On The Survival Assessment of Diabetic Patients Using Machine Learning Techniques.

International Journal of Research and Innovation in Applied Science (IJRIAS) | Volume VII, Issue I, January 2022 | ISSN 2454–6194

On The Survival Assessment of Diabetic Patients Using Machine Learning Techniques.

¹Adeboye, Nureni Olawale (PhD), ²Adesanya, Kehinde Kazeem.
¹Department of Mathematics & Statistics, Federal Polytechnic, Ilaro, Ogun State, Nigeria
²Department of Health Information Management, Ogun State College of Health Technology. Ilese ijebu Ode, Ogun state Nigeria.

Abstract
The extraordinary improvement in biotech and medical sciences have given rise to an impactful data production from stour Electronic Health Records (EHRs), and it has contributed significantly to the Kaggle source from which the data for this research was obtained. The dataset consists of 1416 recorded cases of diabetic patients from 130 various hospitals in the United States. This study thus assesses the survival rate of diabetic patients using machine learning techniques, and determined the duration it will take a diabetic patient to survive based on the application of the most appropriate algorithm. The research tested the application of four different algorithms which include support vector machine, logistic regression, decision tree and k-nearest neighbors’ algorithm. In line with their accuracy measured by f1-score, precision, recall and support metrics; k-nearest neighbors is seen to outperform all other algorithms for predicting the survival rate of the patients. The research also revealed that it takes a diabetic patient 30 days to survive if the patient is placed on medications according to the available information, and that the medication given to the diabetic patients is less effective in the aged patients and more effective among the younger patients.
Keywords: Accuracy, Algorithms, Diabetic Patients, Machine Learning, Survival rate.

1. Introduction
Pertinent researches in the area of technology that uses biological systems and living organisms to develop different outputs results unremittingly in a self-evident and economic data production, thereby heralding the science of biotechnology into the realm of big data. In addition to this lofty performance, there is a myriad of electronic machines from various research fields which culminate into data generation, and these include Super-Resolution Microscopy, Spectrometry Technologies for biomolecules and small molecules, Magnetic Resonance Spectroscopy, just to mention few. Though these technologies produce valuable data, but they do not give researchers insight into the analytical meaning of the generated data. Thus, Knowledge Discovery in Biological data has becomes essential and logically inescapable; the primary aim is mainly to research into the rapidly increasing body of such official data and set the basis for providing genuine responses to fundamental questions in biological and medical sciences ref. [6]. In the hybrid field of biotechnology, Diabetes Mellitus (DM) is one of the mostly diagnosed ailments in the categories of human-threatening and life quality reducing diseases ref. [2].
According to ref. [13], DM is a metabolic disorder in which the amount of sugar in the blood is increased beyond necessary. Insulin deficiency increases the glucose levels in the blood and subdue the metabolism of carbohydrates, fat and proteins. It is the most normal endocrine issue, affecting more than 415 million individuals in the entire world. Diabetes development is emphatically connected to hormonal and metabolic issues, brought about by constant hyperglycemia. Diabetes covers a wide scope of heterogeneous pathophysiological conditions. Difficulties like harmed nerves, eyes, kidneys, and different organs might emerge when high glucose from diabetes is not treated on schedule.
According to [20], diabetes is categorized into two major clinical types according to the etiopathology of the disorder, which are Type 1 diabetes (T1D) and Type 2 diabetes (T2D). 90% of all diabetic patients are known to be suffering from T2D and thus regarded as the most common form of diabetes, and [21] emphasized that the increasing burden of T2D has become a major concern in healthcare management. T2D is mainly characterized by insulin resistance and the main causes include but not limited to poor Medicare, dietary habits and heredity. T1D on the other hand, is opinioned to be due to auto immunological destruction caused by a chronic condition in which the pancreas produces little or no insulin. T1D usually manifest in adolescence and it has been established to affects almost 10% of all diabetic patients globally, resulting in symptoms such as increased thirst, frequent urination, fatigue, blurred vision and hunger. Other classifications of DM on the basis of insulin secretion profile include Endocrinopathies, Gestational, Mitochondrial, MODY (Maturity Onset Diabetes of the Young), Neonatal, and Pregnancy diabetes.

Read More