Purpose: This study aims to evaluate the effectiveness of advanced deep learning models, specifically YOLOv8 and EfficientNetV2, in detecting meniscal tears on magnetic resonance imaging (MRI) using a relatively small data set.Method: Our data set consisted of MRI studies from 642 knees—two orthopaedic surgeons labelled and annotated the MR images. The training pipeline included MRI scans of these knees. It was divided into two stages: initially, a deep learning algorithm called YOLO was employed to identify the meniscus location, and subsequently, the EfficientNetV2 deep learning architecture was utilized to detect meniscal tears. A concise report indicating the location and detection of a torn meniscus is provided at the end.Result: The YOLOv8 model achieved mean average precision at 50% threshold (mAP@50) scores of 0.98 in the sagittal view and 0.985 in the coronal view. Similarly, the EfficientNetV2 model obtained area under the curve scores of 0.97 and 0.98 in the sagittal and coronal views, respectively. These outstanding results demonstrate exceptional performance in meniscus localization and tear detection.Conclusion: Despite a relatively small data set, state‐of‐the‐art models like YOLOv8 and EfficientNetV2 yielded promising results. This artificial intelligence system enhances meniscal injury diagnosis by generating instant structured reports, facilitating faster image interpretation and reducing physician workload.Level of Evidence: Level III.

To evaluate a deep learning pipeline using YOLOv11 for segmentation and a custom CNN for classification to automatically detect and assess chondromalacia patella on axial knee MRI, aligning with expert clinical evaluation. A dataset of 650 axial knee MRIs was analyzed. YOLOv11 segmented the patellofemoral region, and a custom CNN classified chondromalacia. Performance was assessed using segmentation accuracy, classification accuracy, confidence scoring, and Grad-CAM-based visual explainability. The CNN achieved a test accuracy of 82.30% on 113 images, with an AUC of 0.87, indicating promising but preliminary discriminative ability. Grad-CAM maps showed reasonable agreement with expert interpretation. The proposed YOLOv11-CNN pipeline demonstrated promising accuracy and may provide a potentially useful and interpretable solution for the detection and segmentation of chondromalacia patella on MRI, with the possibility of enhancing efficiency and consistency in orthopedic radiology workflows after further validation.

Hong Kong Journal of Radiology

Aim: Surgical management of chronic frontal sinusitis poses significant challenges due to the highly variable anatomy of the frontal sinus. While endoscopic sinus surgery (ESS) is commonly employed, it has limitations in accurately diagnosing and excising frontal sinus pathology, particularly in visualizing drainage pathways. Minitrephination serves as a less invasive alternative; however, anatomical variation often complicates the identification of a reliable entry point. This study aims to determine optimal anatomical landmarks for frontal sinus minitrephination using three-dimensional computed tomography (CT) imaging, with the goal of improving surgical accuracy and safety. Methods: A total of 432 patients (864 frontal sinuses) were evaluated using 64-slice multidetector CT (MDCT) scanning. Frontal sinus dimensions and maximum depth were measured relative to predefined X and Y axes on axial, coronal, and sagittal images. Anatomical variations, including hypoplasia, were also recorded. Results: Hypoplastic frontal sinuses were identified in 4.4% of right and 3.5% of left sinuses. The average depth of the frontal sinus was 14.43 ± 4.73 mm on the right and 14.48 ± 5.43 mm on the left. No significant correlation was found between sinus measurements and patient age. However, male patients exhibited significantly greater Y-coordinate values and frontal sinus depth compared to female patients (p < 0.05). Conclusion: This study quantitatively demonstrates the extent of anatomical variation in the frontal sinus region and highlights sex-related differences. The findings support the implementation of precise CT-based preoperative planning for frontal sinus minitrephination. Such individualized anatomical mapping may reduce procedural complications and improve surgical outcomes in patients undergoing treatment for chronic frontal sinusitis.

To conduct a preliminary single-center feasibility study of a YOLO-based deep-learning model for automated detection of lumbar disc herniation at the L4

Purpose: Atherosclerosis is a systemic vascular disease that can affect visceral arteries, including the superior mesenteric artery (SMA), potentially leading to mesenteric ischemia with high morbidity and mortality rates. Despite its clinical importance, SMA involvement remains under‐reported. This study aimed to evaluate the prevalence, morphological characteristics, anatomical distribution, and associated risk factors of SMA atherosclerotic plaques using multidetector computed tomography (MDCT). Materials and Methods: This retrospective study was conducted using data collected from a University Hospital in Turkey. This study focused on patients referred to the Department of Radiology and examined imaging data obtained between January 2008 and December 2010. Information was gathered by examining MDCT scans, patient reports, laboratory findings, and medical records of 15,000 patients admitted to our University Hospital over a designated period. The anatomical distribution of the aneurysms within the SMA (proximal, middle, distal, or diffuse) was recorded. The presence of concomitant atherosclerosis in other visceral arteries and cardiovascular risk factors, including hypertension (HT), diabetes mellitus (DM), coronary artery disease (CAD), and abnormal lipid profiles, was evaluated. Statistical analysis was performed using SPSS (Version 10.0), and associations were assessed using Pearson’s chi‐square test, with a significance level set at p < 0.05. Results: The prevalence of SMA atherosclerosis was 0.86% in the studied population. Clinical data were available for 102 patients in this study. HT, DM, and CAD were present in 73.5%, 49%, and 44.1% of the patients, respectively. Calcified plaques were the most common type (55.8%), followed by mixed (28.7%) and soft plaques (15.5%).Plaques were predominantly located in the proximal (29.4%) and proximal–middle (22.4%) segments of the SMA, followed by the middle (14.7%), middle–distal (14.7%), distal (4.6%), and diffuse segments (13.9%). All patients demonstrated concomitant abdominal aortic atherosclerosis. Atherosclerotic involvement of other visceral arteries was frequently observed in this study. The renal arteries were the most commonly affected, with bilateral involvement in 49.2% of the patients and unilateral calcified plaques in 23.4% of the patients. Celiac trunk atherosclerosis was detected in 41.8% of patients, splenic artery involvement in 12.4%, and inferior mesenteric artery (IMA) involvement in 27.1%, making it the least affected artery. Lipid profile analysis revealed elevated total cholesterol levels in 38% of the patients and low HDL levels in 25% of the patients although no significant correlation was found between lipid levels and plaque characteristics or distribution. Conclusion: Contrast‐enhanced MDCT is a reliable, noninvasive imaging modality for detecting and characterizing SMA atherosclerosis, allowing for an accurate assessment of plaque morphology, distribution, and associated vascular involvement. Although relatively uncommon, SMA atherosclerosis is strongly associated with systemic atherosclerotic disease and major cardiovascular risk factors. Routine evaluation of mesenteric vessels during abdominal CT examinations may facilitate early diagnosis and improve clinical outcomes by enabling timely management of patients at risk for mesenteric ischemia.