Category Ranking
98%
Total Visits
921
Avg Visit Duration
2 minutes
Citations
20
Article Abstract
Background: Accurate measurement of endometrial thickness (ET) using transvaginal ultrasound (TVUS) imaging is essential for diagnosing various gynecological conditions. However, manual ET measurement remains challenging, especially for junior physicians, due to variability in image quality and patient characteristics.
Methods: A prospective observational study was performed using a dataset of 976 uterine ultrasound videos (82,063 images) measured in 2014-2019 in Tongji Hospital, Huazhong University of Science and Technology. We developed EndoUSScan, a comprehensive system for automated image selection and keyframe identification. The system incorporates MSNet, an improved DenseNet169-based system, to select candidate images with accurate endometrial representation. We also designed a keyframe detection system to assist junior medical staff in identifying frames with the largest ET from the candidate images. Comparative evaluations involved six junior sonographers, who assessed both speed and accuracy.
Findings: MSNet achieved an accuracy of 94.7% and a specificity of 96.7% in selecting candidate images, outperforming conventional models including ResNet50, ResNet101, DenseNet121, and DenseNet169. The automatically selected keyframes were consistent with the expert-defined gold standard. Compared with manual procedures by junior sonographers, EndoUSScan significantly improved both the speed and accuracy of keyframe selection.
Interpretation: This study presents the first fully automated and clinically validated system for keyframe detection in TVUS videos to support ET measurement. By standardizing the image selection process and assisting junior sonographers, EndoUSScan enhances diagnostic efficiency and accuracy, ultimately contributing to improved patient care.
Funding: This study was funded by the National Key Research and Development Program of China (grant number 2022YFC2704100) and Knowledge Innovation Program of Wuhan -Basic Research (No. 2023020201010041).
Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.1016/j.cmpb.2025.108957 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Protection and Analysis of Intangible Cultural Heritage Videos Based on Keyframe Extraction and Adaptive Weight Assignment.
PLoS One
August 2025
School of Fashion Art, Shaanxi University of International Trade & Commerce, Xi'an, China.
To preserve the intangible cultural heritage digitally and effectively manage and analyze the intangible cultural heritage video data, the research creatively employs target recognition algorithms and keyframe extraction to perform video extraction and analysis. The keyframe extraction and target detection model is constructed with the help of shot boundary detection, feature pyramid network, and attention mechanism. The experimental results revealed that the designed keyframe extraction model outperformed all the other methods, achieving an accuracy rate of 0.
View Article and Find Full Text PDFKey frame extraction based abnormal vehicle identification technique using statistical distribution analysis.
Sci Rep
August 2025
Research Associate, Vivit Academy, Tirunelveli-627008, Tamilnadu, India.
In recent days, due to potential growth of vehicle usage, the researchers have to concentrate on abnormal vehicle identification areas to provide solutions to avoid accidents. Though many vehicle identification works have been done by applying machine and deep learning approaches, still there is some problem with handling repetition frames and identifying the abnormal vehicles among vehicles in a camera. To overcome these challenges, this paper introduces KFEAVI (Key Frame Extraction based Abnormal Vehicle Identification) technique that uses statistical feature extraction technique and constrained angular second moment technique.
View Article and Find Full Text PDFEndoUSScan: Keyframe detection in transvaginal ultrasound imaging for measuring endometrial thickness.
Comput Methods Programs Biomed
November 2025
Department of Obstetrics and Gynecology, Tongji Hospital, Huazhong University of Science and Technology, Wuhan, Hubei, China. Electronic address:
Background: Accurate measurement of endometrial thickness (ET) using transvaginal ultrasound (TVUS) imaging is essential for diagnosing various gynecological conditions. However, manual ET measurement remains challenging, especially for junior physicians, due to variability in image quality and patient characteristics.
Methods: A prospective observational study was performed using a dataset of 976 uterine ultrasound videos (82,063 images) measured in 2014-2019 in Tongji Hospital, Huazhong University of Science and Technology.
Hyperparameter optimization of YOLO models for invasive coronary angiography lesion detection and assessment.
Comput Biol Med
September 2025
Instituto de Investigación Biomédica de Málaga y Plataforma en Nanomedicina-IBIMA Plataforma BIONAND, Málaga TechPark, Campanillas, 29590, Spain; Department of Endocrinology and Nutrition, Virgen de la Victoria University Hospital, Málaga, 29010, Spain; Centro de Investigación Biomédica en Re
Coronary artery disease (CAD) remains the leading cause of mortality, creating an urgent need for reproducible, image-based decision support. Although YOLOv8-based detectors underpin much of today's state-of-the-art stenosis detection, their accuracy is sensitive to dozens of interacting hyperparameters. We therefore perform a systematic study of hyperparameter optimizers for YOLO-style models, pairing YOLOv8 and its Double Coordinate Attention (DCA) variant with three model-based engines: Covariance-Matrix-Adaptation Evolution Strategy (CMA-ES), Tree-structured Parzen Estimator (TPE), and a Gaussian-process sampler; and contrasting them with Random Search and the mutation-only routine that serves as the default optimizer in the ultralytics package.
View Article and Find Full Text PDFApproach to Semantic Visual SLAM for Bionic Robots Based on Loop Closure Detection with Combinatorial Graph Entropy in Complex Dynamic Scenes.
Biomimetics (Basel)
July 2025
School of Information Science and Technology, Yunnan Normal University, Kunming 650500, China.
In complex dynamic environments, the performance of SLAM systems on bionic robots is susceptible to interference from dynamic objects or structural changes in the environment. To address this problem, we propose a semantic visual SLAM (vSLAM) algorithm based on loop closure detection with combinatorial graph entropy. First, in terms of the dynamic feature detection results of YOLOv8-seg, the feature points at the edges of the dynamic object are finely judged by calculating the mean absolute deviation (MAD) of the depth of the pixel points.
View Article and Find Full Text PDF