返回
AWOLSE: Adaptive Weight-Optimized Level Set Evolution-Based Blood Cell Segmentation
DOI:10.1109/TIM.2023.3331410.png)
摘要
En 中文
Proper and precise segmentation of blood cells is a crucial task in the detection of hematological disorders. The presence of intensity inhomogeneity, noise, and weak edges makes it more challenging. To mitigate these issues, a novel adaptive weight-optimized level set evolution (AWOLSE)-based segmentation scheme is developed. In this proposed AWOLSE scheme, weights of edge and area terms are adaptively updated in each iteration by minimizing the energy function, resulting in efficient contour detection. More importantly, a novel mathematical model is proposed to make the adaptive optimization of energy terms computationally efficient for exhibiting more accurate segmentation. Moreover, AWOLSE is hybridized with marker-controlled watershed segmentation integrating the benefits of both, leading to precise blood cell segmentation. In addition, k-means-based color segmentation is employed before AWOLSE to remove undesired cells, which can boost overall performance. The assets of the proposed method are: 1) improved performance; 2) computationally efficient adaptive optimization of energy terms; and 3) more accurate segmentation with precise boundary detection. The experimental results demonstrate that the proposed AWOLSE-based scheme outperforms other recent level set evolution (LSE) methods with the best performances.
Keyword:
Image segmentation
Image edge detection
Optimization
Databases
Watersheds
Image color analysis
Feature extraction
Acute lymphoblastic leukemia (ALL)
acute myeloid leukemia (AML)
blood cancer
hematological disorder
image segmentation
level set evolution (LSE)
sickle cell anemia (SCA)
期刊
IF:
5.9
论文数:
2.0W
被引数:
5.8W
机构
引用论文
Development and Validation of the University of Washington Clinical Assessment of Music Perception Test华盛顿大学音乐知觉临床评估测试的开发和验证
Electrochemically assisted micro localized grafting of aptamers in a microchannel engraved in fluorinated thermoplastic polymer Dyneon THV
RSC Advances
IF0

