Our Main Areas of Focus:
Creating preventative medicine methods and technologies that can stop non-infectious disease in its tracks
At the intersection of biology, data collection, and machine learning, faculty and students at MIT are working on the development of new methods and technologies that have the potential to change the course of noninfectious disease by stopping it in its tracks. The promise is better outcomes across a wide range of conditions that affect millions—from cancer and neurodegenerative disorders to chronic illnesses such as diabetes, kidney disease, and asthma—as well lower costs associated with disease detection.
Making An Impact
Led by Regina Barzilay, the Delta Electronics Professor of Electrical Engineering and Computer Science, MIT researchers focused on breast cancer are collaborating with Massachusetts General Hospital (MGH) to develop new machine learning technologies that combine multifaceted patient data with biological insights targeted toward specific patient characteristics.
Developing cost-effective diagnostic tests to detect and alleviate health problems
MIT researchers are working on embedding machine learning techniques into devices such as wearables and wireless biosensors that may be able to both detect and alleviate health problems. With these low-cost devices, machine learning and data inference can be performed in a small footprint, mitigating the need for large, cloud-based infrastructures. The potential for broad reach and global impact is high.
Making An Impact
Feng Zhang, the James and Patricia Poitras Professor in Neuroscience, and James Collins, the Termeer Professor of Medical Engineering and Science, have pioneered a technology called SHERLOCK (Specific High-sensitivity Enzymatic Reporter unLOCKing) that adapts the CRISPR gene-editing technology to target RNA and can be used as a rapid, inexpensive diagnostic tool.
Discovering and developing new pharmaceuticals that can be tailored to
the individual patient
Conventional methods of discovering new therapeutics are declining in productivity while rising in cost. Machine learning has the powerful potential to address these challenges. This revolutionary approach could change drug manufacturing by determining the most efficient and cost-effective way to produce a given chemical compound. MIT’s work focuses on designing molecules with target biological properties, which is essential for drug tailoring and personalized medicine.
Making An Impact
A cross-disciplinary team of MIT researchers including William Green, Klavs Jensen, Regina Barzilay, and Tommi Jaakkola have developed a machine learning algorithm trained to predict outcomes of chemical reactions, assess product properties, and select optimal conditions for reactions. After training on close to a million reactions, the algorithm outperforms human experts.
