Alvaro Ulloa

Sushravya Raghunath, Alvaro E. Ulloa Cerna, Linyuan Jing et al.

The electrocardiogram (ECG) is a widely-used medical test, typically consisting of 12 voltage versus time traces collected from surface recordings over the heart. Here we hypothesize that a deep neural network can predict an important future clinical event (one-year all-cause mortality) from ECG voltage-time traces. We show good performance for predicting one-year mortality with an average AUC of 0.85 from a model cross-validated on 1,775,926 12-lead resting ECGs, that were collected over a 34-year period in a large regional health system. Even within the large subset of ECGs interpreted as 'normal' by a physician (n=297,548), the model performance to predict one-year mortality remained high (AUC=0.84), and Cox Proportional Hazard model revealed a hazard ratio of 6.6 (p<0.005) for the two predicted groups (dead vs alive one year after ECG) over a 30-year follow-up period. A blinded survey of three cardiologists suggested that the patterns captured by the model were generally not visually apparent to cardiologists even after being shown 240 paired examples of labeled true positives (dead) and true negatives (alive). In summary, deep learning can add significant prognostic information to the interpretation of 12-lead resting ECGs, even in cases that are interpreted as 'normal' by physicians.

Alvaro E. Ulloa Cerna, Marios Pattichis, David P. vanMaanen et al.

The majority of biomedical studies use limited datasets that may not generalize over large heterogeneous datasets that have been collected over several decades. The current paper develops and validates several multimodal models that can predict 1-year mortality based on a massive clinical dataset. Our focus on predicting 1-year mortality can provide a sense of urgency to the patients. Using the largest dataset of its kind, the paper considers the development and validation of multimodal models based on 25,137,015 videos associated with 699,822 echocardiography studies from 316,125 patients, and 2,922,990 8-lead electrocardiogram (ECG) traces from 631,353 patients. Our models allow us to assess the contribution of individual factors and modalities to the overall risk. Our approach allows us to develop extremely low-parameter models that use optimized feature selection based on feature importance. Based on available clinical information, we construct a family of models that are made available in the DISIML package. Overall, performance ranges from an AUC of 0.72 with just ten parameters to an AUC of 0.89 with under 105k for the full multimodal model. The proposed approach represents a modular neural network framework that can provide insights into global risk trends and guide therapies for reducing mortality risk.

Alvaro Ulloa, Linyuan Jing, Christopher W Good et al.

Predicting future clinical events helps physicians guide appropriate intervention. Machine learning has tremendous promise to assist physicians with predictions based on the discovery of complex patterns from historical data, such as large, longitudinal electronic health records (EHR). This study is a first attempt to demonstrate such capabilities using raw echocardiographic videos of the heart. We show that a large dataset of 723,754 clinically-acquired echocardiographic videos (~45 million images) linked to longitudinal follow-up data in 27,028 patients can be used to train a deep neural network to predict 1-year mortality with good accuracy (area under the curve (AUC) in an independent test set = 0.839). Prediction accuracy was further improved by adding EHR data (AUC = 0.858). Finally, we demonstrate that the trained neural network was more accurate in mortality prediction than two expert cardiologists. These results highlight the potential of neural networks to add new power to clinical predictions.

Alvaro Ulloa, Sergey Plis, Vince Calhoun

The wide variety of brain imaging technologies allows us to exploit information inherent to different data modalities. The richness of multimodal datasets may increase predictive power and reveal latent variables that otherwise would have not been found. However, the analysis of multimodal data is often conducted by assuming linear interactions which impact the accuracy of the results. We propose the use of a multimodal multi-layer perceptron model to enhance the predictive power of structural and functional magnetic resonance imaging (sMRI and fMRI) combined. We also use a synthetic data generator to pre-train each modality input layers, alleviating the effects of the small sample size that is often the case for brain imaging modalities. The proposed model improved the average and uncertainty of the area under the ROC curve to 0.850+-0.051 compared to the best results on individual modalities (0.741+-0.075 for sMRI, and 0.833+-0.050 for fMRI).