All Publications

  • Machine Learned Cellular Phenotypes Predict Outcome in Ischemic Cardiomyopathy. Circulation research Rogers, A. J., Selvalingam, A., Alhusseini, M. I., Krummen, D. E., Corrado, C., Abuzaid, F., Baykaner, T., Meyer, C., Clopton, P., Giles, W. R., Bailis, P., Niederer, S. A., Wang, P. J., Rappel, W., Zaharia, M., Narayan, S. M. 2020


    RATIONALE: Susceptibility to ventricular arrhythmias (VT/VF) is difficult to predict in patients with ischemic cardiomyopathy either by clinical tools or by attempting to translate cellular mechanisms to the bedside.OBJECTIVE: To develop computational phenotypes of patients with ischemic cardiomyopathy, by training then interpreting machine learning (ML) of ventricular monophasic action potentials (MAPs) to reveal phenotypes that predict long-term outcomes.METHODS AND RESULTS: We recorded 5706 ventricular MAPs in 42 patients with coronary disease (CAD) and left ventricular ejection fraction (LVEF) {less than or equal to}40% during steady-state pacing. Patients were randomly allocated to independent training and testing cohorts in a 70:30 ratio, repeated K=10 fold. Support vector machines (SVM) and convolutional neural networks (CNN) were trained to 2 endpoints: (i) sustained VT/VF or (ii) mortality at 3 years. SVM provided superior classification. For patient-level predictions, we computed personalized MAP scores as the proportion of MAP beats predicting each endpoint. Patient-level predictions in independent test cohorts yielded c-statistics of 0.90 for sustained VT/VF (95% CI: 0.76-1.00) and 0.91 for mortality (95% CI: 0.83-1.00) and were the most significant multivariate predictors. Interpreting trained SVM revealed MAP morphologies that, using in silico modeling, revealed higher L-type calcium current or sodium calcium exchanger as predominant phenotypes for VT/VF.CONCLUSIONS: Machine learning of action potential recordings in patients revealed novel phenotypes for long-term outcomes in ischemic cardiomyopathy. Such computational phenotypes provide an approach which may reveal cellular mechanisms for clinical outcomes and could be applied to other conditions.

    View details for DOI 10.1161/CIRCRESAHA.120.317345

    View details for PubMedID 33167779

  • DIFF: a relational interface for large-scale data explanation VLDB JOURNAL Abuzaid, F., Kraft, P., Suri, S., Gan, E., Xu, E., Shenoy, A., Ananthanarayan, A., Sheu, J., Meijer, E., Wu, X., Naughton, J., Bailis, P., Zaharia, M. 2020
  • Machine Learning to Classify Intracardiac Electrical Patterns during Atrial Fibrillation. Circulation. Arrhythmia and electrophysiology Alhusseini, M. I., Abuzaid, F., Rogers, A. J., Zaman, J. A., Baykaner, T., Clopton, P., Bailis, P., Zaharia, M., Wang, P. J., Rappel, W., Narayan, S. M. 2020


    Background - Advances in ablation for atrial fibrillation (AF) continue to be hindered by ambiguities in mapping, even between experts. We hypothesized that convolutional neural networks (CNN) may enable objective analysis of intracardiac activation in AF, which could be applied clinically if CNN classifications could also be explained. Methods - We performed panoramic recording of bi-atrial electrical signals in AF. We used the Hilbert-transform to produce 175,000 image grids in 35 patients, labeled for rotational activation by experts who showed consistency but with variability (kappa=0.79). In each patient, ablation terminated AF. A CNN was developed and trained on 100,000 AF image grids, validated on 25,000 grids, then tested on a separate 50,000 grids. Results - In the separate test cohort (50,000 grids), CNN reproducibly classified AF image grids into those with/without rotational sites with 95.0% accuracy (CI 94.8-95.2%). This accuracy exceeded that of support vector machines, traditional linear discriminant and k-nearest neighbor statistical analyses. To probe the CNN, we applied Gradient-weighted Class Activation Mapping which revealed that the decision logic closely mimicked rules used by experts (C-statistic 0.96). Conclusions - Convolutional neural networks improved the classification of intracardiac AF maps compared to other analyses, and agreed with expert evaluation. Novel explainability analyses revealed that the CNN operated using a decision logic similar to rules used by experts, even though these rules were not provided in training. We thus describe a scaleable platform for robust comparisons of complex AF data from multiple systems, which may provide immediate clinical utility to guide ablation.

    View details for DOI 10.1161/CIRCEP.119.008160

    View details for PubMedID 32631100

  • MACHINE LEARNING IDENTIFIES SITES WHERE ABLATION TERMINATES PERSISTENT ATRIAL FIBRILLATION Alhusseini, M., Abuzaid, F., Clopton, P., Rogers, A., Rodrigo, M., Baykaner, T., Wang, P., Rappel, W., Narayan, S. ELSEVIER SCIENCE INC. 2019: 301
  • To Index or Not to Index: Optimizing Exact Maximum Inner Product Search Abuzaid, F., Sethi, G., Bailis, P., Zaharia, M., IEEE IEEE. 2019: 1250–61
  • MacroBase: Prioritizing Attention in Fast Data ACM TRANSACTIONS ON DATABASE SYSTEMS Abuzaid, F., Bailis, P., Ding, J., Gan, E., Madden, S., Narayanan, D., Rong, k., Suri, S. 2018; 43 (4)

    View details for DOI 10.1145/3276463

    View details for Web of Science ID 000457123200001

  • DIFF: A Relational Interface for Large-Scale Data Explanation PROCEEDINGS OF THE VLDB ENDOWMENT Abuzaid, F., Kraft, P., Suri, S., Gan, E., Xu, E., Shenoy, A., Ananthanarayan, A., Sheu, J., Meijer, E., Wu, X., Naughton, J., Bailis, P., Zaharia, M. 2018; 12 (4): 419–32
  • Machine Learning Reveals That Drivers for Persistent Atrial Fibrillation at Termination Sites Show Irregular Rotational Cycles and Domain Size Alhusseini, M., Abuzaid, F., Swerdlow, M., Clopton, P., Meckler, G. L., Maniar, N. M., Rogers, A., Rodrigo, M., Baykaner, T., Zaman, J., Kowalewski, C., Shenasa, F., Tamboli, M., Viswanathan, M., Wang, P., Atienza, F., Rappel, W. J., Bailis, P., Zaharia, M., Narayan, S. M. LIPPINCOTT WILLIAMS & WILKINS. 2018
  • Sites Where Ablation Terminated Atrial Fibrillation Identified by Machine Learning Models Alhusseini, M., Abuzaid, F., Swerdlow, M., Meckler, G., Clopton, P., Rogers, A., Rodrigo, M., Baykaner, T., Zaman, J., Kowalewski, C., Shenasa, F., Atienza, F., Mohan, N., Wang, P. J., Rappel, W. J., Bailis, P., Zaharia, M., Narayan, S. LIPPINCOTT WILLIAMS & WILKINS. 2018
  • NoScope: Optimizing Neural Network Queries over Video at Scale PROCEEDINGS OF THE VLDB ENDOWMENT Kang, D., Emmons, J., Abuzaid, F., Bailis, P., Zaharia, M. 2017; 10 (11): 1586–97
  • Caffe con Troll: Shallow Ideas to Speed Up Deep Learning. Proceedings of the Fourth Workshop on Data analytics at sCale (DanaC 2015) : May 31st, 2015, Melbourne, Australia. Workshop on Data Analytics in the Cloud (4th : 2015 : Melbourne, Vic.) Hadjis, S., Abuzaid, F., Zhang, C., Ré, C. 2015; 2015


    We present Caffe con Troll (CcT), a fully compatible end-to-end version of the popular framework Caffe with rebuilt internals. We built CcT to examine the performance characteristics of training and deploying general-purpose convolutional neural networks across different hardware architectures. We find that, by employing standard batching optimizations for CPU training, we achieve a 4.5× throughput improvement over Caffe on popular networks like CaffeNet. Moreover, with these improvements, the end-to-end training time for CNNs is directly proportional to the FLOPS delivered by the CPU, which enables us to efficiently train hybrid CPU-GPU systems for CNNs.

    View details for PubMedID 27314106