You are now in the main content area


At IAMLAB, we specialize in the design of AI algorithms that extract insights from medical images. A major focus is on robust design, so AI algorithms can be applied to large, multi-institutional datasets reliably to identify causes and to model progression of diseases.

These technologies enable the development of personalized therapies that can drastically improve patient outcomes by delivering the right treatment at the right time. They also allow for more accurate and efficient diagnosis of diseases.

AI for Neuroimaging

Collage of graphics that illustrate Image Analysis & Machine Learning for Neurological MRI

We are developing novel algorithms that measure biomarkers from magnetic resonance images (MRI) of the brain in order to quantify neurological diseases such as dementia, vascular disease, multiple sclerosis and developmental disorders. These tools are being used to improve accuracy of diagnosis, measure disease progression and to understand more about disease causation.

AI for Digital Pathology

Collage of graphics that illustrate 
Image Analysis & Machine Learning for Breast Cancer Digital Pathology

Just as radiology transformed into a digital practice over 20 years ago, pathology is making the same transition. Wholeslide imaging scanners are digitizing tissue specimens with extremely high resolution, generating digital pathology images that are hundreds of thousands of pixels. We are developing novel algorithms that automatically analyze H&E and IHC breast cancer, renal and colon pathology images. These tools are being used to understand disease etiology and causation, and to improve diagnostic accuracy.

AI for Large Clinical Databases

Collage of graphics that illustrate Artificial Intelligence and Big Data Analytics for Large Clinical Databases

Now that most clinical data is digitized, and stored in central repositories, in presents us with a unique opportunity to leverage computational power and artificial intelligence algorithms to analyze these data sources as a whole. We are developing AI systems to model, integrate, quantify and analyze these large datasets on a much deeper level than previous technology has allowed for. Since these datasets describe the same disease process for the same patient, the result is the ability to model disease in a more personalized manner.