Saudi Arabia is making significant strides in AI-powered healthcare, particularly through the efforts of two young researchers from the Mohamed bin Zayed University of Artificial Intelligence. Mohammed Firdaus Ridzuan and Tooba Tehreem Sheikh are developing essential tools for hospitals in the Kingdom that will enhance diagnosis and prognosis capabilities.

Credit: Arab News

Their research aligns with Saudi Arabia’s healthcare transformation plans under Vision 2030, focusing on creating intelligent, interpretable, and scalable systems. As the country implements AI systems across various diagnostic units, the emphasis is shifting towards ensuring these technologies provide clarity and real-time support.

Ridzuan, who holds a PhD in machine learning, has created the Human-in-the-Loop for Prognosis (HuLP), a cancer survival prediction system that reintegrates doctors into AI-driven decision-making. He noted, “While AI has made significant strides in diagnosing diseases, predicting individual survival outcomes, especially in cancer, is still a challenging task,” adding that his model allows for real-time clinician intervention.

Unlike conventional AI models that function independently, HuLP is designed for collaboration, enabling medical professionals to modify and refine predictions based on their expertise. Ridzuan explained, “Doctors and medical professionals can actively engage with the system. Their insights don’t just influence the result — they actually help the model learn.”

This collaborative approach ensures that predictions are explainable and tailored to individual patient needs, which is vital for the AI integration occurring in Saudi hospitals. Ridzuan emphasized that enabling clinicians to adjust predictions enhances the system’s adaptability to local healthcare challenges.

Saudi Arabia’s healthcare institutions are undergoing a digital transformation supported by national entities such as the Saudi Data and AI Authority (SDAIA) and the Ministry of Health. These organizations are focused on incorporating new AI tools into clinical workflows, a key area where Ridzuan believes HuLP can make a significant impact.

“Smart hospitals are already integrating AI diagnostic tools for medical imaging and patient data analysis,” he stated. “Our model can take this to the next level by empowering clinicians to interact with and guide the system’s predictions.” This could address a persistent issue in AI, known as the black box effect, where decision-making processes are opaque.

Additionally, Ridzuan highlighted the importance of local data for developing context-sensitive AI models. “Local data provides insights into the unique health conditions and medical practices within the Gulf region,” he said, ensuring that AI solutions meet the specific needs of patients in the area.

Sheikh, an MSc graduate in computer vision, is also making significant contributions with her project, Med-YOLOWorld, a next-generation imaging system capable of interpreting nine different medical scans in real time. Unlike traditional AI tools that focus on single modalities, Med-YOLOWorld employs open-vocabulary detection, allowing it to identify a broader range of anomalies.

“Most models are confined to a single modality like CT or X-ray,” Sheikh explained, noting that her system supports multiple imaging types, including ultrasound and histopathology. With a capability of processing up to 70 frames per second, it is designed for use in high-demand clinical settings.

Sheikh faces technical challenges, such as managing the varying preprocessing requirements for different imaging modalities. “The biggest challenge was managing the diverse preprocessing requirements across imaging modalities,” she stated. To address data imbalance issues, she developed custom augmentation techniques to ensure consistent performance across all imaging types.

Moreover, Sheikh is exploring the integration of Med-YOLOWorld with vision-language models to enhance its functionality. “MiniGPT-Med does a great job at explaining radiology images,” she noted, emphasizing that combining it with her system could provide comprehensive diagnostic capabilities.

Both Ridzuan and Sheikh are exemplifying how innovation can be aligned with the healthcare realities and goals of the region. As Saudi Arabia continues to invest in advanced healthcare solutions, their work on HuLP and Med-YOLOWorld may soon lead to a more intelligent and human-centered approach to medical care.