Unsupervised Image Segmentation on 2D Echocardiogram

被引:1
作者
Cacao, Gabriel Farias [1 ]
Du, Dongping [1 ]
Nair, Nandini [2 ]
机构
[1] Texas Tech Univ, Dept Ind Mfg & Syst Engn IMSE, Lubbock, TX 79409 USA
[2] Milton S Hershey Med Ctr, Heart & Vasc Inst, PennState Coll Med, 500 Univ Dr, Hershey, PA 17033 USA
关键词
unsupervised learning; image segmentation; echocardiography; convolutional neural networks; medical imaging;
D O I
10.3390/a17110515
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Echocardiography is a widely used, non-invasive imaging technique for diagnosing and monitoring heart conditions. However, accurate segmentation of cardiac structures, particularly the left ventricle, remains a complex task due to the inherent variability and noise in echocardiographic images. Current supervised models have achieved state-of-the-art results but are highly dependent on large, annotated datasets, which are costly and time-consuming to obtain and depend on the quality of the annotated data. These limitations motivate the need for unsupervised methods that can generalize across different image conditions without relying on annotated data. In this study, we propose an unsupervised approach for segmenting 2D echocardiographic images. By combining customized objective functions with convolutional neural networks (CNNs), our method effectively segments cardiac structures, addressing the challenges posed by low-resolution and gray-scale images. Our approach leverages techniques traditionally used outside of medical imaging, optimizing feature extraction through CNNs in a data-driven manner and with a new and smaller network design. Another key contribution of this work is the introduction of a post-processing algorithm that refines the segmentation to isolate the left ventricle in both diastolic and systolic positions, enabling the calculation of the ejection fraction (EF). This calculation serves as a benchmark for evaluating the performance of our unsupervised method. Our results demonstrate the potential of unsupervised learning to improve echocardiogram analysis by overcoming the limitations of supervised approaches, particularly in settings where labeled data are scarce or unavailable.
引用
收藏
页数:28
相关论文
共 50 条
[21]   Method and evaluation test design for 2D/3D image segmentation [J].
Giorgini, F ;
Dellepiane, S ;
Incardona, G ;
Piotto, F .
APPLICATIONS AND SCIENCE OF NEURAL NETWORKS, FUZZY SYSTEMS, AND EVOLUTIONARY COMPUTATION, 1998, 3455 :332-339
[22]   Clustering and Graph Convolution of Sub-Regions for Unsupervised Image Segmentation [J].
Jiao, Xue .
IEEE ACCESS, 2022, 10 :15506-15515
[23]   Unsupervised color image segmentation [J].
Liu, RJ ;
Yuan, BZ .
2002 6TH INTERNATIONAL CONFERENCE ON SIGNAL PROCESSING PROCEEDINGS, VOLS I AND II, 2002, :744-747
[24]   IMAGE SEGMENTATION BASED ON 2D RENYI GRAY ENTROPY AND FUZZY CLUSTERING [J].
Cheng, Chuanqi ;
Hao, Xiangyang ;
Liu, Songlin .
2014 12TH INTERNATIONAL CONFERENCE ON SIGNAL PROCESSING (ICSP), 2014, :738-742
[25]   Image segmentation and classification based on a 2D distributed hidden Markov model [J].
Ma, Xiang ;
Schonfeld, Dan ;
Khokhar, Ashfaq .
VISUAL COMMUNICATIONS AND IMAGE PROCESSING 2008, PTS 1 AND 2, 2008, 6822
[26]   Contextual embedding learning to enhance 2D networks for volumetric image segmentation [J].
Wang, Zhuoyuan ;
Sun, Dong ;
Zeng, Xiangyun ;
Wu, Ruodai ;
Wang, Yi .
EXPERT SYSTEMS WITH APPLICATIONS, 2024, 253
[27]   A Novel Programmatic Hybrid 2D / 3D Image Segmentation and Labeling Methodology [J].
Busch, Mike ;
Funk, Andrew ;
McIlvride, Matt ;
Thomas, Trevor ;
Fuller, Brian ;
Blair, Dan ;
Chapman, John .
2022 IEEE APPLIED IMAGERY PATTERN RECOGNITION WORKSHOP, AIPR, 2022,
[28]   A Dynamically Weighted Loss Function for Unsupervised Image Segmentation [J].
Guermazi, Boujemaa ;
Ksantini, Riadh ;
Khan, Naimul .
2022 IEEE EIGHTH INTERNATIONAL CONFERENCE ON MULTIMEDIA BIG DATA (BIGMM 2022), 2022, :73-78
[29]   UNSUPERVISED DETECTION OF DISTURBANCES IN 2D RADIOGRAPHS [J].
Estacio, Laura ;
Ehlke, Moritz ;
Tack, Alexander ;
Castro, Eveling ;
Lamecker, Hans ;
Mora, Rensso ;
Zachow, Stefan .
2021 IEEE 18TH INTERNATIONAL SYMPOSIUM ON BIOMEDICAL IMAGING (ISBI), 2021, :367-370
[30]   Unsupervised Performance Evaluation of Image Segmentation [J].
Sebastien Chabrier ;
Bruno Emile ;
Christophe Rosenberger ;
Helene Laurent .
EURASIP Journal on Advances in Signal Processing, 2006