Three-dimensional analysis of left ventricle regional wall motion by using gated blood pool tomography

Abstract Background The deep neural network assisted in automated echocardiography interpretation joint to cardiologist final confirmation has now been gradually emerging. There were applications applied in echocardiography views classification, chamber size and myocardium mass evaluation, and certain disease detections already published. Our aim, instead of frame-by-frame “image-level” interpretation in previous studies, is to apply deep neural network in echocardiography temporal relationship analysis – “video-level” – and applied in automated left ventricle myocardium regional wall motion abnormalities recognition. Methods We collected all echocardiography performed in 2017, and preprocessed them into numeric arrays for matrix computations. Regional wall motion abnormalities were approved by authorized cardiologists, and processed into labels whether regional wall motion abnormalities presented in anterior, inferior, septal, or lateral walls of the left ventricle, as the ground truth. We then first developed a convolutional neural network (CNN) model to do view selection, and gathered parasternal long/short views, and apical four/two chamber views from each exam, as well as developing view prediction confidence for strict image quality control. Within these images, we annotated part of images to develop the second CNN model, known as U-net, for image segmentation and mark each regional wall. Finally, we developed the major three-dimensional CNN model with the inputs composed of four views of echocardiography videos and then output the final label for motion abnormalities in each wall. Results In total we collected 13,984 series of echocardiography, and gathered four main views with quality confidence level above 90%, which resulted in 9,323 series for training. Within these images, we annotated 2,736 frames for U-net model and resulted in dice score of segmentation 73%. With the join of segmentation model, the final three-dimensional CNN model predict regional wall motion with accuracy of 83%. Conclusions Deep neural network application in regional wall motion recognition is feasible and should mandate further investigation for promoting performance. Acknowledgement/Funding None

Download Full-text

Nuclear cardiology: a comparison of equilibrium gated blood pool and gamma camera first-pass measurements for the assessment of left-ventricular regional wall motion

British Journal of Radiology ◽

10.1259/0007-1285-59-705-901 ◽

1986 ◽

Vol 59 (705) ◽

pp. 901-905

Author(s):

D. N. Taylor ◽

S. Choraria ◽

J. Mills ◽

J. A. McIntosh ◽

J. Pilcher

Keyword(s):

Nuclear Cardiology ◽

Gamma Camera ◽

Wall Motion ◽

Regional Wall Motion ◽

Blood Pool ◽

Left Ventricular ◽

Regional Wall ◽

Gated Blood Pool ◽

First Pass

Download Full-text

Assessment of left ventricular regional wall motion abnormalities using regional time-volume curves obtained by real time three-dimensional echocardiography

The Egyptian Heart Journal ◽

10.1016/j.ehj.2018.02.002 ◽

2018 ◽

Vol 70 (3) ◽

pp. 189-194

Author(s):

Ghada Youssef ◽

Zakarya Saad ◽

Waleed Ammar ◽

Yasser Sharaf

Keyword(s):

Real Time ◽

Wall Motion ◽

Three Dimensional ◽

Regional Wall Motion ◽

Left Ventricular ◽

Regional Wall ◽

Regional Wall Motion Abnormalities ◽

Wall Motion Abnormalities ◽

Dimensional Echocardiography ◽

Three Dimensional Echocardiography

Download Full-text

Assessment of Regional Wall Motion Abnormality in Patients with Acute Anteroseptal ST Segment Elevation Myocardial Infarction

University Heart Journal ◽

10.3329/uhj.v15i1.41443 ◽

2019 ◽

Vol 15 (1) ◽

pp. 28-33

Author(s):

Tunaggina Afrin Khan ◽

Saiful Ahmed ◽

Mostashirul Haque ◽

Md Rasul Amin ◽

ATM Iqbal Hasan ◽

...

Keyword(s):

Myocardial Infarction ◽

Left Ventricle ◽

Wall Motion Abnormality ◽

Wall Motion ◽

Regional Wall Motion ◽

Regional Wall Motion Abnormality ◽

Regional Wall ◽

St Segment Elevation ◽

Segment Elevation Myocardial Infarction ◽

St Segment

Post myocardial infarction (MI) short and long term clinical outcome is largely determined by the size of the infarcted area. It is generally assumed that as the lead involvement in the 12 lead electrocardiography (ECG) is less in anteroseptal ST segment elevation myocardial infarction (AS-STEMI), where ST segment elevation (STE) is limited to leads V1 to V3, myocardial damage is likely to be less. This study was intended to assess regional wall motion abnormality (RWMA) in acute anteroseptal STEMI patients. 90 patients with AS-STEMI admitted in between October 2012 and September 2013, were included. For each patient, a transthoracic echocardiogram (TTE) was performed within 24-48 hours of MI and was interpreted by an independent investigator blinded to the patient’s ECG data.The mean (± SD) age of the patients was 51.57 (± 12.02) years with mean (± SD) age of the patients was 52.58 (± 12.02) years with a range of 23 - 80 years. There were 91.1% male and 8.9% female. The mean (± SD) EF% was 38.80 %( ± 5.78). All the segments of left ventricle, except basal and mid inferolateral segments, were affected in anteroseptal STEMI. So, the term AS-STEMI may be a misnomer, as it implies that only the anteroseptal segments of the left ventricle are involved. This study shows that regional dysfunction in patients with AS-STEMI extends beyond the anteroseptal region and may be it is as much extensive as extensive anterior myocardial infarction. So, any patients with anterior wall involvement should be treated with utmost importance. University Heart Journal Vol. 15, No. 1, Jan 2019; 28-33

Download Full-text

Floating axis does not reduce motion artifacts in a model of left ventricular wall motion in dogs

AJP Heart and Circulatory Physiology ◽

10.1152/ajpheart.1993.264.2.h631 ◽

1993 ◽

Vol 264 (2) ◽

pp. H631-H638

Author(s):

T. N. Nguyen ◽

S. A. Glantz

Keyword(s):

Left Ventricle ◽

Measurement Method ◽

Wall Motion ◽

Regional Wall Motion ◽

Left Ventricular ◽

Motion Artifacts ◽

Regional Wall ◽

Myocardial Wall Motion ◽

Myocardial Wall ◽

Deformation Patterns

Methods of measuring regional wall motion of the left ventricle superimpose end-diastolic and end-systolic images. Differences in dimensions between images are assumed to be due to contraction, but they are also due to motion artifacts. To determine whether the errors caused by motion artifacts are reduced when measured with floating-axis referencing, and whether the measurement method affects these errors, we simulated end-systolic angiograms of a pure contraction (control) and contractions affected by motion artifacts and then measured differences in wall motion between angiograms with hemichord, radial, and trapezoid methods, using floating-axis and fixed-axis referencing. We chose these three methods because they form the basis for other methods, e.g., the center line method. For the simulations, we applied deformation patterns of the left ventricle, computed from the motion of tantalum markers implanted in the endocardiums of six dogs, to end-diastolic angiograms. This marker method measured the myocardial wall motion directly, independent of the angiogram. We found that differences caused by motion artifacts were not significantly reduced when measured with floating-axis referencing in our model. Normalized differences measured by radial and trapezoid methods were not significantly different, but they were significantly smaller than those measured by the hemichord method. We conclude that the axis referencing system has no significant effect on errors caused by motion artifacts in regional wall motion in our model. The measurement method, however, does affect these errors, with the radial and trapezoid methods being superior to the hemichord method.

Download Full-text