Fully automated and comprehensive MRI-based left-ventricular contractility analysis in post-chemotherapy breast cancer patients

Automated segmentation of the left-ventricle from MRI with a fully convolutional network to investigate CTRCD in breast cancer patients

Purpose: The goal of this study was to develop a fully convolutional network (FCN) tool to automatedly segment the left-ventricular (LV) myocardium in displacement encoding with stimulated echoes MRI. The segmentation results are used for LV chamber quantification and strain analyses in breast cancer patients susceptible to cancer therapy-related cardiac dysfunction (CTRCD). Approach: A DeepLabV3+ FCN with a ResNet-101 backbone was custom-designed to conduct chamber quantification on 45 female breast cancer datasets (23 training, 11 validation, and 11 test sets). LV structural parameters and LV ejection fraction (LVEF) were measured, and myocardial strains estimated with the radial point interpolation method. Myocardial classification validation was against quantization-based ground-truth with computations of accuracy, Dice score, average perpendicular distance (APD), Hausdorff-distance, and others. Additional validations were conducted with equivalence tests and Cronbach's alpha (C - α ) intraclass correlation coefficients between the FCN and a vendor tool on chamber quantification and myocardial strain computations. Results: Myocardial classification results against ground-truth were Dice = 0.89 , APD = 2.4    mm , and accuracy = 97 % for the validation set and Dice = 0.90 , APD = 2.5    mm , and accuracy = 97 % for the test set. The confidence intervals (CI) and two one-sided t-test results of equivalence tests between the FCN and vendor-tool were CI = - 1.36 % to 2.42%, p-value < 0.001 for LVEF (58 ± 5 % versus 57 ± 6 % ), and CI = - 0.71 % to 0.63%, p-value < 0.001 for longitudinal strain (- 15 ± 2 % versus - 15 ± 3 % ). Conclusions: The validation results were found equivalent to the vendor tool-based parameter estimates, which show that accurate LV chamber quantification followed by strain analysis for CTRCD investigation can be achieved with our proposed FCN methodology.

Machine Learning in Cardio-Oncology: New Insights from an Emerging Discipline

A growing body of evidence on a wide spectrum of adverse cardiac events following oncologic therapies has led to the emergence of cardio-oncology as an increasingly relevant interdisciplinary specialty. This also calls for better risk-stratification for patients undergoing cancer treatment. Machine learning (ML), a popular branch discipline of artificial intelligence that tackles complex big data problems by identifying interaction patterns among variables, has seen increasing usage in cardio-oncology studies for risk stratification. The objective of this comprehensive review is to outline the application of ML approaches in cardio-oncology, including deep learning, artificial neural networks, random forest and summarize the cardiotoxicity identified by ML. The current literature shows that ML has been applied for the prediction, diagnosis and treatment of cardiotoxicity in cancer patients. In addition, role of ML in gender and racial disparities for cardiac outcomes and potential future directions of cardio-oncology are discussed. It is essential to establish dedicated multidisciplinary teams in the hospital and educate medical professionals to become familiar and proficient in ML in the future.

Can global longitudinal strain (GLS) with magnetic resonance prognosticate early cancer therapy-related cardiac dysfunction (CTRCD) in breast cancer patients, a prospective study?

PURPOSE

To determine if Artificial Intelligence-based computation of global longitudinal strain (GLS) from left ventricular (LV) MRI is an early prognostic factor of cancer therapy-related cardiac dysfunction (CTRCD) in breast cancer patients. The main hypothesis based on the patients receiving antineoplastic chemotherapy treatment was CTRCD risk analysis with GLS that was independent of LV ejection fraction (LVEF).

METHODS

Displacement Encoding with Stimulated Echoes (DENSE) MRI was acquired on 32 breast cancer patients at baseline and 3- and 6-month follow-ups after chemotherapy. Two DeepLabV3+ Fully Convolutional Networks (FCNs) were deployed to automate image segmentation for LV chamber quantification and phase-unwrapping for 3D strains, computed with the Radial Point Interpolation Method. CTRCD risk (cardiotoxicity and adverse cardiac events) was analyzed with Cox Proportional Hazards (PH) models with clinical and contractile prognostic factors.

RESULTS

GLS worsened from baseline to the 3- and 6-month follow-ups (-19.1 ± 2.1%, -16.0 ± 3.1%, -16.1 ± 3.0%; P < 0.001). Univariable Cox regression showed the 3-month GLS significantly associated as an agonist (hazard ratio [HR]-per-SD: 2.1; 95% CI: 1.4-3.1; P < 0.001) and LVEF as a protector (HR-per-SD: 0.8; 95% CI: 0.7-0.9; P = 0.001) for CTRCD occurrence. Bivariable regression showed the 3-month GLS (HR-per-SD: 2.0; 95% CI: 1.2-3.4; P = 0.01) as a CTRCD prognostic factor independent of other covariates, including LVEF (HR-per-SD: 1.0; 95% CI: 0.9-1.2; P = 0.9).

CONCLUSIONS

The end-point analyses proved the hypothesis that GLS is an early, independent prognosticator of incident CTRCD risk. This novel GLS-guided approach to CTRCD risk analysis could improve antineoplastic treatment with further validation in a larger clinical trial.

Myocardial Contractility Pattern Characterization in Radiation-Induced Cardiotoxicity Using Magnetic Resonance Imaging: A Pilot Study with ContractiX

Radiation therapy (RT) plays an integral role in treating thoracic cancers, despite the risk of radiation-induced cardiotoxicity. We hypothesize that our newly developed magnetic resonance imaging (MRI)-based contractility index (ContractiX) is a sensitive marker for early detection of RT-induced cardiotoxicity in a preclinical rat model of thoracic cancer RT. Adult salt-sensitive rats received image-guided heart RT and were imaged with MRI at 8 weeks and 10 weeks post-RT or sham. The MRI exam included cine and tagging sequences to measure left-ventricular ejection fraction (LVEF), mass, myocardial strain, and ContractiX. Furthermore, ventricular torsion, diastolic strain rate, and mechanical dyssynchrony were measured. Statistical analyses were performed between the sham, 8 weeks post-RT, and 10 weeks post-RT MRI parameters. The results showed that both LVEF and myocardial mass increased post-RT. Peak systolic strain and ContractiX significantly decreased post-RT, with a more relative reduction in ContractiX compared to strain. ContractiX showed an inverse nonlinear relationship with LVEF and continuously decreased with time post-RT. While early diastolic strain rate and mechanical dyssynchrony significantly changed post-RT, ventricular torsion changes were not significant post-RT. In conclusion, ContractiX measured via non-contrast MRI is a sensitive early marker for the detection of subclinical cardiac dysfunction post-RT, and it is superior to other MRI cardiac measures.

Direct left-ventricular global longitudinal strain (GLS) computation with a fully convolutional network

This study's purpose was to develop a direct MRI-based, deep-learning semantic segmentation approach for computing global longitudinal strain (GLS), a known metric for detecting left-ventricular (LV) cardiotoxicity in breast cancer. Displacement Encoding with Stimulated Echoes cardiac image phases acquired from 30 breast cancer patients and 30 healthy females were unwrapped via a DeepLabV3 + fully convolutional network (FCN). Myocardial strains were directly computed from the unwrapped phases with the Radial Point Interpolation Method. FCN-unwrapped phases of a phantom's rotating gel were validated against quality-guided phase-unwrapping (QGPU) and robust transport of intensity equation (RTIE) phase-unwrapping. FCN performance on unwrapping human LV data was measured with F1 and Dice scores versus QGPU ground-truth. The reliability of FCN-based strains was assessed against RTIE-based strains with Cronbach's alpha (C-α) intraclass correlation coefficient. Mean squared error (MSE) of unwrapping the phantom experiment data at 0 dB signal-to-noise ratio were 1.6, 2.7 and 6.1 with FCN, QGPU and RTIE techniques. Human data classification accuracies were F1 = 0.95 (Dice = 0.96) with FCN and F1 = 0.94 (Dice = 0.95) with RTIE. GLS results from FCN and RTIE were -16 ± 3% vs. -16 ± 3% (C-α = 0.9) for patients and -20 ± 3% vs. -20 ± 3% (C-α = 0.9) for healthy subjects. The low MSE from the phantom validation demonstrates accuracy of phase-unwrapping with the FCN and comparable human subject results versus RTIE demonstrate GLS analysis accuracy. A deep-learning methodology for phase-unwrapping in medical images and GLS computation was developed and validated in a heterogeneous cohort.

Validation of a deep-learning semantic segmentation approach to fully automate MRI-based left-ventricular deformation analysis in cardiotoxicity

OBJECTIVE

Left-ventricular (LV) strain measurements with the Displacement Encoding with Stimulated Echoes (DENSE) MRI sequence provide accurate estimates of cardiotoxicity damage related to chemotherapy for breast cancer. This study investigated an automated and supervised deep convolutional neural network (DCNN) model for LV chamber quantification before strain analysis in DENSE images.

METHODS

The DeepLabV3 +DCNN with three versions of ResNet-50 backbone was designed to conduct chamber quantification on 42 female breast cancer data sets. The convolutional layers in the three ResNet-50 backbones were varied as non-atrous, atrous and modified, atrous with accuracy improvements like using Laplacian of Gaussian filters. Parameters such as LV end-diastolic diameter (LVEDD) and ejection fraction (LVEF) were quantified, and myocardial strains analyzed with the Radial Point Interpolation Method (RPIM). Myocardial classification was validated with the performance metrics of accuracy, Dice, average perpendicular distance (APD) and others. Repeated measures ANOVA and intraclass correlation (ICC) with Cronbach's α (C-Alpha) tests were conducted between the three DCNNs and a vendor tool on chamber quantification and myocardial strain analysis.

RESULTS

Validation results in the same test-set for myocardial classification were accuracy = 97%, Dice = 0.92, APD = 1.2 mm with the modified ResNet-50, and accuracy = 95%, Dice = 0.90, APD = 1.7 mm with the atrous ResNet-50. The ICC results between the modified ResNet-50, atrous ResNet-50 and vendor-tool were C-Alpha = 0.97 for LVEF (55±7%, 54±7%, 54±7%, p = 0.6), and C-Alpha = 0.87 for LVEDD (4.6 ± 0.3 cm, 4.6 ± 0.3 cm, 4.6 ± 0.4 cm, p = 0.7).

CONCLUSION

Similar performance metrics and equivalent parameters obtained from comparisons between the atrous networks and vendor tool show that segmentation with the modified, atrous DCNN is applicable for automated LV chamber quantification and subsequent strain analysis in cardiotoxicity.

ADVANCES IN KNOWLEDGE

A novel deep-learning technique for segmenting DENSE images was developed and validated for LV chamber quantification and strain analysis in cardiotoxicity detection.

A deep-learning semantic segmentation approach to fully automated MRI-based left-ventricular deformation analysis in cardiotoxicity

Left-ventricular (LV) strain measurements with the Displacement Encoding with Stimulated Echoes (DENSE) MRI sequence provide accurate estimates of cardiotoxicity damage related to breast cancer chemotherapy. This study investigated an automated LV chamber quantification tool via segmentation with a supervised deep convolutional neural network (DCNN) before strain analysis with DENSE images. Segmentation for chamber quantification analysis was conducted with a custom DeepLabV3+ DCNN with ResNet-50 backbone on 42 female breast cancer datasets (22 training-sets, eight validation-sets and 12 independent test-sets). Parameters such as LV end-diastolic diameter (LVEDD) and ejection fraction (LVEF) were quantified, and myocardial strains analyzed with the Radial Point Interpolation Method (RPIM). Myocardial classification was validated against ground-truth with sensitivity-specificity analysis, the metrics of Dice, average perpendicular distance (APD) and Hausdorff-distance. Following segmentation, validation was conducted with the Cronbach's Alpha (C-Alpha) intraclass correlation coefficient between LV chamber quantification results with DENSE and Steady State Free Precession (SSFP) acquisitions and a vendor tool-based method to segment the DENSE data, and similarly for myocardial strain analysis in the chambers. The results of myocardial classification from segmentation of the DENSE data were accuracy = 97%, Dice = 0.89 and APD = 2.4 mm in the test-set. The C-Alpha correlations from comparing chamber quantification results between the segmented DENSE and SSFP data and vendor tool-based method were 0.97 for LVEF (56 ± 7% vs 55 ± 7% vs 55 ± 6%, p = 0.6) and 0.77 for LVEDD (4.6 ± 0.4 cm vs 4.5 ± 0.3 cm vs 4.5 ± 0.3 cm, p = 0.8). The validation metrics against ground-truth and equivalent parameters obtained from the SSFP segmentation and vendor tool-based comparisons show that the DCNN approach is applicable for automated LV chamber quantification and subsequent strain analysis in cardiotoxicity.