publications

2024

1. 

   GPT versus Resident Physicians — A Benchmark Based on Official Board Scores

   Uriel Katz , Eran Cohen , Eliya Shachar , Jonathan Somer , Adam Fink , Eli Morse , Beki Shreiber , and Ido Wolf

   NEJM AI, 2024

   abstract Cite paper

   

   A nationwide study found that an artificial intelligence language model outperformed physicians in medical board residency examinations across multiple specialties.

   @article{doi:10.1056/AIdbp2300192,
     author = {Katz, Uriel and Cohen, Eran and Shachar, Eliya and Somer, Jonathan and Fink, Adam and Morse, Eli and Shreiber, Beki and Wolf, Ido},
     title = {GPT versus Resident Physicians — A Benchmark Based on Official Board Scores},
     journal = {NEJM AI},
     volume = {0},
     number = {0},
     pages = {AIdbp2300192},
     year = {2024},
     doi = {10.1056/AIdbp2300192},
     eprint = {https://ai.nejm.org/doi/pdf/10.1056/AIdbp2300192},
     google_scholar_id = {_FxGoFyzp5QC},
   }

2. 

   Temporal tissue dynamics from a single snapshot

   Jonathan Somer , Shie Mannor , and Uri Alon

   bioRxiv, 2024

   abstract Cite paper

   

   Physiological and pathological processes such as inflammation or cancer emerge from the interactions between cells over time. However, methods to follow cell populations over time within the native context of a human tissue are lacking, since tissue biopsy offers only a single snapshot. Here we present one-shot tissue dynamics reconstruction (OSDR), an approach to estimate a dynamical model of cell populations based on a single tissue sample. OSDR uses spatial proteomics data to learn how the composition of cellular neighborhoods influences division rate, providing a dynamical model of cell population change over time. We apply OSDR to human breast cancer data, and reconstruct two fixed points of fibroblasts and macrophage interactions. These fixed points correspond to hot and cold fibrosis, in agreement with co-culture experiments that measured dynamics directly. We then use OSDR to discover a pulse-generating excitable circuit of T and B cells in the tumor microenvironment, suggesting temporal flares of adaptive anti-cancer responses. OSDR can be applied to a wide range of spatial transcriptomic or proteomic assays to enable analysis of tissue dynamics based on patient biopsies.

   @article{somer2024temporal,
     title = {Temporal tissue dynamics from a single snapshot},
     author = {Somer, Jonathan and Mannor, Shie and Alon, Uri},
     journal = {bioRxiv},
     pages = {2024--04},
     year = {2024},
     publisher = {Cold Spring Harbor Laboratory},
     eprint = {https://www.biorxiv.org/content/10.1101/2024.04.22.590503v1},
     google_scholar_id = {LkGwnXOMwfcC},
   }

2022

1. 

   Analyzing continuous physiologic data to find hemodynamic signatures associated with new brain injury after congenital heart surgery

   Jessica Nicoll , Jonathan Somer , Danny Eytan , Vann Chau , Davide Marini , Jessie Mei Lim , Robert Greer , Safwat Aly , Mike Seed , Steven P Miller , and  others

   Critical Care Explorations, 2022

   abstract Cite paper

   

   Continuously acquired physiologic measurements for up to 72 hours after cardiac surgery were analyzed for association with new brain injury by MRI. Mixed-effects regression analyses characterized relationships between HR, BP, and Spo2 and new brain injury over time while accounting for variation between patients, measurement heterogeneity, and missingness.

   @article{nicoll2022analyzing,
     title = {Analyzing continuous physiologic data to find hemodynamic signatures associated with new brain injury after congenital heart surgery},
     author = {Nicoll, Jessica and Somer, Jonathan and Eytan, Danny and Chau, Vann and Marini, Davide and Lim, Jessie Mei and Greer, Robert and Aly, Safwat and Seed, Mike and Miller, Steven P and others},
     journal = {Critical Care Explorations},
     volume = {4},
     number = {9},
     pages = {e0751},
     year = {2022},
     publisher = {LWW},
     google_scholar_id = {W7OEmFMy1HYC}
   }

2021

1. 

   Development and validation of a machine learning model predicting illness trajectory and hospital utilization of COVID-19 patients: a nationwide study

   Michael Roimi , Rom Gutman , Jonathan Somer , Asaf Ben Arie , Ido Calman , Yaron Bar-Lavie , Udi Gelbshtein , Sigal Liverant-Taub , Arnona Ziv , Danny Eytan , and  others

   Journal of the American Medical Informatics Association, 2021

   abstract Cite paper

   

   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.

   @article{roimi2021development,
     title = {Development and validation of a machine learning model predicting illness trajectory and hospital utilization of COVID-19 patients: a nationwide study},
     author = {Roimi, Michael and Gutman, Rom and Somer, Jonathan and Ben Arie, Asaf and Calman, Ido and Bar-Lavie, Yaron and Gelbshtein, Udi and Liverant-Taub, Sigal and Ziv, Arnona and Eytan, Danny and others},
     journal = {Journal of the American Medical Informatics Association},
     volume = {28},
     number = {6},
     pages = {1188--1196},
     year = {2021},
     publisher = {Oxford University Press},
     google_scholar_id = {2osOgNQ5qMEC}
   }

2. 

   Blood pressure in critically ill children: Exploratory analyses of concurrent invasive and noninvasive measurements

   Andrew Goodwin , Mjaye L Mazwi , Jonathan Somer , Steven M Schwartz , Alistair McEwan , and Danny Eytan

   Critical Care Explorations, 2021

   abstract Cite paper

   

   Blood pressure is a basic vital sign reflecting the cardiovascular status of the patients that can be sampled either invasively using an arterial line (IABP) or intermittingly using an oscillometric blood pressure cuff (NIBP). Differences and biases between the invasive and cuff measurements have been reported in adults and children and yet, at the bedside, clinicians are often faced with the challenge of assigning a confidence to a specific cuff blood pressure measurement and acting upon the measurement. We hypothesized that big data could be used to quantify the relationship between NIBP and IABP measurements and its dependence on patient characteristics in a manner that may aid bedside decision making. We developed an online tool to explore these distributions interactively with the goal of supporting decisions on the need for invasive monitoring in a data-driven manner.

   @article{goodwin2021blood,
     title = {Blood pressure in critically ill children: Exploratory analyses of concurrent invasive and noninvasive measurements},
     author = {Goodwin, Andrew and Mazwi, Mjaye L and Somer, Jonathan and Schwartz, Steven M and McEwan, Alistair and Eytan, Danny},
     journal = {Critical Care Explorations},
     volume = {3},
     number = {12},
     pages = {e0586},
     year = {2021},
     publisher = {LWW},
     google_scholar_id = {IjCSPb-OGe4C}
   }

3. 

   Hospital load and increased COVID-19 related mortality in Israel

   Hagai Rossman , Tomer Meir , Jonathan Somer , Smadar Shilo , Rom Gutman , Asaf Ben Arie , Eran Segal , Uri Shalit , and Malka Gorfine

   Nature communications, 2021

   abstract Cite paper

   

   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

   @article{rossman2021hospital,
     title = {Hospital load and increased COVID-19 related mortality in Israel},
     author = {Rossman, Hagai and Meir, Tomer and Somer, Jonathan and Shilo, Smadar and Gutman, Rom and Ben Arie, Asaf and Segal, Eran and Shalit, Uri and Gorfine, Malka},
     journal = {Nature communications},
     volume = {12},
     number = {1},
     pages = {1904},
     year = {2021},
     publisher = {Nature Publishing Group UK London},
     google_scholar_id = {Y0pCki6q_DkC}
   }

2020

1. 

   Developing a COVID-19 mortality risk prediction model when individual-level data are not available

   Noam Barda , Dan Riesel , Amichay Akriv , Joseph Levy , Uriah Finkel , Gal Yona , Daniel Greenfeld , Shimon Sheiba , Jonathan Somer , Eitan Bachmat , and  others

   Nature communications, 2020

   abstract Cite paper

   

   At the COVID-19 pandemic onset, when individual-level data of COVID-19 patients were not yet available, there was already a need for risk predictors to support prevention and treatment decisions. Here, we report a hybrid strategy to create such a predictor, combining the development of a baseline severe respiratory infection risk predictor and a post-processing method to calibrate the predictions to reported COVID-19 case-fatality rates. With the accumulation of a COVID-19 patient cohort, this predictor is validated to have good discrimination (area under the receiver-operating characteristics curve of 0.943) and calibration (markedly improved compared to that of the baseline predictor). At a 5% risk threshold, 15% of patients are marked as high-risk, achieving a sensitivity of 88%. We thus demonstrate that even at the onset of a pandemic, shrouded in epidemiologic fog of war, it is possible to provide a useful risk predictor, now widely used in a large healthcare organization.

   @article{barda2020developing,
     title = {Developing a COVID-19 mortality risk prediction model when individual-level data are not available},
     author = {Barda, Noam and Riesel, Dan and Akriv, Amichay and Levy, Joseph and Finkel, Uriah and Yona, Gal and Greenfeld, Daniel and Sheiba, Shimon and Somer, Jonathan and Bachmat, Eitan and others},
     journal = {Nature communications},
     volume = {11},
     number = {1},
     pages = {4439},
     year = {2020},
     publisher = {Nature Publishing Group UK London},
     google_scholar_id = {Tyk-4Ss8FVUC},
   }