Hybrid Technique forEstimating Fetal Head Circumference Using Ultrasound Imaging
Journal Title: International Journal of Innovations in Science and Technology - Year 2024, Vol 6, Issue 3
Abstract
Introduction: Analysis of fetal head shape is crucial for assessing head growth and detecting abnormalities in fetuses. In traditional clinical practice, Head Circumference (HC) is determined by manually fitting an ellipse to the fetal skull based on 2D ultrasound images. Novelty Statement: To address this, an automated method integrating image processing techniques with U-net variant have been developed to achieve maximum accuracy of fetal head circumference detection on HC18 dataset. This method aims to enhance precision in HC delineation, thereby improving clinical reliability. Material and Method: This study proposed a method that combines image processing techniques (noise removal, edge detection, segmentation) with a Residual U-net model for detecting the boundary of the fetal skull using HC18 dataset. Results and Discussion: The results of this method outperformed a simple residual u-net model in terms of accuracy. The proposed method is evaluating using the HC18 challenge dataset, achieving a Dice coefficient of 97.99%, a mean difference of 5.86 mm, and a mean Hausdorff distance of 0.56 mm compared to manual annotations. These results demonstrate the effectiveness of the proposed method in accurately delineating the fetal skull boundary. Concluding remarks: Furthermore, the proposed method shows comparability with state-ofthe-art techniques in the field.
Authors and Affiliations
Ameera Shafique, Zobia Suhail, Hafiz Muhammad Danish
Keywords
Related Articles
- EP ID EP760365
- DOI -
- Views 19
- Downloads 0
Analyzing the Shadows:Machine Learning Approaches for Depression Detection on Twitter
Depression is a leading cause of disability worldwide, affecting approximately 4.4% of the global population. It can escalate from mild symptoms to severe outcomes, including suicide, if not treated early. Thus, develo...
EvaluatingFasterR-CNNandYOLOv8forTrafficObject Detection andClass-Based Counting
Real-time traffic object detection is a critical component necessary for achieving a fully autonomous traffic system. Traffic object detection, along with background classification, is a significant area of research ai...
Python Based Estimation ofGroundwater Quality Along Hudaira Drain
During periods of restricted access to fresh surface water, enterprises depend on underground water reserves to meet their growing demands. Groundwater is crucial for fulfilling the growing demands of families, agricul...
Empowering Growth: Implementation of Sustainable Software Requirement Engineering Practices in Pakistan
Introduction/Importance of Study: Sustainability must be integrated into Software Requirements Engineering due to the environmental implications of software systems. Novelty Statement: This research addresses the curre...
Machine Learning in Livestock Management: A Systematic Exploration of Techniques and Outcomes
This Systematic Literature Review (SLR) examines the growing field of leveraging Machine Learning (ML) to improve livestock productivity. Through a meticulous analysis of peer-reviewed articles, the study categorizes r...