Automated assessment of cardiac dynamics in aging and dilated cardiomyopathy Drosophila models using machine learning

Abstract

The Drosophila model is pivotal in deciphering the pathophysiological underpinnings of various human ailments, notably aging and cardiovascular diseases. Cutting-edge imaging techniques and physiology yield vast high-resolution videos, demanding advanced analysis methods. Our platform leverages deep learning to segment optical microscopy images of Drosophila hearts, enabling the quantification of cardiac parameters in aging and dilated cardiomyopathy (DCM). Validation using experimental datasets confirms the efficacy of our aging model. We employ two innovative approaches deep-learning video classification and machine-learning based on cardiac parameters to predict fly aging, achieving accuracies of 83.3% (AUC 0.90) and 79.1%, (AUC 0.87) respectively. Moreover, we extend our deep-learning methodology to assess cardiac dysfunction associated with the knock-down of oxoglutarate dehydrogenase (OGDH), revealing its potential in studying DCM. This versatile approach promises accelerated cardiac assays for modeling various human diseases in Drosophila and holds promise for application in animal and human cardiac physiology under diverse conditions.

Fig. 1. Proposed deep learning pipeline.

a One pipeline enables frame-level segmentation of heart walls. For this task, users upload videos for analysis, perform segmentation using a trained neural network, select validated parameters, and extract beat-level parameters. This feeds into the middle pipeline which enables machine learning-based age-prediction of hearts via calculated cardiac statistics, further enabling transparent modeling of physiological parameters. Lastly, the bottom-most pipeline details an age-prediction task using a neural network, thus providing an alternate method for age-prediction. b Schematic of the presented segmentation model. We use an attention-UNet with rectangular convolution kernels. A diagram of the attention gates is also shown.

Fig. 2. Representative output of the neural network.

a Select frame extracted from cardiac video, with the overlaid mask from a neural network denoting the heart wall. b, c Using annotated frames, an annotated mechanical mode (M-Mode) image is generated. d, e Detected heart masks are used to calculate average stationary diameter on a per frame resolution. b, d, f represent generated M-Mode, time-series beating pattern, and calculated cardiac parameters for a representative 1w male heart. c, e, g represents generated M-Mode, time-series beating pattern, and calculated cardiac parameters for a representative 5w male heart. Green (start of DI) and red (end of DI) lines are annotated in b, c, and gray periods in d, e. Times of max contraction (red) and max relaxation (green) are annotated by x indicators in d, e.

Fig. 3. Deep learning recovers aging phenotypes in the Drosophila cardiac model.

a Diastolic diameter, in micron, is calculated by our model (left), SOHA (middle), and agreement between two datasets (right). b Systolic diameter, in microns, calculated by our model (left), SOHA (middle), and agreement between two datasets (right). c Radial contractility (fractional shortening), in percentage, calculated by our model (left), SOHA (middle), and agreement between two datasets (right). d Heart rate, in Hertz, is calculated by our model (left), SOHA (middle), and agreement between two datasets (right). e Beating dysrhythmia (arrhythmia index) calculated by our model (left), SOHA (middle), and agreement between two datasets (right). All error bars report ± SEM. Age-dependent statistics compared with one-way ANOVA with two-sided unpaired t-test. Statistics are calculated via the use of a restricted ROI, selected by a trained user.

Fig. 4. Contractile dynamics detected by deep learning.

a Stroke volume for male and female aging groups is depicted, in picolitres. A significant reduction in stroke volume is exhibited with aging in both genders. b Cardiac output was visualized via integration of per-beat stroke volumes and normalized by beat times, measured in pL s-1. Aging depicts a strong reduction in cardiac output. c Time to peak contraction (negative) heart-wall velocity visualized for aging. d The time between peak contraction (negative) and relaxation (positive) velocities visualized for aging. Both c and d reflect no significant change with aging. All error bars report ± SEM. Age-dependent statistics compared with one-way ANOVA with two-sided unpaired t-test t.

Fig. 5. Machine learning classification of aging data.

Logistic classification models enable accurate prediction of fly age from model-calculated cardiac statistics. a Heatmap of test-dataset predictions, indicating high agreement. b Average ROC-curve of all folds from the presented model, reporting a mean AUROC of 0.87. c Physiological feature relevancies for predicting fly age, as determined by the SHAP study. SHAP study suggests a large dependence on fractional shortening (FS) in aging.

Fig. 6. Deep learning classification of aging videos (n = 497).

Deep learning models enable accurate prediction of fly age from only video data. a Heatmap of test-dataset predictions, indicating high agreement. b Average ROC curve of all folds from the presented model, reporting a mean test accuracy of 83.3% and mean AUROC of 0.900. c Model class log-likelihood distribution for investigated test predictions.

Fig. 7. Cardiac-specific knock-down of Odgh leads to cardiac dilation and compromised cardiac performance.

Histograms of cardiac physiological parameters 3-weeks control (Hand/+) and Hand>Ogdh RNAi has compromised a diastolic diameter (DD), b systolic diameter (SD), c % fractional shortening (FS), d heart rate (HR), and e arrhythmia Index (AI) analyzed using machine learning (ML) left, and SOHA right panels. n = 35 (Hand/+) and 30 Hand>Ogdh RNAi for ML; and n = 34, (Hand/+) and 31 Hand>Ogdh RNAi for SOHA were represented as mean ± SEM. Statistical analysis of each cardiac parameter between control and Ogdh knock-down was carried out using one-way ANOVA with two-sided unpaired t-test.