Machine Learning for Personalized Medicine: Clinical Outcome Prediction and Diagnosis : Plenary Talk
Žarko Ćojbašić · 2019
Machine learning and computational intelligence have been applied to a wide range of medical problems to assist medical professionals in decision-making. Especially artificial neural networks, fuzzy systems and powerful hybrid neuro-fuzzy approaches have already proven their strong potentials in medicine. This is especially important and interesting in emerging field of personalized medicine, which is often described as providing ”the right patient with the right drug at the right dose at the right time” and represents tailoring of medical treatment to the individual patient characteristics, needs and preferences. Machine learning focuses on the development of computer programs that can access data and use it to learn by themselves. Regarding medical applications concerning prediction of patient's clinical outcome, machine learning can be considered as a data-driven analytic approach that specializes in the integration of multiple risk factors into a predictive tool. In this talk hypothesis has been considered that machine learning based models may help to improve prediction of clinical outcome in various medical fields, in comparison to traditional statistical and scoring approaches. This hypothesis has been tested with our own results from several studies. Our experiences have been considered regarding artificial neural networks based prediction of cerebral palsy in infants with central coordination disturbance, adaptive neuro-fuzzy estimation of autonomic nervous system parameters effect on heart rate variability, machine learning leukemia clinical outcome prediction, neural prediction of mortality in spontaneous intracerebral hemorrhage based on initial clinical parameters and others. Finally, to demonstrated potentials of machine learning in medical diagnosis, our results concerning cervical cancer detection by improving standard screening tests have been considered.