MiniGPT-Med Model
The head of the scientific team at SDAIA, who worked on creating the model, Dr. Ahmed bin Zaki Al-Sannan, the AI consultant at the National Center for Artificial Intelligence, said: “The MiniGPT-Med model was designed in a way that allows users to input medical images and perform a number of tasks, including generating medical reports, answering medical visual questions, detecting the location of the disease, recognizing diseases, and producing medically-based descriptions.”
X-rays, CT scans, MRIs
He added that the model has been trained on various medical images (X-rays, CT scans, and MRIs), noting that recent advancements in artificial intelligence have contributed to the development of the healthcare field. However, these advancements are often limited by specific functions. Therefore, the MiniGPT-Med model will address this issue. It is derived from large-scale language models and is specifically designed for medical applications, demonstrating remarkable versatility across different imaging methods. It also included X-rays, CT scans, and MRIs, which enhances its utility.
Al-Senan pointed out that the MiniGPT-Med model was developed by AI specialists from SDAIA and KAUST. It achieved advanced performance in generating medical reports that is 19% higher than the best previous model. It serves as a general interface for radiology diagnostics, enhancing diagnostic efficiency across a wide range of medical imaging applications.
KAUST, VQA
Related Topics :
SADAIA, UNESCO Discuss Cooperation in AI Ethics
How Has Kingdom Achieved Leadership in Information Security?
National Guard Health Prepares Medical Clinics, Sunstroke Center for Hajj 2024