Aviv A. Rosenberg

I’m a PhD student in the VISTA Lab at the Technion Computer Science Faculty, advised by Prof. Alex Bronstein. I am also a machine learning engineer and data scientist at Sibylla, where I lead the design and implementation of our ML platform.

My research focuses on implementing machine learning tools for practical use in medicine and biology, including modeling and analysis of heart rate variability, detection of abnormalities in ECG signals, statistical methods such as conformalization and quantile regression, quantification of uncertainty, and reliability in deep learning systems in medical applications.

I hold an MSc in Biomedical Engineering and a BSc in Electrical Engineering, both from the Technion. Previously head TA of the CS faculty’s Deep Learning course (2019-2022).

In addition to a my research pursuits, I have a strong technical background in software engineering. Before starting my PhD I worked in multiple technical roles. I have extensive experience writing robust, testable, well-designed code, leading software teams and architecting large-scale systems.

news

May 20, 2022	New paper in Nature Communications.
Dec 8, 2021	Gutwirth Excellence Scholarship for PhD research awarded.
Jul 23, 2021	Our Deep Learning in Cardiology paper featured by Technion.
Jun 14, 2021	New paper in PNAS about deep learning in cardiology.
Aug 24, 2020	MLIS Data Science Scholarship awarded.
Jul 19, 2020	Our new paper was published in Nature Scientific Reports.

selected publications

Preprint

Fast Nonlinear Vector Quantile Regression

Aviv A. Rosenberg, Sanketh Vedula, Yaniv Romano, and Alex M. Bronstein

May 2022

Abs HTML PDF

Quantile regression (QR) is a powerful tool for estimating one or more conditional quantiles of a target variable Y given explanatory features X. A limitation of QR is that it is only defined for scalar target variables, due to the formulation of its objective function, and since the notion of quantiles has no standard definition for multivariate distributions. Recently, vector quantile regression (VQR) was proposed as an extension of QR for high-dimensional target variables, thanks to a meaningful generalization of the notion of quantiles to multivariate distributions. Despite its elegance, VQR is arguably not applicable in practice due to several limitations: (i) it assumes a linear model for the quantiles of the target Y given the features X; (ii) its exact formulation is intractable even for modestly-sized problems in terms of target dimensions, number of regressed quantile levels, or number of features, and its relaxed dual formulation may violate the monotonicity of the estimated quantiles; (iii) no fast or scalable solvers for VQR currently exist. In this work we fully address these limitations, namely: (i) We extend VQR to the non-linear case, showing substantial improvement over linear VQR; (ii) We propose vector monotone rearrangement, a method which ensures the estimates obtained by VQR relaxations are monotone functions; (iii) We provide fast, GPU-accelerated solvers for linear and nonlinear VQR which maintain a fixed memory footprint with number of samples and quantile levels, and demonstrate that they scale to millions of samples and thousands of quantile levels; (iv) We release an optimized python package of our solvers as to widespread the use of VQR in real-world applications.
NatComm

Codon-specific Ramachandran plots show amino acid backbone conformation depends on identity of the translated codon

Aviv A. Rosenberg, Ailie Marx, and Alex M. Bronstein

Nature Communications May 2022

Abs HTML PDF

Synonymous codons translate into chemically identical amino acids. Once considered inconsequential to the formation of the protein product, there is evidence to suggest that codon usage affects co-translational protein folding and the final structure of the expressed protein. Here we develop a method for computing and comparing codon-specific Ramachandran plots and demonstrate that the backbone dihedral angle distributions of some synonymous codons are distinguishable with statistical significance for some secondary structures. This shows that there exists a dependence between codon identity and backbone torsion of the translated amino acid. Although these findings cannot pinpoint the causal direction of this dependence, we discuss the vast biological implications should coding be shown to directly shape protein conformation and demonstrate the usefulness of this method as a tool for probing associations between codon usage and protein structure. Finally, we urge for the inclusion of exact genetic information into structural databases.
PNAS

Meeting the unmet needs of clinicians from AI systems showcased for cardiology with deep-learning–based ECG analysis

Yonatan Elul, Aviv A. Rosenberg, Assaf Schuster, Alex M. Bronstein, and Yael Yaniv

Proceedings of the National Academy of Sciences May 2021

Abs HTML PDF

The use of artificial intelligence (AI) in medicine, particularly deep learning, has gained considerable attention recently. Although some works boast superior capabilities compared to clinicians, actual deployments of AI systems in the clinic are scarce. We describe four important gaps on the machine-learning side responsible for this discrepancy by first formulating them in a way that is actionable by AI researchers and then systematically addressing these needs. Aiming beyond the search for better model architectures or improved accuracy, we focus directly on the challenges of clinical usefulness as stated by medical professionals in the literature. Our results show that deep-learning systems can be robust, trustworthy, explainable, and transparent while retaining the superior level of performance these algorithms are known for.Despite their great promise, artificial intelligence (AI) systems have yet to become ubiquitous in the daily practice of medicine largely due to several crucial unmet needs of healthcare practitioners. These include lack of explanations in clinically meaningful terms, handling the presence of unknown medical conditions, and transparency regarding the system’s limitations, both in terms of statistical performance as well as recognizing situations for which the system’s predictions are irrelevant. We articulate these unmet clinical needs as machine-learning (ML) problems and systematically address them with cutting-edge ML techniques. We focus on electrocardiogram (ECG) analysis as an example domain in which AI has great potential and tackle two challenging tasks: the detection of a heterogeneous mix of known and unknown arrhythmias from ECG and the identification of underlying cardio-pathology from segments annotated as normal sinus rhythm recorded in patients with an intermittent arrhythmia. We validate our methods by simulating a screening for arrhythmias in a large-scale population while adhering to statistical significance requirements. Specifically, our system 1) visualizes the relative importance of each part of an ECG segment for the final model decision; 2) upholds specified statistical constraints on its out-of-sample performance and provides uncertainty estimation for its predictions; 3) handles inputs containing unknown rhythm types; and 4) handles data from unseen patients while also flagging cases in which the model’s outputs are not usable for a specific patient. This work represents a significant step toward overcoming the limitations currently impeding the integration of AI into clinical practice in cardiology and medicine in general.All study data are included in the article and/or SI Appendix.
Nat. Sci. Rep.

Signatures of the autonomic nervous system and the heart’s pacemaker cells in canine electrocardiograms and their applications to humans

Aviv A Rosenberg, Ido Weiser-Bitoun, George E Billman, and Yael Yaniv

Nature Scientific Reports May 2020

Abs HTML PDF

Heart rate and heart rate variability (HRV) are mainly determined by the autonomic nervous system (ANS), which interacts with receptors on the sinoatrial node (SAN; the heart’s primary pacemaker), and by the “coupled-clock” system within the SAN cells. HRV changes are associated with cardiac diseases. However, the relative contributions of the ANS and SAN to HRV are not clear, impeding effective treatment. To discern the SAN’s contribution, we performed HRV analysis on canine electrocardiograms containing basal and ANS-blockade segments. We also analyzed human electrocardiograms of atrial fibrillation and heart failure patients, as well as healthy aged subjects. Finally, we used a mathematical model to simulate HRV under decreased “coupled-clock” regulation. We found that (a) in canines, the SAN and ANS contribute mainly to long- and short-term HRV, respectively; (b) there is evidence suggesting a similar relative SAN contribution in humans; (c) SAN features can be calculated from beat-intervals obtained in-vivo, without intervention; (d) ANS contribution can be modeled by sines embedded in white noise; (e) HRV changes associated with cardiac diseases and aging can be interpreted as deterioration of both SAN and ANS; and (f) SAN clock-coupling can be estimated from changes in HRV. This may enable future non-invasive diagnostic applications.
Frontiers

PhysioZoo: a novel open access platform for heart rate variability analysis of mammalian electrocardiographic data

Joachim A Behar, Aviv A Rosenberg, Ido Weiser-Bitoun, Ori Shemla, Alexandra Alexandrovich, and 2 more authors

Frontiers in Physiology May 2018

HTML PDF