Related Experiment Video
Updated: Jan 14, 2026
Simultaneous Brightfield, Fluorescence, and Optical Coherence Tomographic Imaging of Contracting Cardiac Trabeculae Ex Vivo
Published on: October 2, 2021
Multi-Scale Fusion for Real-Time Image Observation and Data Analysis of Athletes after Soft Tissue Injury
Jinhui Li1, Yang Yu2, Jiaxing Han1
1College of Physical Education, Qiqihar University, Qiqihar 161006, Heilongjiang, China.
Abstract:
Objective: To address insufficient segmentation accuracy in athletes' soft tissue injury analysis, this study proposes an enhanced Swin-Unet model with multiscale feature fusion via the FPN (Feature Pyramid Network) and an adaptive window selection mechanism for dynamic receptive field adjustment.
Methods: A weighted hybrid loss function integrating Dice Loss, Cross-Entropy Loss, and boundary auxiliary loss optimizes segmentation precision and boundary recognition.
Results: Evaluated on the OAI-ZIB dataset using 10-fold cross-validation, the model achieves a DSC (Dice Similarity Coefficient) of 0.978, outperforming baseline Swin-Unet and mainstream architectures. Superior performance is demonstrated in IoU (Intersection over Union) (0.968) and boundary Hausdorff distance (3.21), alongside significantly reduced diagnosis time (6.0 minutes vs. 16.8 minutes manually).
Conclusion: This framework enhances real-time medical imaging analysis for athlete injuries, offering improved accuracy, efficiency, and clinical utility in soft tissue segmentation tasks.
.
