Real Time Monitoring and Controlling Device for Elderly Personnel with Hypoglycemia

Size: px

Start display at page:

Download "Real Time Monitoring and Controlling Device for Elderly Personnel with Hypoglycemia"

Isabella Hill
5 years ago
Views:

1 Real Time Monitoring and Controlling Device for Elderly Personnel with Hypoglycemia Dr. S.SHIYAMALA Associate Professor/ ECE Veltech Dr. RR &Dr. SR University, Avadi, Chennai Dr. T.KAVITHA Associate Professor/ ECE Veltech Dr. RR &Dr. SR University, Avadi, Chennai ABSTRACT Hypoglycemia is a common side-effect of insulin therapy for patients with type 1 diabetes mellitus (T1DM). Due to this the deficiency in glucose counter-regulation may even lead to severe hypoglycemia. It leads to seizures or loss of consciousness and the possible development of permanent brain death under certain circumstances. Separated elderly personnel without assistant struggle a lot with hypoglycemia, for themtwo different methods are proposed to save the elderly personal from hypoglycemia, firstly with the help of biosensor continuously monitor the sugar level especially during night time and inject the Glipizide when it needs and secondly automatically induce the nasal spray based on embedded system. Keywords Hypoglycemia, Embedded System, Elder personnel I. INTRODUCTION In India 34 million people in rural and 28 million people in urban affected by the Type 1 sugar.phyo et al. [1] discussed extreme learning machine (ELM) based hypoglycemia detection system is developed to recognize the presence of hypoglycemic episodes. From a clinical study of 16 children with T1DM, natural occurrence of nocturnal hypoglycemic episodes are associated with increased heart rates (p < 0.06) and increased corrected QT intervals (p < 0.001). The overall data were organized into a training set with 8 patients (320 data points) and a testing set with 8 patients (269 data points). By using the ELM trained feed-forward neural network (ELM-FFNN), the testing sensitivity (true positive) and specificity (true negative) for detection of hypoglycemia is 78 and 60% respectability. For comparison and analysis purposes, five methods were applied to the hypoglycemia detection problem as following: 1. Proposed feed-forward neural network trained with ELM with 15 hidden nodes (FFNN -ELM); 2. Feed-forward neural network trained with particle swarm optimization (FFNN -PSO); 3. Second order multiple regress fuzzy inference system 4. Fuzzy inference system with 5 membership functions for each input (FIS); 5. Linear multiple regression (LMR). Xue Mo et al. [2] described extreme learning machines (ELM) and regularized ELM (RELM). Under three different prediction horizons, 10, 20, and 30 min, these two methods were compared systematically in terms of root mean square error (RMSE), sensitivity, and specificity. The area under curve (AUC) value was used the evaluate the ROC performance for different test accurately. The experiment results demonstrate that these two methods can predict hypoglycemia pretty good. As expect, the bigger prediction horizon (PH), induce the worse performance. As hypoglycemia threshold is increased, sensitivity improves at cost of specificity. Both methods can get good specificity and acceptable sensitivity. Good specificity can make sure each alarm is effective for patients to take correct actions. In terms of AUC, ELM and RELM have comparable performance for hypoglycemia prediction. 128

2 Identification of Hypoglycemia and Hyperglycemia in Type 1 Diabetic patients using ECG parameters was studied by Linh et al.[3]. An ECG Acquisition and Analysis System based on LabVIEW software has been developed for collecting ECG signals and extracting features with abnormal changes for type 1 diabetic patients experienced by both hypoglycemic and hyperglycemic. ECG parameters included Heart rate (HR), corrected QT interval (QTe C), PR interval, corrected RT interval (RTC) and corrected TpTe interval (TpTe C). Blood glucose levels were used to classify glycemic states in subjects as hypoglycemic state ( 60 mml/l, Hypo), as normoglycemic state (80 to 110 mml/l, Normo), and as hyperglycemic state (150 mml/l, Hyper). The results indicated that hypoglycemic and hyperglycemic states produce significant inverse changes on those ECG parameters. [4]HypoMon was a realtime non-invasive monitor that measures relevant physiological parameters continuously to provide detection of hypoglycemic episodes in Type 1 diabetes mellitus patients (T1DM). Based on heart rate and corrected QT interval of the ECG signal, we have continued to develop effective algorithms for early detection of nocturnal hypoglycemia. Using the optimal Bayesian neural network which was derived from the training set with the highest log evidence, the estimated blood glucose profiles produced a significant correlation (P<0.02) against measured values in the test set. Quantitative measurement of sugar concentration using in house fabricated microgap biosensor was discussed by Rao et al. [5]. In this work, studies and analysis were conducted based on different concentration of sugar on constant sized microgap and based on different size of microgap structure on constant sugar concentration. As the result, the experiment has been successfully yielded a high sensitive microgap sensor and the lowest detected sugar concentration sample is gml-1. This method of biosensing provides a very simple and promising detection technique for any kind of biomolecules that includes proteins, DNA, enzymes, antibody and antigen. Recent Advances in Application of Biosensors in Tissue Engineering was discussed by Hasan et al. [6]. Thesensing efficiency and detection limit of the graphene-based glucose biosensors were found to increase when silver nanoparticle (AgNP)/AuNP hybrid was used to catalyze electrochemical reaction of GOD. Akhtar et al. [7] described thecurrent Trends in Nanomaterial-Based Amperometric Biosensors Amperometric biosensors based on nanomaterials may offer various advantages for enhancing and superseding the capabilities of current analytical methodology by permitting rapid and highly accurate analysis. The main aim of this paper is to design the device which is useful for elderly personal those who felt with type 1 diabetes. The rest of the paper has been organized as follows. Section 2 presents the overall view of hypoglycaemia Section 3 discusses the various types of sweat sensors. Section 4 deals the innovative product viz. nasal spray. Limitation while designing the new product is discussed in section 5. Finally, conclusions with further scope are presented in section 6. II. HYPOGLYCEMIA AN OVER VIEW Diabetes mellitus is a syndrome of disordered metabolism due to a combination of hereditary and environmental causes. Classification of diabetes are Type 1: Lack of insulin. Type 2: Cells resistance to insulin For the Type 1 insulin must be injected or inhaled and for the Type 2 treatment is food control and exercise. Hypoglycemia (60mg/ dl or less) is a common side-effect of insulin therapy for patients with type 1 diabetes mellitus (T1DM). It is the major limiting factor to maintain tight glycemic control. Possible symptoms of hypoglycemia include: Anxiety Blurred vision 129

Confusion Dizziness Fatigue, sleepiness, weakness Heart palpitations Hunger Nervousness Nightmares Shakiness Sweating Tremor III. SWEAT SENSORS A.

3 Confusion Dizziness Fatigue, sleepiness, weakness Heart palpitations Hunger Nervousness Nightmares Shakiness Sweating Tremor III. SWEAT SENSORS A. Sweatronics Patch Design Multiple layer patch design that manages sweat flow to enable measurement of analytes in real-time coupled with the wireless transmission of the measured data to remote, interpretive applications and monitoring/alert systems. A typical patch system is composed of layered materials and electronics that include: 1. sweat-porous adhesive 2. modular, flexible electronic circuitry and sensors 3. sweat management layer, and, 4. breathable top protective textile Although all layers of the patch system contribute to its function, and may be altered for market-specific requirements, most critical is the flexible electronics and sensing layer. Fig 1 shows the patch design. Fig 1: Sweatronics Patch Design B. Sleep Sentry Diabetes brings the risk of hypoglycemia, low blood sugar. Most often the result of imbalance between food, exercise, and medications, a "hypo," a low blood sugar reaction, can cause disorientation, unconsciousness, and sometimes, the need for an ambulance. The Sleep Sentry is intended to warn type 1 diabetics of a hypoglycemic reaction, low blood sugars, occurring while they are asleep. It is actually designed to monitor the temperature and moisture level of the skin; symptoms normally associated with insulin reactions. If it does not detect one of these symptoms, if a low blood sugar reaction occurs without them, the alarm will not sound, and it will probably be of no use to the diabetic. Remember, the alarm is intended to awaken the user, if there has been a temperature drop, or if it is detecting increased perspiration. The alarm does not directly indicate hypoglycemia. If the alarm goes off, and, after checking your blood sugar, you find it is in the normal range, the alarm could have been triggered by its wrist band being too loose, by an extreme change in the room temperature, or if the user perspires excessively or 130

4 exhibits a drop in skin temperature for reasons other than hypoglycemia, as examples, fever, menopause, or as a side effect of medications. Diabetes Sentry Products tells me the Sleep Sentry was successful, in initial clinicals 90 percent of the time, in detecting hypoglycemia. The 10 percent failure rate was generally attributed to user error. Some people do not wear the device during the daytime because they sweat more then, which can trigger a false alarm. If you have concerns about undetectable low blood sugar, you might wear the device during the day, as a precaution. Again, the Sleep Sentry is intended for a type 1 diabetic, but can be worn by type 2s as well, to warn them of hypoglycemia. There is a condition called hypoglycemia unawareness, in which the diabetic loses the ability to tell that s/he is going down, and the next sound you hear may be a siren. What many of us need is a device to warn us we are getting low when we cannot tell for ourselves -- sufficient to warn us to act before we are so far into the hypo we can t help ourselves. (NOTE: The FDA has approved this device for nighttime use.) Fig 2 : Sleep Sentry Wrist watch IV. NASAL SPRAY The nasal spray needs no mixing. Either diabetic or a bystander can shoot it up the person s nose, where the glucagon is absorbed by the mucous membranes in the nasal passages. The nasal spray needs no mixing. Either the diabetic or a by stander can shoot up the person s nose where the glucagon is absorbed by the mucous membrane in the nasal passages said Dr. Deena (waiting for FDA approval) V. CONSTRAINTS Limitations to be consider to design a new device Different Age Group Different types of Sugar level (Type 1, Type 2) Sensor Selection Embed the latest ELM in the Glucose monitoring System Low Cost Tiny Size, low weight machine Low power A. Different Age Group While designing the device, according to the patient age, insulin injection get varied. Youth with type 1 diabetes T1D (Kellee Miller et al.)ages fourteen to seventeen had more episodes of Diabetic Ketoacidosis (DKA), whereas children under the age of five had more episodes of Severe Hypoglycemia (SH) compared to other age groups. B.Different types of Sugar level (Type 1, Type 2) 131

5 Type 1 people only felt with hypoglycemia. Due to high sugar long term damage to the body for the Type 2 patient. If the sugar level crossing beyond the limitation, as per FDP approval insulin level may be reduced. Single device can manage type 1 and type 2 at critical condition may encourage. C.Sensor Selection Sensor selection is a crucial one. Embed the latest biosensor or sweat sensor in the device is useful for increasing the device efficiency. D.Embed the latest ELM in the Glucose monitoring System Latest early learning machine should be embed with the device to avoid the time delay. E.Low Cost This product should be reached to the below poverty people also. While designing this device, cost is the predominant factor. F.Tiny Size, low weight machine With the help of nano technology size of the component should be reduced since throughout night time patient should wear this machine. G.Low power Power consumption is the crucial scenario in the electronic device. By changing algorithm and architecture we can achieve mitigated power. VI.CONCLUSION Various techniques are available in the market to monitor the sugar level. But mere monitoring is not sufficient to avoid the effects due to hypoglycemia. In this crucial condition, have to integrate the monitoring device with controlling device. In this paper, two different concepts are discussed with limitations. REFERENCES [1] PhyoPhyo San, Sai Ho Ling Senior Member, IEEE, Ni NiSoe, and Hung T. Nguyen Senior Member, IEEE A Novel Extreme Learning Machine for Hypoglycemia Detection 2014, IEEE 36 th Annual International Conference on Engineering. pp [2] Xue Mo ; Youqing Wang ; Xiangwei Wu Hypoglycemia prediction using extreme learning machine (ELM) and regularized ELM 25th Chinese Control and Decision Conference (CCDC), 2013, pp [3]Linh Lan Nguyen ; ; Su, S. ; Nguyen, H.T. Identification of Hypoglycemia and Hyperglycemia in Type 1 Diabetic patients using ECG parameters IEEE Annual international conference on Engineering in Medicine and Biology Society (EMBS),2012,pp [4]Hung T. Nguyen, NejhdehGhevondian, Timothy W. Jones Real-time Detection of Nocturnal Hypoglycemic Episodes using a Novel Non-invasive Hypoglycemia Monitor IEEE EMBS 31st Annual International Conference, 2009, pp [5] Rao, B.S.; Nurfaiz, M.; Hashim, U. Quantitative measurement of sugar concentration using in house fabricated microgap biosensor IEEE Regional Symposium on Micro and Nanoelectronics (RSM), 2013, pp [6] Hasan, Anwarul, MdNurunnabi, MahboobMorshed, Arghya Paul, Alessandro Polini, Tapas Kuila, Moustafa Al Hariri, Yongkyu Lee, and Ayad A. Jaffa Recent Advances in Application of Biosensors in Tissue Engineering. BioMed Research International 2014 (1): [7] Akhtar Hayat ; GaëlleCatanante ;Jean Louis Marty, Current Trends in Nanomaterial-Based Amperometric Biosensors Sensors 2014, 14, ; doi: /s

Non-invasive Hypoglycemia Monitoring System using Extreme Learning Machine for Type 1 diabetes

Non-invasive Hypoglycemia Monitoring System using Extreme Learning Machine for Type 1 diabetes Sai Ho Ling, Phyo Phyo San, and Hung T. Nguyen Abstract Hypoglycemia (low blood glucose level) is a medical