Uri Shalit

Uri Shalit

Auto-generated profile, based on publicly available information

Uri Shalit

Johns Hopkins University Technion Energy Research Technion

Research Focus & Metrics

Academic Career Length

9Years (2018-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
19 April 2021 in Nature Medicine
Verified
Journal Article
Closed access
PDF file
HRHagai Rossman
Hagai Rossman
SSSmadar Shilo
Smadar Shilo
TMTomer Meir
Tomer Meir
MGMalka Gorfine
Malka Gorfine
USUri Shalit
Uri Shalit
ESEran Segal
Eran Segal

20 May 2021 in Computational Linguistics
Verified
Journal Article
Open access
PDF file
AFAmir Feder
Amir Feder
NONadav Oved
Nadav Oved
USUri Shalit
Uri Shalit
RRRoi Reichart
Roi Reichart
Understanding predictions made by deep neural networks is notoriously difficult, but also crucial to their dissemination. As all machine learning–based methods, they are as good as their training data, and can also capture unwanted biases. While there are tools that can help understand whether such biases exist, they do not distinguish between correlation and causation, and might be ill-suited for text-based models and for reasoning about high level language concepts. A key problem of estimating the causal effect of a concept of interest on a given model is that this estimation requires the generation of counterfactual examples, which is challenging with existing generation technology. To bridge that gap, we propose CausaLM, a framework for producing causal model explanations using counterfactual language representation models. Our approach is based on fine-tuning of deep contextualized embedding models with auxiliary adversarial tasks derived from the causal graph of the problem. Concretely, we show that by carefully choosing auxiliary adversarial pre-training tasks, language representation models such as BERT can effectively learn a counterfactual representation for a given concept of interest, and be used to estimate its true causal effect on model performance. A byproduct of our method is a language representation model that is unaffected by the tested concept, which can be useful in mitigating unwanted bias ingrained in the data.

1 March 2021 in Pancreas
Verified
Journal Article
Open access
BKBarbara Kenner
Barbara Kenner
SCSuresh T. Chari
Suresh T. Chari
DKDavid Kelsen
David Kelsen
DKDavid S. Klimstra
David S. Klimstra
SPStephen J. Pandol
Stephen J. Pandol
BWBrian Wolpin
Brian Wolpin
Despite considerable research efforts, pancreatic cancer is associated with a dire prognosis and a 5-year survival rate of only 10%. Early symptoms of the disease are mostly nonspecific. The premise of improved survival through early detection is that more individuals will benefit from potentially curative treatment. Artificial intelligence (AI) methodology has emerged as a successful tool for risk stratification and identification in general health care. In response to the maturity of AI, Kenner Family Research Fund conducted the 2020 AI and Early Detection of Pancreatic Cancer Virtual Summit (www.pdac-virtualsummit.org) in conjunction with the American Pancreatic Association, with a focus on the potential of AI to advance early detection efforts in this disease. This comprehensive presummit article was prepared based on information provided by each of the interdisciplinary participants on one of the 5 following topics: Progress, Problems, and Prospects for Early Detection; AI and Machine Learning; AI and Pancreatic Cancer—Current Efforts; Collaborative Opportunities; and Moving Forward—Reflections from Government, Industry, and Advocacy. The outcome from the robust Summit conversations, to be presented in a future white paper, indicate that significant progress must be the result of strategic collaboration among investigators and institutions from multidisciplinary backgrounds, supported by committed funders.

26 March 2021 in Nature Communications
Verified
Journal Article
Open access
PDF file
HRHagai Rossman
Hagai Rossman
TMTomer Meir
Tomer Meir
JSJonathan Somer
Jonathan Somer
SSSmadar Shilo
Smadar Shilo
RGRom Gutman
Rom Gutman
MGMalka Gorfine
Malka Gorfine
The spread of Coronavirus disease 19 (COVID-19) has led to many healthcare systems being overwhelmed by the rapid emergence of new cases. Here, we study the ramifications of hospital load due to COVID-19 morbidity on in-hospital mortality of patients with COVID-19 by analyzing records of all 22,636 COVID-19 patients hospitalized in Israel from mid-July 2020 to mid-January 2021. We show that even under moderately heavy patient load (>500 countrywide hospitalized severely-ill patients; the Israeli Ministry of Health defined 800 severely-ill patients as the maximum capacity allowing adequate treatment), in-hospital mortality rate of patients with COVID-19 significantly increased compared to periods of lower patient load (250–500 severely-ill patients): 14-day mortality rates were 22.1% (Standard Error 3.1%) higher (mid-September to mid-October) and 27.2% (Standard Error 3.3%) higher (mid-December to mid-January). We further show this higher mortality rate cannot be attributed to changes in the patient population during periods of heavier load.

26 February 2021 in Journal of the American Medical Informatics Association
Verified
Journal Article
Open access
PDF file
MRMichael Roimi
Michael Roimi
RGRom Gutman
Rom Gutman
JSJonathan Somer
Jonathan Somer
AAAsaf Ben Arie
Asaf Ben Arie
ICIdo Calman
Ido Calman
USUri Shalit
Uri Shalit
Objective: The spread of coronavirus disease 2019 (COVID-19) has led to severe strain on hospital capacity in many countries. We aim to develop a model helping planners assess expected COVID-19 hospital resource utilization based on individual patient characteristics. Materials and Methods: We develop a model of patient clinical course based on an advanced multistate survival model. The model predicts the patient's disease course in terms of clinical states—critical, severe, or moderate. The model also predicts hospital utilization on the level of entire hospitals or healthcare systems. We cross-validated the model using a nationwide registry following the day-by-day clinical status of all hospitalized COVID-19 patients in Israel from March 1 to May 2, 2020 (n = 2703). Results: Per-day mean absolute errors for predicted total and critical care hospital bed utilization were 4.72 ± 1.07 and 1.68 ± 0.40, respectively, over cohorts of 330 hospitalized patients; areas under the curve for prediction of critical illness and in-hospital mortality were 0.88 ± 0.04 and 0.96 ± 0.04, respectively. We further present the impact of patient influx scenarios on day-by-day healthcare system utilization. We provide an accompanying R software package. Discussion: The proposed model accurately predicts total and critical care hospital utilization. The model enables evaluating impacts of patient influx scenarios on utilization, accounting for the state of currently hospitalized patients and characteristics of incoming patients. We show that accurate hospital load predictions were possible using only a patient’s age, sex, and day-by-day clinical state (critical, severe, or moderate). Conclusions: The multistate model we develop is a powerful tool for predicting individual-level patient outcomes and hospital-level utilization.

30 March 2020 in medRxiv
Verified
Preprint
Open access
PDF file
GCGil Caspi
Gil Caspi
USUri Shalit
Uri Shalit
SKSøren Lund Kristensen
Søren Lund Kristensen
DADoron Aronson
Doron Aronson
LCLilac Caspi
Lilac Caspi
OCOren Caspi
Oren Caspi
Background: COVID-19 outbreak poses an unprecedented challenge for societies, healthcare organizations and economies. In the present analysis we coupled climate data with COVID-19 spread rates worldwide, and in a single country (USA). Methods: Data of confirmed COVID-19 cases was derived from the COVID-19 Global Cases by the CSSE at Johns Hopkins University up to March 19, 2020. We assessed disease spread by two measures: replication rate (RR), the slope of the logarithmic curve of confirmed cases, and the rate of spread (RoS), the slope of the linear regression of the logarithmic curve. Results: Based on predefined criteria, the mean COVID-19 RR was significantly lower in warm climate countries (0.12±0.02) compared with cold countries (0.24±0.01), (P<0.0001). Similarly, RoS was significantly lower in warm climate countries 0.12±0.02 vs. 0.25 ± 0.01 than in cold climate countries (P<0.001). In all countries (independent of climate classification) both RR and RoS displayed a moderate negative correlation with temperature R= -0.69, 95% confidence interval [CI], -0.87 to -0.36; P<0.001 and R= -0.72, 95% confidence interval [CI], -0.87 to -0.36; P<0.001, respectively. We identified a similar moderate negative correlation with the dew point temperature. Additional climate variables did not display a significant correlation with neither RR nor RoS. Finally, in an ancillary analysis, COVID-19 intra-country model using an inter-state analysis of the USA did not identify yet correlation between climate parameters and RR or RoS as of March, 19, 2020. Conclusions: Our analysis suggests a plausible negative correlation between warmer climate and COVID-19 spread rate as defined by RR and RoS worldwide. This initial correlation should be interpreted cautiously and be further validated over time, the pandemic is at different stages in various countries as well as in regions within these countries. As such, some associations may be more affected by local transmission patterns rather than by climate. Importantly, we provide an online surveillance dashboard ( https://covid19.net.technion.ac.il/ ) to further assess the association between climate parameters and outbreak dynamics worldwide as time goes by. Research in context: Evidence before this study: The coronavirus, COVID-19 pandemic caused by the novel SARS-CoV 2, challenges healthcare organizations and economies worldwide. There have been previous reports describing the association between seasonal climactic variance and SARS-CoV 1 as well as the MERS infections, but the association with SARS-CoV 2 and climate has not been described extensively. Added value of this study: Our analysis demonstrates a plausible negative correlation between warmer climate and COVID-19 spread rate as defined by RR and RoS worldwide in all countries with local transmission as of March 9, 2020. This initial correlation should be interpreted cautiously and be further validated over time. Importantly, we provide an online surveillance dashboard available at ( https://covid19.net.technion.ac.il/ ) for further dynamic tracking of climate effect on COVID-19 disease spread rate worldwide and on intra-country analysis between USA states. Implications of all the available evidence: Our findings of decreased replication and spread rates of COVID-19 in warm climates may suggest that the inevitable seasonal variance will alter the dynamic of the disease spread in both hemispheres in the coming months. However, we warrant a cautious interpretation of these findings given the fact that we are in the initial steps of this outbreak in many “warm” climate countries, the high variance of the data and the dynamic changes in the disease surveillance and the lack of correlation based on the limited data in the US. We hope that the online tool coupling COVID-19 data with climate data will assist in tracking the disease and tailoring the needed measures to contain it.