Presentation on theme: "FLIPT (Fuzzy Logic and Insulin Pump Technology) University: Tallinn Technical University Professor: Toomas Parve Course: Physiology and Engineering IEM3220."— Presentation transcript:
FLIPT (Fuzzy Logic and Insulin Pump Technology) University: Tallinn Technical University Professor: Toomas Parve Course: Physiology and Engineering IEM3220 Student: Jose Angel Miranda Calero Date: 07/12/2012 1
2 DIABETES DEFINITION: Diabetes mellitus is a group of metabolic diseases in which a person has high blood sugar, either because the pancreas does not produce enough insulin, or because cells do not respond to the insulin that is produced. AMOUNT: 250 million (2011 statistics), 315 million (2012). Representing a 8.2% of the total mortality for any cause. TYPES: Type I. Insulin-dependent. They are the most dependent of insulin injections. Auto-immune reaction -> the bodys defense system attacks the insulin-producing cells. These people dont have or produce enough insulin. High risk. Type II. Non-Insulin dependent. This kind account for more or less 80% of all cases of diabetes. Due to insulin resistance and sometimes combined with a relatively reduced insulin secretion. It appears later and the risk is lower. Gestational Diabetes. Increase the glucose levels during the pregnancy. INSULIN: slow, prorogated action, intermediate action, regular, and fast insulin.
New Approach to diabetic control based in the current GBL and KCAL estimated for each patient and using Fuzzy Controller methodology and Insulin Pump Technology. It does focus on patients with diabetes Type I (most insulin dependent). The goals of this project are: Control the GBL (Glucose Blood Level) continuously in the time as if it was an artificial pancreas. Make the life of diabetic easy and safe. Demonstrate that Fuzzy Controller is the most suitable way to control the glucose blood level in a diabetic patient. Provide a fully automatic response in terms of insulin control and injection. The inclusion of nanotechnology in order to solve medical problems. 3 FLIPT – GOAL Control for Health Monitoring in Diabetic Patients with Fuzzy Controller and Pump Insulin Technology
In order to achieve the goal, this project has to get the next steps: 1. Two basic inputs: current GBL and calories. 2. Step 1. Take the GBL measurement. 3. Step 2. Send this information to the microprocessor. 4. Step 3. Apply the method of control to the measurements taken and estimated through patient profile. 5. Step 4. Send the information to the actuator of the system with the correct amount of insulin to inject. 6. Step 5. Inject the units of insulin specified. 7. Step 6. Repeat this each time interval chosen. 4 FLIPT – CLUSTERS System Clusters SENSING FUZZY LOGIC CONTROLLER
5 FLIPT – CLUSTERS System Clusters NANOSENSOR OR SENSOR NON-INVASIVE TAKEN GBL NANOSENSOR OR SENSOR NON-INVASIVE TAKEN GBL DSP MICRO-CONTROLLER APPLY FUZZY CONTROLLER DSP MICRO-CONTROLLER APPLY FUZZY CONTROLLER INSULIN PUMP INJECTION INSULIN NEEDED INSULIN PUMP INJECTION INSULIN NEEDED HUMAN BODY OR PATIENT BODY HUMAN BODY OR PATIENT BODY GBL MEASUREMENT INSULIN ESTIMATED MEASUREMENT PATIENT PROFILE: -HEIGHT -WEIGHT -AGE -GENDER -PHYSICAL ACTIVITY TOTAL: KCAL (ESTIMATED PER DAY) PATIENT PROFILE: -HEIGHT -WEIGHT -AGE -GENDER -PHYSICAL ACTIVITY TOTAL: KCAL (ESTIMATED PER DAY) 15 MINUTES FAST INSULIN: -React faster to high glucose picks. -Live more flexible life, within a limit. FAST INSULIN: -React faster to high glucose picks. -Live more flexible life, within a limit.
1. Emerging nanotechnologies: Advance and better diagnosis and new devices for medicine through the manufacturing of nanoelectronics based on new CMOS (Complementary Metal Oxide Semiconductor) technologies. This technology is still in progress of manufacturing. 6 FLIPT – CLUSTERS 1. SENSING. 2. Non-Invasive sensor: Substitute to Nanosensor. Aspire TM sensor (Glucon Inc.). It is a sensor with photoacustic technology. Ultrasound waves are generated by illuminating the tissue with laser pulses.
WHY A FUZZY CONTROLLER? 1. Fuzzy rule-based technique is very suitable for biomedical systems. 2. Physiological parameters vary from patient to another. 3. The different life styles difficult to manage the time and the amount to insulin to inject. Thus, we have to have the closest control to real decisions. 4. Variables: defined by linguistic descriptions instead of numerical values. 5. IF-THEN rules describe the relationships between the variables of a system. 9 FLIPT – CLUSTERS 2. Fuzzy Controller.
The embedded system will be: TMS320C6748 DSP, provided by Texas Instruments. 1. TMS320C6748 DSP software and development kit to jump-start real-time signal processing innovation for biometric analytics applications, audio and more. 2. Reduces design work with downloadable and duplicable board schematics and design files 3. Fast and easy development. 4. Low-power TMS320C6748 applications processor. 5. Scalable platform enables a variety of performance, power, peripheral and price options. 6. 128-MByte DDR2 SDRAM. 7. 128-MByte NAND Flash memory. 8. Micro SD/MMC slot. 9. USB and SD connectors. 10 FLIPT – CLUSTERS 3. Embedded System.
Continuous Infusion against multiples injections Insulin PumpInjections Uses only fast insulinFast insulin + slow or mixes ± 2,8% of variabilityMore than 52,0% of variability Very predictableQuite unpredictable Low risk of hypoglycemia High risk of hypoglycemia Enough to achieve a good control In so many cases, it is not enough to achieve a good control Flexible life styleConditioned life by therapy 11 FLIPT – CLUSTERS 4. Pump Insulin. This part of the project is still under development, but we are considering using a specific kind of insulin pumps, which dont have any kind of cannula; thus, it will be better for the patient, lets say less invasive.
1. The inclusion of nanotechnology in order to solve problems related, in this case, with the glucose level is possible. 2. We have achieved to imitate the pancreas behavior controlling the GBL continuously in the time. 3. Has been demonstrated that Fuzzy Controller is the most suitable way to control the glucose blood level in a diabetic patient. Thats because fuzzy adapt perfectly and quickly to unexpected changes in GBL. 4. Our system has a full automatic response, which, in comparison with an insulin pump, is better because we are reacting at the current moment without following any kind of schedule known. And also, the patient doesnt need to interact with the user interface or device, only at the first time during the set up in order to introduce the personal data. 5.The quality of life of a Diabetes Patient will be better using this system to control his GBL. 6.This topic has totally suitability for the course. 12 FLIPT – CONCLUSION
A. A PI-fuzzy logic controller for the regulation of blood glucose level in diabetic patients. M.Ibbini. Balqa Applied Universtity, Al Huson University College, Jordan. 2006. B. Medical nanorobotics for diabetes control. Calvacanti A. Shirinzadeh B. Kretly L. Medical Nanorobotics. 2008. C. A new approach to diabetic control: Fuzzy logic and insulin pump technology. Grant P. Science Direct. August 2006. D. Fuzzy-Based Controller for Glucose Regulation in Type-1 Diabetic Patients by Subcutaneous Route. D.U. Campos-Delgado, M. Hernández-Ordóñez, R. Femat, A. Gordillo-Moscoso. IEEE Transactions on biomedical engineering, Vol 53. NO. 11, November 2006. E. The control of blood glucose in the critical diabetic patient. A neuro-fuzzy method. Dazzi D. Taddei F. Gavarini A. Uggeri E. Negro R. Pezzarossa A. Journal of Diabetes and its Complications 15. (2001). 14 FLIPT – REFERENCES
University: Tallinn Technical University Professor: Toomas Parve Course: Physiology and Engineering IEM3220 Student: Jose Angel Miranda Calero Date: 07/12/2012