Medical Information Analytics Lab

In the Medical Information Analytics Labs, we research ideas for innovative healthcare technologies using Artificial Intelligence. We focus on combining different data sources and training multimodal prediction models. We collaborate with several hospitals including Soroka medical center, Shamir medical center (formerly Assaf Harofeh medical center), and Beit Rivka geriatric center.

Our research covers the following areas:

Expanding drug knowledge, we use more than twenty data sources to train machine learning models for a variety of tasks:
1. Classifying drug safety properties such as drug-drug interactions, side effects, and drug safety during pregnancy
2. Drug repurposing
3. Synergistic drug classification
4. Molecule generation using reinforcement learning
Text mining using medical data
Treatment outcome prediction

Materials:

Watch Guy Shtar describing some of our key research subjects (available only in Hebrew at the moment). The video can be downloaded from here.
A presentation given by Guy Shtar @ WSDM 2021 doctoral consortium is also available.
A website for drug pregnancy safety prediction developed in our lab.

(12/2021) We have identified an approved drug with anti-cancer potential using graph neural networks. In a collaboration with Prof. Shimon Ben-Shabat this result was validated in-vitro in currently being tested in-vitro.

(01/2021) Our medical information lab received funding from the Israeli innovation authority (bioconvergence track)

Published papers:

Guy Shtar, Lior Rokach, Bracha Shapira, Elkana Kohn, Matitiahu Berkovitch, Maya Berlin, Explainable multimodal machine learning model for classifying pregnancy drug safety, Bioinformatics, 2021;, btab769, https://doi.org/10.1093/bioinformatics/btab769
Shtar G. Multimodal Machine Learning for Drug Knowledge Discovery. InProceedings of the 14th ACM International Conference on Web Search and Data Mining 2021 Mar 8 (pp. 1115-1116).
Shauloff, N., Morag, A., Yaniv, K. et al. Sniffing Bacteria with a Carbon-Dot Artificial Nose. Nano-Micro Lett. 13, 112 (2021). https://doi.org/10.1007/s40820-021-00610-w
Itchaki G, Rokach L, Benjamini O, Bairey O, Sabag A, Vernitsky H, Cohen H, Rotem S, Elia U, Raanani P, Bar-Haim E. Cellular Immune Responses to BNT162b2 mRNA COVID-19 Vaccine in Patients with Chronic Lymphocytic Leukemia. Blood. 2021 Nov 23;138(Supplement 1):638-.
Tadmor T, Benjamini O, Braester A, Rahav G, Rokach L. Antibody persistence 100 days following the second dose of BNT162b mRNA Covid19 vaccine in patients with chronic lymphocytic leukemia. Leukemia. 2021 Sep;35(9):2727-30.
Cohen S, Dagan N, Cohen-Inger N, Ofer D, Rokach L. ICU Survival Prediction Incorporating Test-Time Augmentation to Improve the Accuracy of Ensemble-Based Models. IEEE Access. 2021 Jun 22;9:91584-92.

Shtar G, Rokach L, Shapira B, Nissan R, Hershkovitz A. Using Machine Learning to Predict Rehabilitation Outcomes in Post-acute Hip Fracture Patients. Archives of Physical Medicine and Rehabilitation. 2020 Sep 16.
Shtar G, Rokach L, Shapira B, Berkovitch M, Dinavitser N, Cohen R, De Haan T, Kohn E, Berlin M. Artificial Intelligence-Game Changer in the Teratology Information Service. Reproductive Toxicology (Elmsford, NY). 2020 Oct 1;97:3-4.
Benjamini O, Rokach L, Itchaki G, Braester A, Shvidel L, Goldschmidt N, Shapira S, Dally N, Avigdor A, Rahav G, Lustig Y. Safety and efficacy of BNT162b mRNA Covid19 Vaccine in patients with chronic lymphocytic leukemia. Haematologica. 2020.
Shtar G, Rokach L, Shapira B. Detecting drug-drug interactions using artificial neural networks and classic graph similarity measures. PloS one. 2019 Aug 1;14(8):e0219796.