Prajakta Pardeshi
Position:
Department: Lab: eMail: Homepage: |
PhD Researcher
Instrumentation and Control Embedded Systems Lab prjakta.pardeshi@gmail.com, prajakta.pardeshi@mitwpu.edu.in http://www.coepembeddedlab.com/pardeshi |
Biography
Prajakta Pardeshi is a research scholar at Department of Instrumentation and Control, College of Engineering, Pune. Her field of research includes Model Predictive Control and Embedded Control, Communication and Signal processing. She had earned her post-graduation degree in Electronics & Telecommunication from Smt. Kashibai Navale College of Engineering, Pune and graduation in Signal Processing- Electronics & Telecommunication from Smt. Kashibai Navale College of Engineering, Pune. Currently she is working on the Biomedical research, she has previously worked on Communication and Signal processing and on projects like implementing smart laboratory, virtual laboratory. Currently she is working with Dr. Vishwanath Karad MIT World Peace University, as an Assistant Professor in Electronics and Communication Engineering since July 2015. She has 4.6 years of experience in teaching.
Personal
Research Interests
- Model Predictive Control
- Embedded Systems
- Optimization
PhD Thesis
Learning-based Predictive Control for Glucose Regulation in Type-1 Diabetes
Abstract:
Diabetes mellitus (DM) also called diabetes, is prolonged metabolic disorder due to high blood sugar level. Type 1 or insulin-dependent is the more severe form of diabetes. It usually develops in children, teenagers or at any stage of life. The control of diabetes is an interdisciplinary attempt which includes a significant biomedical engineering component. With identifying individuals at risk, developing robust control model is equally important. Despite the fact, that technologies on diabetes treatment keep advancing from time to time, blood glucose control problem is still being a crucial issue raised by many scientist and researchers. In early days it started with injecting insulin, then began with modelling of insulin-glucose system, and progressed to various in-silico experiments, and automated to close-loop control Firstly it started with classic minimal modelling approach and discuss a number of subsequent models which has resulted in the first in silico model accepted as substitute to animal trials in the need for optimal diabetes control. The said research focuses on studying existing diabetes pk-pd control models and developing a full-fledged close loop glucose regulation system comprising of various Learning based MPC algorithms. A universal, accurate, reliable and simple mathematical model is being focused in developing a control algorithm on blood glucose level. As there’s no benchmark control system for diabetes. One would hope that the insulin delivery mechanism, can be simplified, and life quality of patient can be improved. It is hoped that newer generations of continuous glucose monitors will have improved signal transmission and accuracy, and avoid the need for finger-stick glucose calibration.
Abstract:
Diabetes mellitus (DM) also called diabetes, is prolonged metabolic disorder due to high blood sugar level. Type 1 or insulin-dependent is the more severe form of diabetes. It usually develops in children, teenagers or at any stage of life. The control of diabetes is an interdisciplinary attempt which includes a significant biomedical engineering component. With identifying individuals at risk, developing robust control model is equally important. Despite the fact, that technologies on diabetes treatment keep advancing from time to time, blood glucose control problem is still being a crucial issue raised by many scientist and researchers. In early days it started with injecting insulin, then began with modelling of insulin-glucose system, and progressed to various in-silico experiments, and automated to close-loop control Firstly it started with classic minimal modelling approach and discuss a number of subsequent models which has resulted in the first in silico model accepted as substitute to animal trials in the need for optimal diabetes control. The said research focuses on studying existing diabetes pk-pd control models and developing a full-fledged close loop glucose regulation system comprising of various Learning based MPC algorithms. A universal, accurate, reliable and simple mathematical model is being focused in developing a control algorithm on blood glucose level. As there’s no benchmark control system for diabetes. One would hope that the insulin delivery mechanism, can be simplified, and life quality of patient can be improved. It is hoped that newer generations of continuous glucose monitors will have improved signal transmission and accuracy, and avoid the need for finger-stick glucose calibration.
Publications
Book Chapters:
- Gulve S.P., Khoje S.A., Pardeshi P. (2017) Implementation of IoT-Based Smart Video Surveillance System. In: Behera H., Mohapatra D. (eds) Computational Intelligence in Data Mining. Advances in Intelligent Systems and Computing, vol 556. Springer, Singapore
- M. Chavan, P. Pardeshi, S. A. Khoje and M. Patil, "Study of Health Monitoring System," 2018 Second International Conference on Intelligent Computing and Control Systems (ICICCS), Madurai, India, 2018, pp. 1779-1784, doi: 10.1109/ICCONS.2018.8663063.
- R. Dak, S. Khoje and P. J. Pardeshi, "A review on cost-efficient virtual laboratory: A teaching aid," 2017 International Conference on Computing Methodologies and Communication (ICCMC), Erode, 2017, pp. 663-667, doi: 10.1109/ICCMC.2017.8282549.
- P. Balap, S. Khoje and P. Pardeshi, "A critical review on wavelet based feature extraction for Iris Recognition," 2017 11th International Conference on Intelligent Systems and Control (ISCO), Coimbatore, 2017, pp. 235-240, doi: 10.1109/ISCO.2017.7855988.