In Abu Dhabi, doctors can now use an artificial intelligence-based assistant to search and summarize data from extensive patient medical records. Even information collected in different locations and at different times—such as an allergy recorded in one hospital, blood test results from another, or an old scan—can contain important details about a person's health.
The Abu Dhabi Department of Health has integrated such a physician assistant into Malaffi, the emirate's unified medical information network. This allows specialists to ask the system questions instead of manually reviewing numerous documents.
This assistant is capable of extracting laboratory reports, identifying documented illnesses, and providing summaries of a patient's medical history. It is available to doctors across Abu Dhabi to reduce the time spent studying medical charts and allow them to focus more on treatment itself.
The system relies on a network that includes over four billion clinical records, covering more than 14 million patient profiles, according to data from the Department of Health and M42. These records include health status reports, allergy information, prescribed medications, test results, vital signs, and diagnostic images collected from various medical institutions.
Malaffi, which translates to 'my card' in Arabic, was launched in 2019 to connect medical information between clinics and hospitals in Abu Dhabi. It enables authorized healthcare workers to access patient records even if previous consultations or tests were conducted at a different facility.
Although the network already provided access to this information, the new AI assistant is intended to speed up the search for specific details, especially when a patient's medical history spans several years and involves different medical organizations.
Karim Shahin, CEO of Abu Dhabi Health Data Services, which manages Malaffi, noted: 'Malaffi has played an important role in unifying medical information among healthcare providers, and this integration adds a practical tool that helps doctors better utilize this data in their daily work.'
Querying Patient Information via AI
During an appointment, a doctor may be required to review past diagnoses, check for registered allergies, or analyze changes in laboratory results over time. With the new assistant, medical professionals can ask questions about available records and receive answers summarizing relevant information. The system is designed to support the review of a patient's medical history, especially when data is scattered across multiple documents.
The technology operates based on Ossi—a clinical AI platform developed by the Abu Dhabi-based company M42 and hosted on Microsoft Azure infrastructure in the UAE. Ossi was introduced in October and activated in seven M42 medical facilities in the UAE and Bahrain. The broader platform includes features such as clinical documentation drafting and consultation support.
Its implementation in Malaffi focuses on extracting and summarizing information from existing patient medical records, bringing this capability into the network used by healthcare workers across Abu Dhabi.
Dimitris Molavasilis, Group CEO of M42, stated that the technology was developed with the needs of doctors in mind, including helping them study medical information more effectively. He added: 'The integration of the Assistant into Malaffi brings these capabilities into the environment of connected medical data. We will continue to develop and evaluate its performance in collaboration with healthcare professionals, supporting clinical use while maintaining the central role of the physician in assessment and decision-making.'
Responsibility for Decisions Remains with Doctors
The assistant functions within the existing Malaffi access permissions and patient data protection requirements. It is created for healthcare professionals, not for patients who directly consult the AI chatbot for advice. Doctors retain responsibility for interpreting the received information, verifying its accuracy, and making decisions regarding diagnosis and treatment.
Studies on AI-generated medical summaries have demonstrated potential in assisting doctors with complex medical histories. However, a 2025 study concerning heart failure records did not show a statistically significant improvement in overall task completion time, although doctors maintained their accuracy when using AI summaries.
M42 reported that the implementation will be evaluated based on physician feedback and daily usage results, with subsequent phases planning scheduled improvements. For patients visiting various hospitals, clinics, or specialists, the system provides an additional way for doctors to access information already stored in the interconnected Abu Dhabi healthcare network.

