Examination of Health Data Depending on Creative Use of Optimization Methods and Machine Learning Algorithms

Abstract

Health science is one of the areas where predictions and foresightedness are significant. This study aims to show that with the help of machine learning, decisions being made in medical life can be supported for better results. The precision of mathematics and the combination of existing technology and existing systems can help improve health care. A decision system in this regard is not intended to replace doctors. Undoubtedly, the final medical decision lies with the experts. Therefore, these systems based on machine learning approaches are generally for suggestions only. Some automated systems are useful for two important reasons. The first is to avoid the time-consuming simple, usual things and avoid overloading the medical equipment with many activities that can be easily performed by a machine, for example, verifying whether a patient’s medical characteristics are recommending that person for a particular job or the best time to supply a particular drug to the store. Another is to indicate medical conditions that are hardly detectable by humans; a suggestion or warning can change the severity of the patient and that would mean life or death, for example, to detect certain cancerous cells or to predict the risk of a genetic disease in an image. It can eliminate or minimize problems that may arise due to human-specific reasons (stress, fatigue, reduced attention). To show the appliance of these technologies, a case to find cancer types on a patient will be shown. The proposed combination of the optimization algorithms with the selected machine learning methods also shows that predictions in health care can be improved by using these algorithms creatively and can be improved drastically if done correctly. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.