The Department of Computer Science (DCS) at Notre Dame University-Louaize North Lebanon Campus (NDU-NLC) organized on Wednesday, February 21, 2018, a lecture titled, “Artificial Intelligence in Healthcare,” presented by guest speaker Dr. Sandy Rihana, associate professor and head of the Department of Biomedical Engineering at Holy Spirit University Kaslik (USEK). Artificial Intelligence (AI) was introduced to students and a closer look was taken at how AI is solving healthcare-related challenges. The adoption of AI in healthcare is on the rise and is greatly impacting patients, hospitals, and the healthcare industry in general.
This presentation began by demonstrating the need for AI, driven by the amount of digital data, which is growing at a mind-blowing speed, doubling every two years. Dr. Rihana then presented the different areas in which AI is already implemented and successfully used (for example in our cars, in Google Searches, Amazon suggestions, and in many devices). She cited several examples of AI applications in healthcare (for example Google DeepMind Health, which is an acute kidney management tool). The application alerts the clinician when a sudden deterioration of kidney function occurs. Another application of AI in healthcare is IBM Watson, which has the ability to understand the context in a medical file (using Natural Language Processing) and analyze relevant portions of the electronic medical record.
Dr. Rihana explained that AI is divided into three categories:
- AI itself, defined broadly and covering all possible approaches to simulating intelligence.
- A subset of AI is Machine Learning, which uses data and experience automatically to tune algorithms.
- A subset of Machine Learning is Deep Learning, which uses brain inspired algorithms, such as Neural Networks to simulate the learning process.
She also went on to cite other examples in detail of each of the above categories, and the advantages of each, concluding with the following quote from Gartner: “By 2018, deep learning (deep neural networks) will be a standard component in 80% of data scientists’ toolboxes.”
A Q&A session followed where attendees asked a host of questions related to the current and future trends of AI in healthcare.