AI Can Help Doctors Diagnose and Treat Patients Faster, Researcher Says
Hamed Akbari, an associate professor of bioengineering at Santa Clara University, uses machine learning to analyze medical data and develop personalized treatment plans for conditions including brain tumors, epilepsy, stroke, and heart disease. He said AI tools could help doctors spend less time on paperwork and more time evaluating patients.
The core challenge in his work is straightforward: people respond differently to medicine, and most patient data remains locked behind privacy protections like HIPAA. AI models need diverse datasets to identify patterns and provide reliable insights. Without access to that data, developing personalized treatments remains difficult.
Where AI Could Make the Biggest Difference
Akbari sees the greatest potential in underserved areas where patients have limited access to specialists. AI could provide advanced diagnostic support in regions where doctors are scarce, allowing physicians to make faster treatment decisions without wasting time or resources.
The efficiency gains extend to the clinic itself. If doctors spend less time documenting cases, they can focus on patient evaluation and diagnosis. This shift could make healthcare more accessible for patients who need immediate attention.
The Research in Practice
Akbari's lab focuses on specific applications: predicting seizure types from MRI scans, analyzing EEG signals to support diagnosis of disorders like schizophrenia and alcoholism, and creating three-dimensional maps of heart ischemia using electrocardiogram data. His students have also developed AI models for seizure prediction and tools to assess COVID-19 severity.
Data preparation and model testing form the foundation of this work. Students propose ideas that often lead to new research directions, making collaboration essential to advancing projects.
What Healthcare Professionals Should Know
Akbari recommends Deep Learning by Ian Goodfellow, Yoshua Bengio, and Aaron Courville for anyone working in biomedical AI. The book covers both theory and practice, with particular value for researchers applying models to medical images, EEG, and ECG signals.
For healthcare professionals looking to understand how AI for Healthcare actually works, the book provides practical insight into model operation and clinical application. Understanding AI Data Analysis methods is increasingly important as these tools move from research labs into clinical practice.
Your membership also unlocks:
