Analysing functional implications of differences in left ventricular morphology using statistical shape modelling

Figures

• 

  Figure 1. Statistical shape modelling (SSM) framework. Step 1: Cardiovascular magnetic resonance (CMR) cine short axis images segmentation to create 3D volume meshes of LV endocardium (in red) at end-diastole using Segment software; Step 2: Indication of the reconstructed shape population; Step 3: A shape mode showing a deformation as being correlated to LVEF%, exemplifying the feasibility of statistical analysis to capture parameters of interest.

• 

  Figure 2. Morphometric measurements. (a) Calculation of LV sphericity as short (B) to long (A) axis ratio, with short axis set as the perpendicular intersect of the long axis mid-point (A/2) while long axis being the distance from the mitral valve mid-point (C) to apex, and calculation of LV conicity as apical (E) to short axis ratio, with apical axis being the diameter of the best fitting sphere (D) (calculations were taken directly on the scans). (b) Calculation of aortic gothicity as height to width ratio (H/W) [The aortic arch width was measured as the perpendicular distance from the centreline at the sinotubular junction level to the centreline at the mid-descending aorta. The height of the arch was measured as the maximum vertical distance from the width measurement to the highest point of the centreline in the arch], and aortic tortuosity as 1-(W/l) with l being the length of the aortic arch as shown by T (highlighted in blight-blue on the centreline). T shows the endpoints set to indicate the aortic arch defined by using the pulmonary artery as a landmark (calculation were taken using Mimics software).

• 

  Figure 3. LV shape templates, showing the average 3D configuration of the ventricles. As labelled, the figure illustrates the average shapes of each group separately, both side and top views. The whole population template is also shown.

• 

  Figure 4. Shape modes. Illustration of the five computed shape modes capturing LV variation (± 2.7 SD from the mean configuration, representing the shape mode extremities, or more extreme anatomic features in this population). For each mode, it is indicated what the mode overall captures, e.g., Mode 1 = size. SD = standard deviation.

• 

  Figure 5. Correlations for shape modes 3 and 4 and corresponding morphometric LV measurements (sphericity and conicity). Sphericity is significantly positively correlated with Mode 3. SD = standard deviation.

• 

  Figure 6. GLS was associated with Modes 2 and 3, showing that increased height and sphericity correlate with worse LV function.

• 

  Figure 7. LV shape correlation with function. This figure illustrates the association between GRS and GCS with Mode 3, sphericity and conicity.