CardiacField: Computational Echocardiography for Automated Heart Function Estimation Using 2DE Probes

European Heart Journal - Digital Health

Chengkang Shen1, Hao Zhu1, You Zhou1, Yu Liu2, Si Yi1, Lili Dong2, Weipeng Zhao2, David J. Brady3, Xun Cao1, Zhan Ma1, Yi Lin2
1Nanjing University, 2Fudan University, 3University of Arizona
Paper Code

Abstract

Background: Accurate heart function estimation is vital for detecting and monitoring cardiovascular diseases. While two-dimensional echocardiography (2DE) is widely accessible and used, it requires specialized training, is prone to inter-observer variability, and lacks comprehensive three-dimensional (3D) information.

Aim: We introduce CardiacField, a computational echocardiography system using a 2DE probe for precise, automated left ventricular (LV) and right ventricular (RV) ejection fraction (EF) estimations, especially easy-to-use for non-cardiovascular health care practitioners. We assess the system's usability among amateur users and evaluate its performance against expert interpretations and advanced deep learning tools.

Methods: We developed an implicit neural representation network to reconstruct a 3D cardiac volume from sequential multi-view 2D echocardiographic images, followed by automatic segmentation of left ventricular (LV) and right ventricular (RV) areas to calculate volume sizes and ejection fraction (EF) values. Our study involved 127 patients to assess EF estimation accuracy against expert readings and 2D video-based deep learning models. A subset of 56 patients was utilized to evaluate image quality and 3D accuracy, and another 50 to test usability by amateurs and across various ultrasound machines.

Results: CardiacField generates a 3D heart from 2D echocardiograms with <2 minute processing time. The LVEF predicted by our method has a mean absolute error (MAE) of 2.48%, while the RVEF has an MAE of 2.65%.

Conclusions: Employing a straightforward apical ring scan with a cost-effective 2DE probe, our method achieves a level of EF accuracy for assessing LV and RV function that is comparable to that of 3DE probes.

The workflow of the CardiacField

Description of the image
The workflow of the CardiacField and EF calculation. (a) The whole 3D heart is represented as an implicit function, where the input is a 3D coordinate of the heart and the output is the corresponding intensity. This continuous 3D function is approximated by an MLP network with multiresolution hash-table. The implicit function is determined by minimizing a physical-informed loss function. (b) 2D echocardiographic images are acquired by rotating the 2DE probe in 360 degrees around the apex of the heart. Then we synchronize multiple cardiac views and select images at end-diastole and end-systole based on concurrently recorded ECG. (c) The 3D rendering of the reconstructed heart by the CardiacField. (d) The CardiacField represents a 3D heart in a continuous implicit function, leading to less artifacts in slices compared with the conventional interpolation as indicated by the yellow arrows. (e) The workflow of EF calculation based on the CardiacField. We first perform the uniform sampling on the reconstructed 3D heart to generate about 20-30, 3mm-thick 2D slices parallel to the apical four-chamber view, and then use the segmentation model developed in EchoNet to classify the LV and RV regions. We calculate the volumes of the LV and RV by summing the area of each slice. The EF is defined as the ratio of changes in the ESV and EDV of LV/RV. (MLP: multilayer perceptron. ECG: electrocardiogram. LV: left ventricular. RV: right ventricular. ESV: end-systolic volume. EDV: end-diastolic volume.)

Results

Analysis of 3D Heart Reconstruction Accuracy

Description of the image
(a) Visualization of initialized and refined positional parameters, where the red and blue lines illustrate the least squares regression lines between the estimated positional parameters and the ground truth. (b) Visualization of the positional parameter trajectories. The initial and refined positional parameters are obtained using PlaneInVol and CardiacField, respectively.

Assessment of Image Quality

Description of the image
(a) We evaluated the image quality of 3D hearts reconstructed by CardiacField and a conventional interpolation method, comparing them against real-captured 3D hearts. For quality assessment, we used PSNR scores, where higher values indicate better image quality, across newly generated long-axis views, short-axis views, and the overall 3D volume. (b) We demonstrated the 'continuous-slicing' capability of CardiacField in contrast to the conventional interpolation method. By extracting and comparing the same views from different methods, we showed that CardiacField can produce arbitrary views while avoiding the interpolation artifacts, which are marked with yellow arrows. (c) We use the CardiacField to reconstruct realistic 3D heart with real-captured 2D echocardiographic images. (d) We compared cross-sectional views (apical four-chamber view) of hearts reconstructed by CardiacField with those captured by a 3D probe across 10 patients. The results highlight that CardiacField’s reconstructions are more detailed and exhibit fewer artifacts compared to those by the 3D probe.

3D dynamic heart within one cardiac cycle

Description of the image
We show some snapshots of a reconstructed 3D dynamic heart within one cardiac cycle using the CardiacField, compared with that acquired by the 3DE probe. The ECG is used to indicate the phases within one cardiac cycle.

EF estimation

Description of the image
(a) We compare our LVEF results with EchoNet, where the EFs obtained by the 3D ultrasound machine (after calibration by the experienced echocardiographers) are set as the ground truth. The red and blue lines represent the least squares regression line between model prediction and ground truth, respectively. (b) We also compare our RVEF results with RVENet. (LVEF = left ventricular ejection fraction, RVEF = right ventricular ejection fraction)

Volume Segmentation Results

BibTeX

@article{shen2024cardiacfield,
  author = {Shen, Chengkang and Zhu, Hao and Zhou, You and Liu, Yu and Yi, Si and Dong, Lili and Zhao, Weipeng and Brady, David and Cao, Xun and Ma, Zhan and Lin, Yi},
  title = {CardiacField: Computational Echocardiography for Automated Heart Function Estimation Using 2DE Probes},
  journal = {European Heart Journal - Digital Health},
  pages = {ztae072},
  year = {2024},
  month = {09},
  issn = {2634-3916},
  doi = {10.1093/ehjdh/ztae072}
}