International Research Journal of Engineering and Technology (IRJET) Volume: 09 Issue: 05 | May 2022 www.irjet.net
e-ISSN: 2395-0056 p-ISSN: 2395-0072
GLUCAGON: AI-Based Insulin Dosage Prediction Application Jayakrishnan S1, Job Shaji2, Sandeep S3, Vrinda Sukumar S4, Sherin M Wilson5 1,2,3,4 UG
Scholar, Department of Computer Science and Engineering, Prof, Department of Computer Science and Engineering, UKF College of Engineering and Technology, Kerala, India ------------------------------------------------------------------***------------------------------------------------------------------------Abstract:- The automation of insulin treatment is the most challenging aspect of glucose management for type 1 diabetes 5Asst.
owing to unexpected exogenous events (e.g., meal intake). In this article, we propose a reinforcement learning (RL) algorithm based on artificial intelligence (AI) for an application which predicts the optimized insulin dosage using the datasets obtained from CGM and activity band which continuously collects data from the victim. A bio-inspired RL designing method was developed for automated data integration. This strategy uses reward functions to represent the temporal homeostatic goal, as well as discount factors to represent an individual's unique pharmacological profile. The proposed strategy was tested in virtual patients from the FDA-approved UVA/Padova simulator with unscheduled meal intakes using a training method based on an RL algorithm. The trained policy demonstrated fully automated regulation in both the basal and postprandial phases for a single-meal experiment with pre-prandial fasting. The layer-by-layer relevance propagation gives interpret-able data on AI-driven decisions for sensor noise robustness, automatic postprandial management, and avoidance of insulin stacking. The accuracy of the application was also tested by comparing with conventional manner of blood glucose checking.
Keywords: Diabetes Mellitus, Reinforcement Learning, Artificial Neural Networks 1. INTRODUCTION Medical science has been progressing under the shades of technological advancement and technical support. The capabilities of human being has been overridden by machines and programmed devices that ensures the right treatment for patients and accurate test and validation process that eradicates any risk. This progression is a big leap in comparison with old ways of treatment and care. Despite of any unhealthy life style and biogenic diseases, people now a days suffer from hereditary syndromes. Type 1 diabetes (T1D) is the most common condition which is considerably in above average levels of diagnosis. It is also called Diabetes mellitus or Juvenile diabetes as it is commonly diagnosed in children below age 13. It is an irreversible condition where the immune cells of the victim's body accidentally attacks the beta cells of the pancreas and completely destroys its ability to produce necessary insulin and thus shuts down the endocrine system. Any physical body needs energy for existence and it can only be gained when insulin converts the glucose and starch in blood into energy and store it in the cells to be utilized. In case of T1D patient, the victim suffers from terrible fatigue and nausea. Numerous technological advancements has been introduced to overcome and reverse this condition but it's been only managed to control and regulate the blood glucose level manually and doesn't has any progression in rejuvenating the pancreas system other than surgeries. This paper, introduces an ideology of mimicking the beta cells of pancreas so as it could successfully and accurately generate insulin dosages by understanding the victim's physical environment by collecting necessary data instantly. It has been reported that around 70% of people around the world has been using conventional prescribed insulin dosages that may invite short term or long term consequences and risks the physical condition of the victim. This is happening because of miscalculated insulin dosages that doesn't relate to the person's homeostasis at all. A study conducted by University of Texas regarding diabetes mellitus found that numerous critical conditions and consequences such as kidney failure, lung disease, brain cells degradation, muscle degradation, diabetic retinopathy, cataract etc. has been affecting people with T1D even if they are on prescribed medical routine. This issue is the main concern that is being concentrated in this paper.
2. EXISTING SYSTEM 2.1 Review of Existing System A conventional manner of injecting pre-set insulin dosage for a T1D patient is considered a risky and inconvenient strategy. It may result in over dosage of insulin which may be a reason for the hypoglycemic state where a person could experience © 2022, IRJET
|
Impact Factor value: 7.529
|
ISO 9001:2008 Certified Journal
|
Page 894