Limits...
Automatic segmentation of the left ventricle in cardiac MRI using local binary fitting model and dynamic programming techniques.

Hu H, Gao Z, Liu L, Liu H, Gao J, Xu S, Li W, Huang L - PLoS ONE (2014)

Bottom Line: The overlapping dice metric is about 0.91.The regression and determination coefficient between the experts and our proposed method on the LV mass is 1.038 and 0.9033, respectively; they are 1.076 and 0.9386 for ejection fraction (EF).The proposed segmentation method shows the better performance and has great potential in improving the accuracy of computer-aided diagnosis systems in cardiovascular diseases.

View Article: PubMed Central - PubMed

Affiliation: College of Biomedical Engineering, South-Central University for Nationalities, Wuhan, People's Republic of China.

ABSTRACT
Segmentation of the left ventricle is very important to quantitatively analyze global and regional cardiac function from magnetic resonance. The aim of this study is to develop a novel algorithm for segmenting left ventricle on short-axis cardiac magnetic resonance images (MRI) to improve the performance of computer-aided diagnosis (CAD) systems. In this research, an automatic segmentation method for left ventricle is proposed on the basis of local binary fitting (LBF) model and dynamic programming techniques. The validation experiments are performed on a pool of data sets of 45 cases. For both endo- and epi-cardial contours of our results, percentage of good contours is about 93.5%, the average perpendicular distance are about 2 mm. The overlapping dice metric is about 0.91. The regression and determination coefficient between the experts and our proposed method on the LV mass is 1.038 and 0.9033, respectively; they are 1.076 and 0.9386 for ejection fraction (EF). The proposed segmentation method shows the better performance and has great potential in improving the accuracy of computer-aided diagnosis systems in cardiovascular diseases.

Show MeSH

Related in: MedlinePlus

Some segmentation outputs from 9 studies with LVOT by our method; for each case, four images are shown.The left ones are the gray image and its segmentation results from the ES phase; the right ones are derived from the ED phase. The names below the image data are from the data source. The dashed blue curves indicate our contours; while the solid red ones are the ground truth. The epicardial contours are not drawn by experts in the ES phase (cropped for better viewing).
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4263664&req=5

pone-0114760-g005: Some segmentation outputs from 9 studies with LVOT by our method; for each case, four images are shown.The left ones are the gray image and its segmentation results from the ES phase; the right ones are derived from the ED phase. The names below the image data are from the data source. The dashed blue curves indicate our contours; while the solid red ones are the ground truth. The epicardial contours are not drawn by experts in the ES phase (cropped for better viewing).

Mentions: We reveal some segmentation outputs from 9 studies with LVOT by our method in Fig. 5, the names below the image data are from the data source [28]. Segmentation results from our method and clinical experts are in different styles. The contours from experts are solid red curves; while the outputs from ours are dashed blue ones. For each case, four images are shown. The left ones are the gray image and its segmentation result from the ES phase; while the right ones are derived from the ED phase. The ground truth for the epicardial contour of the LV in ES phase is omitted by clinical experts, because it is not used when computing ejection fraction and left ventricle mass by the published evaluation software.


Automatic segmentation of the left ventricle in cardiac MRI using local binary fitting model and dynamic programming techniques.

Hu H, Gao Z, Liu L, Liu H, Gao J, Xu S, Li W, Huang L - PLoS ONE (2014)

Some segmentation outputs from 9 studies with LVOT by our method; for each case, four images are shown.The left ones are the gray image and its segmentation results from the ES phase; the right ones are derived from the ED phase. The names below the image data are from the data source. The dashed blue curves indicate our contours; while the solid red ones are the ground truth. The epicardial contours are not drawn by experts in the ES phase (cropped for better viewing).
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4263664&req=5

pone-0114760-g005: Some segmentation outputs from 9 studies with LVOT by our method; for each case, four images are shown.The left ones are the gray image and its segmentation results from the ES phase; the right ones are derived from the ED phase. The names below the image data are from the data source. The dashed blue curves indicate our contours; while the solid red ones are the ground truth. The epicardial contours are not drawn by experts in the ES phase (cropped for better viewing).
Mentions: We reveal some segmentation outputs from 9 studies with LVOT by our method in Fig. 5, the names below the image data are from the data source [28]. Segmentation results from our method and clinical experts are in different styles. The contours from experts are solid red curves; while the outputs from ours are dashed blue ones. For each case, four images are shown. The left ones are the gray image and its segmentation result from the ES phase; while the right ones are derived from the ED phase. The ground truth for the epicardial contour of the LV in ES phase is omitted by clinical experts, because it is not used when computing ejection fraction and left ventricle mass by the published evaluation software.

Bottom Line: The overlapping dice metric is about 0.91.The regression and determination coefficient between the experts and our proposed method on the LV mass is 1.038 and 0.9033, respectively; they are 1.076 and 0.9386 for ejection fraction (EF).The proposed segmentation method shows the better performance and has great potential in improving the accuracy of computer-aided diagnosis systems in cardiovascular diseases.

View Article: PubMed Central - PubMed

Affiliation: College of Biomedical Engineering, South-Central University for Nationalities, Wuhan, People's Republic of China.

ABSTRACT
Segmentation of the left ventricle is very important to quantitatively analyze global and regional cardiac function from magnetic resonance. The aim of this study is to develop a novel algorithm for segmenting left ventricle on short-axis cardiac magnetic resonance images (MRI) to improve the performance of computer-aided diagnosis (CAD) systems. In this research, an automatic segmentation method for left ventricle is proposed on the basis of local binary fitting (LBF) model and dynamic programming techniques. The validation experiments are performed on a pool of data sets of 45 cases. For both endo- and epi-cardial contours of our results, percentage of good contours is about 93.5%, the average perpendicular distance are about 2 mm. The overlapping dice metric is about 0.91. The regression and determination coefficient between the experts and our proposed method on the LV mass is 1.038 and 0.9033, respectively; they are 1.076 and 0.9386 for ejection fraction (EF). The proposed segmentation method shows the better performance and has great potential in improving the accuracy of computer-aided diagnosis systems in cardiovascular diseases.

Show MeSH
Related in: MedlinePlus