Related Experiment Video
Updated: Jul 17, 2025
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
Published on: November 30, 2022
Intelligent contour extraction approach for accurate segmentation of medical ultrasound images
Tao Peng1,2,3, Yiyun Wu4, Yidong Gu5
1School of Future Science and Engineering, Soochow University, Suzhou, China.
Insights
This study presents an intelligent method for accurate organ contour extraction in ultrasound images, improving diagnostic capabilities. The novel approach enhances precision for interventions and disease diagnosis by overcoming common imaging challenges.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Biomedical Engineering
Background:
- Accurate organ contour extraction in ultrasound images is crucial for image-guided interventions and disease diagnosis.
- Challenges include ambiguous organ outlines, shadow artifacts, and organ shape variability.
- Existing methods struggle with these inherent limitations in ultrasound data.
Purpose of the Study:
- To develop an intelligent and accurate contour extraction method for ultrasound images.
- To address limitations of current techniques in delineating organ boundaries.
- To improve the precision of organ contour extraction for clinical applications.
Main Methods:
- A four-stage method incorporating an improved adaptive principal curve for data acquisition.
- Utilized an enhanced quantum evolution network for optimal neural network selection.
- Employed neural network training with a specific data sequence and mathematical formula for contour smoothing.
Main Results:
- The proposed method achieved superior performance compared to hybrid and Transformer-based deep learning techniques.
- Demonstrated high accuracy with an average Dice similarity coefficient of 95.7 ± 2.4%.
- Achieved excellent Jaccard similarity coefficient (94.6 ± 2.6%) and accuracy (95.3 ± 2.6%).
Conclusions:
- The developed approach offers an intelligent solution for contour extraction in ultrasound imaging.
- Provides more satisfactory outcomes than current state-of-the-art methods.
- Has the potential to significantly enhance disease diagnosis and therapeutic outcomes by defining precise organ boundaries.
Abstract:
Introduction: Accurate contour extraction in ultrasound images is of great interest for image-guided organ interventions and disease diagnosis. Nevertheless, it remains a problematic issue owing to the missing or ambiguous outline between organs (i.e., prostate and kidney) and surrounding tissues, the appearance of shadow artifacts, and the large variability in the shape of organs. Methods: To address these issues, we devised a method that includes four stages. In the first stage, the data sequence is acquired using an improved adaptive selection principal curve method, in which a limited number of radiologist defined data points are adopted as the prior. The second stage then uses an enhanced quantum evolution network to help acquire the optimal neural network. The third stage involves increasing the precision of the experimental outcomes after training the neural network, while using the data sequence as the input. In the final stage, the contour is smoothed using an explicable mathematical formula explained by the model parameters of the neural network. Results: Our experiments showed that our approach outperformed other current methods, including hybrid and Transformer-based deep-learning methods, achieving an average Dice similarity coefficient, Jaccard similarity coefficient, and accuracy of 95.7 ± 2.4%, 94.6 ± 2.6%, and 95.3 ± 2.6%, respectively. Discussion: This work develops an intelligent contour extraction approach on ultrasound images. Our approach obtained more satisfactory outcome compared with recent state-of-the-art approaches . The knowledge of precise boundaries of the organ is significant for the conservation of risk structures. Our developed approach has the potential to enhance disease diagnosis and therapeutic outcomes.
