The process of normal cell division in the human body is quite simple: start dividing in response to a signal, such as a wound, and stop when enough cells have been produced and the skin is healed. But cancerous cells…
Electrical Engineering, Computer Science Researchers Win Artificial Intelligence Award
A research collaboration between electrical engineering and computer science researchers and colleagues at Upstate Medical University on detecting Alzheimer’s disease won a notable award at an artificial intelligence conference. Professors Asif Salekin and Senem Velipasalar, EECS graduate students Fatih Altay and Guillermo Ramon Sánchez, along with doctors Yanli James and Stephen V. Faraone from Upstate Medical University won the IAAI-21 Deployed Application Award at the 33rd Annual Conference on Innovative Applications of Artificial Intelligence.
The team’s research centers on early detection of Alzheimer’s disease. The most common symptoms of Alzheimer’s disease include problems with communicating and abstract thinking, as well as disorientation. Early detection of the disease can help improve cognitive functioning with medication and training. The research paper from the Syracuse University/ Upstate Medical University team proposes two machine learning approaches for detecting Alzheimer’s disease from MRI images to help early detection efforts at a preclinical stage before symptoms have appeared.
In their paper the team described the impact their research could have. “Recent reports on Alzheimer’s disease (AD) suggest that change in the brain may be evident 20 years before dementia symptoms, typically when the disease gets diagnosed. But substantial neuronal loss happens during that latent period of the disease. The early-stage intervention of AD can significantly impact the neuronal degeneration process and treatment of symptoms that would expand the patients’ life expectancy and quality of life. Hence, accurate detection or indication of preclinical AD is a major interest in the medical community. Our research is the first to develop an effective machine learning approach that can identify the latent patterns due to preclinical AD from MRI brain scans, which can significantly improve AD patients’ intervention and treatment.”