A University of Kansas professor of computer science has edited a new book on advanced artificial intelligence (AI) systems that can improve the accuracy, affordability and accessibility of health care.
Arvin Agah, who is also the associate dean for research and graduate programs in the School of Engineering, brought together over 80 experts with 50 unique affiliations from 17 different countries for “Medical Applications of Artificial Intelligence,” which was published in November 2013 by CRC Press. The book provides an overview of AI concepts and techniques before delving into innovative research and continued challenges in integrating AI into medicine.
“As AI techniques are being further developed, their applications to medicine are expanding, and this book is an attempt to capture the breadth and depth of medical applications of AI,” said Dr. Agah.
AI concepts, techniques, and tools have been utilized in medical applications for over four decades with the goal to benefit health care by assisting health care professionals. Improvements in accuracy and efficiency of AI techniques have steadily increased AI’s viability as a choice for tackling problems in medicine.
Current diagnostic tools could lead to incorrect or delayed diagnoses in 10 to 20 percent of cases, according the Journal of the American Medical Association. AI tools, such as data mining and machine learning, improve time to diagnose by combing through patient histories, treatment plans, research, and other information. They summarize data, identify abnormalities, and provide assistance in diagnosis and treatment, says Dr. Agah. By using computer simulations to evaluate numerous treatment options and reassess treatment as new information becomes available, doctors can decide on the best course of action. The book highlights how AI systems are helping doctors in the early detection of vision disorders and melanoma and cervical cancers.
“Medical Applications of Artificial Intelligence” explores a variety of efficiencies and best practices that can lead to cost reductions in health care. Experts examine information retrieval tools, from secure electronic records to searches that highlight similar cases. Another chapter highlights support systems that enable efficient communication among all medical personnel and support staff and serve as additional checks on medication management and patient care. A recent study found AI tools cut patient treatment costs in half, while offering significantly better results.
AI systems provide greater accessibility in research and treatment. Data mining and other AI tools enable the analysis of massive genetic data, leading to a better understanding of cancer and chemical toxicity. Another chapter explores a new more affordable, efficient screening of Pap Smears that will give women in developing countries greater access to potentially life-saving tests. In the operating room, surgeons are using sensors and AI tools to gather additional information during minimally invasive surgeries.
Since joining KU in 1997, Dr. Aagh has published more than 160 refereed articles in artificial intelligence and autonomous robots. He has been a co-investigator on projects that represent more than $32 million in research funding. He has supervised 14 Ph.D. students and 40 M.S. students and received multiple honors for his teaching excellence. He is a senior member of IEEE and ACM.