HBKU AI Job Augmentation
AI impacts jobs in different ways and could improve the efficiency and productivity of humans. This project plans to develop an AI solution to augment jobs at QCRI. The project involves identifying departments or jobs that could benefit from AI augmentation and determining the pain points in these departments. Once the pain points are identified, […]
Examining the impact of AI on jobs from a gender perspective
AI could potentially have an impact on women’s economic empowerment and labour market opportunities by leading to job automation. Recent research by the IMF and the Institute for Women’s Policy Research found that women are at a significantly higher risk of displacement due to job automation than men. In this research, we are investigating the […]
MedQoder – The Automated Medical Coder
Medical coding is an essential step in hospital revenue cycle. Medical coders perform the task of assigning for each patient visit, one or more codes from the international standard of medical coding systems such as ICD for diagnosis or CPT for procedure. This task is performed based on manually reviewing the discharge notes and other […]
Computational Pathology
Digital Pathology has created numerous opportunities for machine learning and artificial intelligence. Several repetitive and tedious tasks done by pathologists can be automated or assisted by AI models. Whilst most existing research focus on adult cancer applications, we aim to focus on pediatrics applications. Accurate and timely diagnosis in pediatric departments would lead to an […]
Robust ML models
The aim of the project is to design ML models which are robust against adversarial attacks. The specific use-case is to make credit scoring models robust. We will not only explore model robustness against adversarial attacks but also concept-drift and covariate-shift.
YouRule: encouraging children’s physical activity by coding the rules of the game
The YouRule system aims to encourage both physical activity and learning to code for children and teenagers. The system lets end-users program the rules of a sports game (any kind they can invent), then allows them to wear sensors and play that game physically by the game rules they coded (It is not a video […]
Visual Analytics of Wearable Data to Improve Health and Wellness
Diabetes and obesity are major health issues in Qatar (Qatar National Vision 2030). We designed InViTAG, a visual analytic platform to support healthcare professionals in improving the health and wellness of patients with diabetes or obesity based on wearable and biometric data. Based on user feedback from previous studies, the present project aims to finalize […]
Safe reinforcement learning
We have two projects related to safe reinforcement learning:
Augmented Intelligence using for Data Preparation
Transformer-based models (e.g., BERT, RoBERTa, XLNet) and giant language models (e.g., GPT-3 and T0pp) have a good potential to learn knowledge from multi-modal data, such as text, tables, and so on. This learned knowledge, if being used appropriately, can significantly help practitioners reduce human cost in terms of laborious data preparation tasks. This project aims […]
Augmented Intelligence for Personalized Patient Lifestyle Improvement based on Wearable Data
Health is one of the pillars of Qatar National Vision 2030. Qatar Foundation has recently restated these objectives as “individualized healthcare and disease prevention driven by emerging research, clinical approach, environment, and lifestyle”. Diabetes and obesity are significant health problems globally and are particularly prevalent in Qatar. These serious conditions are driven by lifestyle factors […]