Precision on Shinjini Kundu, MD, PhD

The Human Brain

Wed, 08 May 2024 00:00:00 +0000

Unraveling the human brain is sometimes referred to as science’s last frontier. With 100 billion neurons, the brain is the most intricate organ in the human body. By accelerating insights into bridging structure and function, machine learning has the potential to speed up the discovery of new therapeutic targets. These research projects focus on human cortical imaging, mild traumatic brain injury, the effects of exercise on the brain, and craniosynostosis.

Early Osteoarthritis Detection

Tue, 16 Mar 2021 00:00:00 +0000

Osteoarthritis is a common but debilitating disease for which there is no cure. Today, detection is possible only after irreversible damage has happened. This study describes a new machine learning approach for mining early markers of cartilage breakdown, which could potentially help detect osteoarthritis in healthy people three years before symptoms manifest.

COVID-19

Fri, 23 Oct 2020 00:00:00 +0000

Artificial intelligence has the potential to expand the role of chest imaging in COVID-19 beyond diagnosis, enabling new insights at the molecular, health-care system, and population scales. The COVID-19 response can be scaled up with the help of health technology. Dr. Shinjini Kundu described some of these applications in this Radiology: Artificial Intelligence editorial. She was also an agenda contributor to the World Economic Forum in 2020.

Transport-Based Learning

Thu, 15 Feb 2018 00:00:00 +0000

Medical imaging may soon be able to see beyond the limitations of human perception. Yet, most standard machine learning classifiers are unable to explain the logic behind their conclusions - the “black box.” This research develops new approaches for machine learning with built-in expainability. Transport-based learning is an innovative method enabling both automatic discovery and direct visualization of imperceptible patterns within a unified framework.