Advancing Diabetic Care Through Non-Invasive Glucose Monitoring Using Optical Sensors and IoT Technologies

<ns7:p>Background Patients diagnosed with diabetes mellitus must monitor their Blood Glucose (BG) to continually control their glycemia. They had to undergo a blood test at least twice a day using a finger prick, and must visit a laboratory every three months for glycated hemoglobin (HbA1c). According to the World Health Organization (WHO), the constitution of Iraqi diabetes patients is 13.9%. An Internet of Things (IoT) based framework for non-invasive BG monitoring was recently developed. The aim is to enhance the Continuous Glucose Monitoring (CGM) device by using IoT technologies. Methods Starting with a review of the latest IoT technologies and some of the CGM commercial devices recommended by ISO 15197 standard. A questionnaire sheet is distributed to determine how much patients in Iraq knew about their health situation and whether they are interested in using new IoT technologies. Women patients are volunteered to check their BG using both finger prick and a CGM device three times a day for comparing. The suggestion is to connect CGM to a personal smart device that has a method controlling alarm messages instead of CGM output. Results The results show how important to educate patients about new technologies. A CGM and finger prick results are checked for similarity using statistical package. Results are presented as (mean ± SD (standard deviation)), the findings demonstrated that, significantly at P < 0.05, there were no differences between the means for tests in both methods based on the standard Ambulatory Glucose Profile (AGP) report. Conclusion This study shows how Iraqi’s patients feel when using a CGM device with new IoT technologies. CGM output is accurate but appeared every 15 min which may be uncomfortable. A method is suggested to transfer CGM output to a smart device to be controlled by algorithms, so that an alarm message is showed every 8h in a normal situation rather than every 15 min. Moreover, GMI% can be calculated every two weeks using data stored in the cloud to estimate a patient’s HbA1c level by using Markov chain. This figure how BG levels are changing in a shorter timeframe, helping in fine-tune diabetes management plans.</ns7:p>