± 0.091; p = 0.012). Logistic regression identified condylar FD as a significant predictor of sex (OR = 0.61, 95% CI: 0.41-0.91; p = 0.014), although discrimination was limited (AUC = 0.628; accuracy = 58.5%). DISCUSSION:Sex-related differences in mandibular trabecular architecture appeared to vary across anatomical regions, with the condylar region showing the most consistent discriminatory ability. This finding may be related to the condyle's greater susceptibility to biomechanical adaptation and hormonal influences, both of which can affect trabecular organization. CONCLUSION:FA of panoramic radiographs may serve as a useful adjunct in forensic sex estimation. However, its moderate discriminative performance suggests that it should be used in combination with complementary analytical methods rather than as an independent diagnostic tool.
INTRODUCTION/OBJECTIVE:Many other diseases can produce a similar pattern; therefore, the diagnosis of IPF is made by exclusion, based on the absence of alternative pathologies. To differentiate IPF from other pathologies, the study developed a deep learning model with an attention mechanism to improve performance. METHODS:This retrospective, single-center study included 96 patients (46 IPF and 50 non-IPF) with a typical UIP pattern on HRCT. The data were split at the patient level into training (70%), validation (15%), and test (15%) sets using stratified randomization (seed = 42). Squeeze-and-Excitation (SE) blocks were incorporated after each convolutional stage of VGG-16 to enable channel-wise feature recalibration. The model was trained using the AdamW optimizer (learning rate = 1×10-4, batch size = 32, 50 epochs, early stopping with a patience of 7), with augmentation applied only to the training set. Test predictions were generated from original, unaugmented slices, and patient-level labels were determined via majority voting. Gradient-weighted Class Activation Mapping (Grad-CAM) was applied post hoc for heatmap visualization. Model performance was evaluated using accuracy, sensitivity, specificity, F1-score, Area Under the Curve (AUC), and Cohen's kappa. RESULTS:The SE-VGG-16 model achieved a patient-level accuracy of 87.2%, sensitivity 86.1%, specificity 88.4%, F1-score 0.870, and AUC 0.91. Ablation studies demonstrated a 4.1% accuracy improvement over baseline VGG-16. Grad-CAM heatmaps consistently highlighted subpleural reticular opacities and honeycombing regions, aligning with established radiological criteria. DISCUSSION:The study has successfully distinguished IPF from other interstitial diseases with a high degree of accuracy. Despite the existence of numerous studies in the literature on the differential diagnosis of IPF, research on deep learning and studies combining deep learning with attention maps are quite limited. Unlike traditional deep learning models, the use of an attention mechanism has enabled the model to focus on pathological regions, thereby producing more reliable results. The findings present a potential approach that could be used in clinical decision support systems. CONCLUSION:Based solely on radiological images, the attention-enhanced VGG-16 model achieved high accuracy in the differential diagnosis of IPF.
INTRODUCTION:Diffusion- and perfusion-based imaging, including Apparent Diffusion Coefficient (ADC) and CT Perfusion (CTP), are standard tools for evaluating ischemic stroke but primarily reflect structural and hemodynamic changes. They provide limited insight into tissue metabolism. Amide Proton Transfer-weighted (APTw) imaging enables noninvasive assessment of pH-related metabolic alterations and may offer complementary diagnostic and prognostic information. MATERIALS AND METHODS:In this prospective study, 54 patients with hyperacute or acute ischemic stroke (mean age 64.39 ± 10.91 years; 41 males) were enrolled. Correlations between APTw and conventional imaging parameters, including ADC, Cerebral Blood Volume (CBV), Cerebral Blood Flow (CBF), and Time to Peak (TTP), were analyzed. Kaplan-Meier analysis assessed 18-month outcomes. Time-dependent Receiver Operating Characteristic (ROC) analysis, decision curve analysis (DCA), calibration curves, and Clinical Impact Curves (CIC) were used to evaluate prognostic performance. RESULTS:APTw values were significantly correlated with CBV (r = 0.6886), CBF (r = 0.6702), ADC (r = 0.6565), and TTP (r = -0.6519) (all p < 0.05). Kaplan-Meier analysis demonstrated significant differences in 18-month outcomes between APTw-based subgroups (log-rank p < 0.05). APTw achieved higher AUCs at 6, 12, and 18 months (0.821, 0.831, and 0.877) than ADC and CTP parameters. DCA, calibration, and CIC analyses confirmed stable predictive performance. DISCUSSION:APTw showed strong concordance with diffusion- and perfusion-based biomarkers while providing complementary metabolic information, suggesting exploratory comparative prognostic relevance in ischemic stroke. Conclusions APTw imaging offers complementary metabolic insight and may enhance diagnostic and prognostic evaluation in ischemic stroke.
INTRODUCTION/BACKGROUND:The diagnosis of benign and malignant primary epithelial tumors of the lacrimal gland (PET-LG) is challenging. This study was designed to obtain a pretreatment diagnosis of adenoma and Adenoid Cystic Carcinoma (ACC) of the LG based on clinical and imaging features. MATERIALS AND METHODS:A retrospective analysis was conducted on 18 patients with histologically confirmed PET-LG (6 adenomas and 12 ACCs). Demographic data, clinical outcomes, and imaging findings were collected. MR findings included lesion size, signals, contrast enhancement patterns, DWI/ADC values, and evidence of orbital involvement. For ACC cases, CT findings were compared with MRI findings using pathological evidence as the reference. Statistical analyses were performed to assess differences between the adenoma and ACC groups. RESULTS:ACCs showed a more aggressive imaging pattern than adenomas. Compared with adenomas, ACCs were larger at presentation and more frequently demonstrated vivid contrast enhancement, orbital wall destruction, and extraocular muscle involvement. Among these findings, orbital wall invasion and extraocular muscle involvement were the most robust imaging features associated with ACC. In contrast, adenoma more commonly showed smooth orbital wall erosion or remodeling. For assessing bone invasion in ACC, MRI showed higher sensitivity and accuracy than CT, whereas CT remained useful for evaluating cortical bone changes. DISCUSSION:Imaging findings can be used to diagnose benign and malignant PET-LG. MRI plays a pivotal role in identifying these imaging features for clinical staging, but CT remains valuable for detecting bone erosion. CONCLUSION:We suggest applying both CT and MR of the orbit as imaging techniques to enhance pretreatment planning.
Optical Coherence Tomography Angiography (OCTA) provides depth-resolved imaging of the retinal vascular system without invasion. Its rapid and non-invasive characteristics make it a promising tool for pediatric retinal imaging. Pediatric myopia and amblyopia can affect visual development and long-term visual prognosis. Early identification and dynamic monitoring are of great clinical significance. This paper retrieved English literature published from January 2006 to May 2026 in PubMed and Web of Science, and reviewed the application of OCTA in pediatric myopia and amblyopia. Existing studies show that pediatric myopia is characterized by decreased retinal vascular density (VD), thinner choroidal thickness (CT), and changes in choroidal perfusion parameters; amblyopia is mainly manifested as decreased retinal VD, but results for the foveal avascular zone (FAZ) and choroidal-related indicators are inconsistent. At present, there are still some deficiencies in the research, such as limited sample size, non-uniform equipment and scanning protocols, and lack of long-term follow-up. OCTA is expected to become an important auxiliary tool for the diagnosis, evaluation, and monitoring of treatment for childhood myopia and amblyopia.
Introduction: It is imperative to perform imaging evaluation following endovascular coil treatment for cerebral aneurysms. Noninvasive vascular assessment using Magnetic Resonance Angiography (MRA), specifically Time-of-Flight (TOF) MRA, is commonly performed. Thus, this study assessed the diagnostic performance of Pointwise Encoding Time Reduction with Radial Acquisition (PETRA) MRA in patients who underwent simple coil embolization. Methods: Occlusion status, overall image quality, and treated-site image quality were compared across Digital Subtraction Angiography (DSA), TOF MRA, and PETRA MRA obtained during the same follow-up session after simple coil embolization of intracranial aneurysms. Results: Overall, 366 patients (median age: 61 ± 13 years) with 373 aneurysms underwent DSA, TOF MRA, and PETRA MRA following simple coil embolization. Occlusion status on PETRA MRA (85 cases, 91.4%) revealed stronger consistency with DSA than with TOF MRA (60 cases, 64.5%). Moreover, PETRA MRA demonstrated higher sensitivity for detecting residual neck and aneurysm (Raymond II and IIIa) compared with TOF MRA (Raymond II: 89% vs. 69%, p < 0.05; Raymond IIIa: 93% vs. 50%, p < 0.05). Notably, no significant differences were observed in overall or treated-site image quality between the techniques (p = 0.9). Discussion: PETRA MRA provides better visualization of residual flow within the coil mass and exhibits greater agreement with DSA than with TOF MRA. It also serves as an effective follow-up imaging modality, preserving image quality while enhancing diagnostic accuracy. Conclusion: PETRA MRA offers superior diagnostic performance compared with TOF MRA for follow-up evaluation following simple coil embolization of cerebral aneurysms.
Objective: We aimed to investigate the efficacy of transcervical esophageal sonoelastography in discriminating between patients with and without Reflux Esophagitis (RE). Methods: Patients who underwent an esophagogastroduodenoscopy following a transcervical sonoelastographic examination in a tertiary center were included. The thickness and Shear Wave Velocity (SWV) of the submucosa-mucosa, muscularis mucosae, and total esophageal wall were measured by using sonoelastography. Patients were classified into RE-positive (grades A-D according to The Los Angeles Classification) and REnegative groups based on endoscopic findings. The elastosonographic features of RE (-) and RE (+) patients were compared statistically Results: Among the 503 included patients, 129 (25.6%) were RE (+) [(80 grade A (62%), 28 grade B (21.7%), 12 grade C (9.3%), 9 grade D (7%)] and 374 were RE (-). A 31.7% positive correlation between age and Total Wall Thickness (TWT) and a 21% negative correlation between TWT and submucosa-mucosa SWV were detected. The thickness of submucosa-mucosa, muscularis propria, and total wall were significantly higher in RE (+) patients compared to RE (-) patients (p < 0.001 for all). The SWVs of submucosa-mucosa, muscularis propria, and total wall were significantly lower in RE (+) patients compared to RE (-) patients (p < 0.001 for all). A cut-off value of 2.35 mm for TWT yielded 67% sensitivity and 57% specificity, while a cut-off of 2.08 m/s for total wall SWV yielded 79% sensitivity and 81% specificity for the diagnosis of RE. Discussion: The significant increase in cervical esophageal wall thickness in RE (+) is consistent with the literature. However, the markedly decreased stiffness is a novel finding demonstrated by this study. Conclusion: The evaluation of esophageal thickness and stiffness using sonoelastography may be added to routine neck ultrasound scans for the prediction of RE.
INTRODUCTION:Congenital Long QT Syndrome (LQTS) is a common cause of potentially life-threatening ventricular arrhythmias and sudden cardiac death in infants and children. Fetal Magnetocardiography (fMCG) is available in only a few specialized centers worldwide and offers accurate measurements of the QT and PR intervals. The aim of this study was to assess the diagnostic accuracy of PW Doppler echocardiographic measurement of fetal QT and PR intervals compared with the available normative data for fMCG. METHODS:576 pregnant women with singleton pregnancies (16-38 weeks' gestation) were studied in this cross-sectional study between 2022 and 2023 with detailed fetal echocardiograms. Standardized pulsed-wave Doppler techniques were used to measure the PR and QT intervals. Results were compared with published normative data for fMCG, adjusted for gestational age, and were also compared with contemporaneous fMCG in a subset of 42 fetuses. RESULTS:Mean QT interval (echo) was 270.0 ± 38.8 and mean PR interval was 119.3 ± 16.2. In the paired subset (n = 42), strong agreement was observed (PR ICC = 0.89 [95% CI: 0.81-0.94]; QT ICC = 0.82 [95% CI: 0.72-0.89]), with mean biases of +2.1 ms and +6.8 ms on Bland-Altman analysis, respectively. Sensitivity and specificity for the full cohort for prolonged intervals (> 95th percentile of GA-adjusted fMCG norms) were 85% and 78% (PPV 68%, NPV 90%). There were no clinically significant differences for PR interval or fetal heart rate, and there was no significant correlation between measured intervals and gestational age (p > 0.05). DISCUSSION:Pulsed-wave Doppler echocardiography is a reliable method for measuring fetal QT and PR intervals and is in good agreement with fMCG. CONCLUSION:It is a practical, non-invasive, and easily accessible substitute for fMCG when it is not available for prenatal screening of cardiac repolarization and conduction abnormalities.
Introduction: Accurate classification of apical hypertrophic cardiomyopathy (ApHCM) subtypes is challenging due to morphological variability and overlapping phenotypes. Conventional echocardiography provides limited visualization of the apex. Artifacts induced during left ventricular opacification (LVO) complicate diagnostic interpretation. A deep learning-based framework enhances image quality and improves subtype classification. Materials and Methods: In this work, a deep learning-based framework is used for ApHCM subtype classification. Apical four-chamber end-diastolic frames from 3,200 individual patients were extracted from the EchoNet-Dynamic Dataset. Two cardiology experts manually annotated images into pure ApHCM, relative ApHCM, mixed ApHCM, and normal classes, based on apical wall thickness and morphological characteristics, using a computer vision annotation tool. A deep learning pipeline integrated multilevel graph-based adaptive particle swarm optimization with a deep denoised convolutional neural network (MG-APSO-DnCNN) to suppress reverberation and clutter artifacts from LVO echocardiograms. Enhanced images were then segmented using a U-Net-based levelset model to delineate the left ventricular (LV) apex. Morphological and LV wall features were extracted from the segmented region, and a graph isomorphism network (GIN) was trained to capture both local hypertrophic patterns and global ventricular morphology for subtype classification. The framework was designed to distinguish among pure ApHCM, relative ApHCM, and mixed ApHCM. Results: The framework achieved a classification accuracy of 96.2%, with a precision, recall, and F1-score of approximately 95%. Cross-validation results indicate stable performance (95.8% ± 0.4%, p < 0.001), and ablation experiments confirmed the contribution of each pipeline component. Discussion: By combining denoising, segmentation, and graph-based learning, the framework addressed limitations caused by LVO artifacts and improved the recognition of subtle ApHCM subtypes. These results demonstrate the clinical potential of integrating morphological feature extraction with deep learning in echocardiography. Conclusion: The framework integrating MG-APSO-DnCNN and GIN enables accurate and robust ApHCM subtype classification, supporting cardiologists in early diagnosis, patient risk stratification, and treatment planning.
Background: Appendiceal neuroendocrine neoplasms (ANENs) are rare tumors of the appendix, most of which are low-grade malignancies. They are often discovered incidentally during surgery for acute appendicitis. This report presents a case of a 29-year-old male with typical manifestations of acute appendicitis. Preoperative imaging and laboratory tests were consistent with acute appendicitis with abscess formation, but postoperative pathology revealed an unexpected appendiceal neuroendocrine tumor. The aim of this report is to improve clinicians’ understanding, diagnostic accuracy, and therapeutic decision-making for this condition. Case Presentation: A 29-year-old male patient presented with a one-day history of migratory right lower quadrant abdominal pain. Physical examination revealed tenderness and rebound tenderness at McBurney’s point, and a positive Rovsing’s sign. Laboratory tests showed elevated inflammatory markers (procalcitonin, interleukin-6, ferritin), consistent with acute appendicitis. Ultrasound and CT imaging revealed a thickened appendix with periappendiceal exudation, consistent with acute appendicitis with abscess formation. Emergency laparoscopic appendectomy was performed. Postoperative pathology confirmed an appendiceal neuroendocrine tumor (G1) with serosal and perineural invasion, without lymphovascular invasion, and with negative margins. The Ki-67 index was 2%+ (G1 according to the 2019 WHO classification). The patient recovered uneventfully, and a 3-month follow-up CT showed no signs of recurrence or metastasis. Conclusion: This case suggests that simple appendectomy with close surveillance might be considered an individualized strategy for selected patients with intermediate-risk G1 tumors measuring 1-2 cm with serosal or perineural invasion but without lymphovascular invasion and with negative margins. However, further validation with larger studies and longer follow-up is needed.
Background: The most common symptoms of transverse-sigmoid sinus dural arteriovenous fistula (TS-S DAVF) are pulsatile tinnitus and headache. The clinical characteristics lack specificity and may lead to misdiagnosis or delayed diagnosis. The noninvasive and convenient nature of neurovascular ultrasound has made it an important screening tool for lesions. This paper reports the use of transcranial Doppler (TCD) ultrasonography for the preoperative screening and postoperative follow-up of a patient diagnosed with a transverse-sigmoid sinus dural arteriovenous fistula (TS-S DAVF). Case Presentation: A 62-year-old lady was admitted with a six-month history of headache and tinnitus on the left side. A structured TCD-based approach was used to evaluate intracranial and extracranial hemodynamics. The protocol included targeted probe positioning based on anatomical landmarks, spectral waveform analysis, and the superficial temporal artery tap maneuver. The preoperative TCD results indicated hemodynamic alterations with increased flow and decreased resistance in the left common carotid artery, external carotid artery, and occipital artery. The presence of a Cognard Type III DAVF in the left transverse sinus-sigmoid sinus was confirmed through DSA. The occipital artery and the posterior branch of the middle meningeal artery were the key feeding arteries with cortical vein drainage to the sigmoid sinus. After endovascular embolization via the venous approach, follow-up TCD showed normal hemodynamics in the previously affected arteries, consistent with complete resolution of the patient’s clinical symptoms. Conclusion: A structured TCD-based approach may help identify shunt-related hemodynamic abnormalities in suspected TS-S DAVF and support postembolization follow-up. Cerebral DSA remains necessary for definitive diagnosis and treatment planning.
In recent years, medical imaging has become an important tool for diagnosing diseases and disorders in healthcare. Advanced imaging technologies are being developed for non-invasive and early detection of diseases and disorders. Analyzing medical images by clinical experts is very expensive. To overcome these challenges, developing automated methods provides an effective solution. Consequently, for processing and analyzing medical images, researchers have adopted the emerging Deep Learning (DL) technologies. It has proven effective across several industries, most notably in healthcare. Even so, it has two significant limitations, such as the training cost and the large amounts of labeled data required. To reduce these limitations, Transfer Learning (TL) and Deep Learning (DL) have been integrated to create Deep Transfer Learning (DTL). This reduces the need to start from scratch and eliminates dependencies by leveraging knowledge from a source task to a target task during training, using fewer datasets. This review addresses the definitions, concepts, modalities, tasks, and techniques of DTL, along with public and private datasets used as source and target data in network-based medical imaging approaches. It also categorizes the last seven years of research by human anatomical area. It offers readers comprehensive coverage of technological advancements, future research directions, and challenges. It also reviews DTL methods by discussing those that have been applied, including Federated Learning (FL) for DL. Empirical evidence from recent studies demonstrates that fine-tuning and network-based DTL strategies, including federated learning, consistently enhance diagnostic accuracy, robustness, and generalization across multiple medical imaging modalities, particularly in data-limited clinical scenarios.
Introduction: Chest X-ray (CXR) still represents the most performed radiological examination, but its interpretation may differ among readers. In the last years, several automatic detection tools have been developed to assist radiologists in CXR interpretation. We aimed to evaluate how artificial intelligence (AI)-based software may increase the performance of young radiologists in CXR interpretation compared to that of an experienced CXR radiologist Materials and Methods: 500 CXRs were selected to generate a well-balanced dataset. For each patient, the CXR reading was conducted by an AI-based software (Lunit INSIGHT CXR) and, independently, by two general radiologists with less than one year of experience. Furthermore, after three months, the radiologists reviewed all the examinations with the assistance of the software. The following CXR findings were searched: atelectasis, calcification, cardiomegaly, consolidation, fibrosis, nodules, mediastinal widening, pleural effusion, pneumoperitoneum, and pneumothorax. Sensitivity and specificity were computed compared to an expert reader allowed to use AI assistance. Chi-squared tests were used to compare the readings. Results: A total of 600 findings were identified by the senior radiologist. The sensitivity of young radiologists without AI was significantly lower than that of AI (p < 0.001), while the specificity was significantly higher for both (p < 0.001). With the assistance of AI, overall sensitivity increased in both radiologists (p < 0.001), while specificity decreased (p ≤ 0.004). Discussion: AI software achieves very high sensitivity, minimizing the number of false negatives as much as possible. As a counterpart, there could be a nonnegligible risk of overdiagnosis. Conclusion: AI-based software assistance can yield a potential enhancement in sensitivity for young radiologists in the interpretation of CXRs. However, there might be a reduction in specificity.
BACKGROUND:Osteoporosis is a widespread skeletal disorder characterized by reduced bone density and increased fracture risk. Early detection is critical to preventing disability and healthcare burden. METHODS:A systematic search was conducted in Google Patents for documents filed between June 2022 and June 2025 that applied artificial intelligence or machine learning to osteoporosis detection using X-ray imaging. Patents were screened in three stages by independent reviewers according to predefined inclusion criteria. Eligible patents were analyzed and categorized according to technical objectives and methodological features. RESULTS:19 patents met the inclusion criteria. Most originated from Asian countries. Deep learning architectures, primarily convolutional and generative models, were central to innovations in bone density estimation, fracture risk prediction, and image preprocessing. Several patents introduced novel approaches such as cross-modal image fusion, anatomical feature transfer, and data augmentation. Persistent challenges included limited interpretability, variability across datasets, and insufficient clinical validation. DISCUSSION:The identified patents demonstrate growing sophistication in AI/ML-based X-ray analysis, reflecting convergence toward standardized workflows for image enhancement, feature extraction, and risk prediction. Nevertheless, important gaps remain, including limited interpretability, variability across imaging conditions, and insufficient clinical validation. The reliance on complex architectures and synthetic data also raises concerns regarding generalizability and real-world deployment. Addressing these challenges will be essential to ensure reliable, scalable translation of AI/ML-driven osteoporosis diagnostics into routine clinical practice. CONCLUSION:The recent patents demonstrate rapid progress in artificial intelligence-driven osteoporosis diagnostics using radiographic imaging. Advances in transparency, model robustness, and clinical validation are essential to enable safe and effective translation of these technologies into clinical practice.
In the originally published version of this article, the elocator number was published incorrectly due to a technical error. This error has now been corrected in the online version of the article [1]. We apologize for any inconvenience caused and appreciate the opportunity to rectify this matter. The original article can be found online at: https://www.benthamscience.com/article/124894 Original: e1875660325101501 Corrected: e290622206508
In the published version of this article [1], the annotations of corresponding authors and first co-authors have been updated on the request of the author. This has now been corrected. The original article can be found online at: https://www.benthamscience.com/article/150901 ORIGINAL: Ying Long1,*, Zhao-ping Chen1,*, Lin-hui Wang2, Xue-qing Liao1, Ming Guo3 and Zhong-qing Huang1,* CORRECTED: Ying Long1,#, Zhao-ping Chen1,#, Lin-hui Wang2, Xue-qing Liao1, Ming Guo3,* and Zhong-qing Huang1,*
