IoT Based Expert System for Diabetes Diagnosis and Insulin Dosage Calculation

Abstract

High blood glucose levels are the defining characteristic of diabetes. Uncontrolled blood glucose levels in diabetic patients might result in mortality. As a result, there is a dire need to control blood glucose levels by constantly monitoring them and delivering the appropriate amount of insulin. However, insulin consumption is affected by several variables, including age, calorific intake, and body weight. The patient must see the doctor on a regular basis in order to determine the appropriate dose. Nonetheless, hospital facilities are finding it increasingly difficult to treat patients as the number of patients rises; thus, the healthcare industry is searching for an efficient method that can alleviate their burden by assisting patients with chronic conditions through remote patient care. In this work, we have developed an expert system to provide remote treatment for diabetic patients. Our expert system consists of two distinct components: one for the patient and one for the hospital. The sole requirement for the patient will be a wearable device that captures and transmits all relevant data to the cloud. On the hospital side, there should be a system in place to process that data in the cloud. The system employs a fuzzy system to handle data in two stages. A fuzzy system is initially employed to identify whether or not a patient is diabetic. In the second stage, a fuzzy system is utilized to determine the insulin dosage for a diabetic patient. Using sensors and the ESP8266 platform, we have developed a prototype of patient-side hardware. The MATLAB fuzzy toolbox is used for the processing part, which includes fuzzy systems, and the results of the MATLAB analysis are presented in the form of simulation results to demonstrate the accuracy of the proposed system in terms of determining insulin dosage. The results of the simulation using the fuzzy toolbox for the insulin dose of the diabetic patient are significantly close to the amount of dosage prescribed by the endocrinologist.

Medicina en la era digital: alcances y perspectivas

“…Los profesionales de la salud deben emplearlos para diagnosticar de manera oportuna y tratar de manera personalizada a los pacientes. En una colaboración internacional, un equipo de investigadores de la Universidad de Dhofar, en Omán, y de la Universidad de Sunderland, en Reino Unido, desarrollaron y probaron un prototipo de sistema experto para calcular la dosis de insulina de pacientes con diabetes mellitus tipo 2 (dm2), basándose en mediciones antropométricas, bioquímicas y clínicas, demostrando su relevancia y utilidad (Valsalan et al, 2022). En México, investigadores del imss compararon los resultados de la evaluación del control metabólico de pacientes con dm2 realizada por médicos y por un sistema experto, obteniendo resultados similares y probando su factibilidad operativa (Meza-Palacios et al, 2018).…”

Predicting the Severity of Diabetes Using ECLAT Algorithm in Data Mining

“…Irresistible blood glucose levels in DM victims may lead to death. So it is necessary to monitor the blood glucose level and provide insulin in the required amount advised by endocrinologists [10].…”

Non-Invasive IR-Based Measurement of Human Blood Glucose

“…In reality, the majority of patients cannot feel the change in their glucose level unless, it is very high or very low, without taking a sample. According to the American Diabetes Association, patients with type 1 undergoing intense therapy should test their blood sugar 4-5 times a day, but those with type 2 require only two measurements per day [4]. Unfortunately, the blood glucose monitoring devices that are available in the market only offer a limited number of measurements per day, while a diabetic needs continuous real-time glucose monitoring (CGM) to be informed in case of severe hyperglycaemia or hypoglycaemia.…”