MLwell lab

In this work, we developed a framework for automatically identifying inconsistencies in electronic medical records by comparing clinical data with established medical knowledge. This approach improves data quality assessment and supports the development of more reliable and trustworthy AI in healthcare.

Dr. Amit Haim in cooperation with the Knowledge Center and Shamgar examined how artificial intelligence is defined in laws and policy frameworks around the world, highlighting the challenges of creating a clear and adaptable legal definition. The report proposes a practical framework for defining AI that supports effective regulation while remaining compatible with international standards.

Our researchers investigated how the size of transformer models affects their ability to solve graph-based algorithmic tasks. They showed that wider, shallower models can match the performance of deeper models while enabling faster training and inference.

Our researchers developed SuperMAN (Super Mixing Additive Networks), a novel AI framework for analyzing irregular and complex real-world data, such as medical records collected at different times. The method improves prediction accuracy while providing interpretable insights that can help researchers better understand disease progression and support clinical decision making.

We are hosting the inaugural conference of the knowledge center for AI policy (KNAIP) at Tel-Aviv University on December 24th 2025

In this work we analyzed electronic medical records, trying to understand the risk factors associated with food allergy. We found a significant increase in the rate of food allergies and some potential risk factors.

In this work we demonstrated how machine learning can be used in epidemiology

We held a mini symposium to discuss the guide for Risk Management and Responsible Use of Artificial Intelligence in the Public Sector. This was a collaboration of the Knowledge Center for AI Policy (KNAIP) and the Shamgar Center for Digital Law and Innovation.

Together with the Ministry of Innovation, Science and Technology of Israel we have established a National Knowledge Center for the Implementation of Artificial Intelligence Applications in Government Ministries

We are honored to have received a grant from SPARK@TAU. This collaboration provides us with a unique opportunity to expand the reach of the algorithms we have developed, enabling us to bring them to a wider audience of potential users. We would like to express our sincere gratitude to SPARK@TAU for their support and partnership

This paper provides a comprehensive framework for reporting the use of Natural Language Processing (NLP) methods in scientific research, focusing on transparency and reproducibility. It highlights the critical role of clear documentation in processes such as data collection, preprocessing, model selection, and performance evaluation.

In this work, we address the challenge of ensuring data quality in Electronic Health Records (EHRs), where the focus often shifts to model development rather than the underlying data itself. To fill this gap, we introduce the Medical Data Pecking Tool (MDPT), an innovative solution that utilizes unit-testing techniques to evaluate EHR data quality and its suitability for specific research questions. By combining a dataframe testing tool with a Large Language Model (LLM), MDPT can automatically generate and execute customized evaluations based on predefined criteria such as population traits and regional health patterns, ensuring that the data aligns with expected patterns for various diseases and geographic regions.