Sensors and smartphone based real-time heart minitoring, alarming and reporting systems | |
| Author | Priyanka Kakria |
| Call Number | AIT Diss. no.RS-16-04 |
| Subject(s) | Smartphones Sensors |
| Note | A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Engineering in Remote Sensing and Geographic Information Systems |
| Publisher | Asian Institute of Technology |
| Series Statement | Dissertation ; no. RS-16-04 |
| Abstract | The World is experiencing a high rate of cardiac diseases during last few decade s in which h eart diseases have become one of the leading causes of death . According to World Health Organization (WHO) , the cardiovascular diseases are the world’s largest killers, claiming 17.1 million lives in a year. The hospital admission mainly focused in the old er adults segment, range from a prevalence of 1.5% to 8.4% in 50 – 65 years and 65 years or older groups, respectively. Admissions to hospital with heart failure have significantly doubled in the last 20 year. Th ese critical issu es in connection with growing demand for better quality of life have raised the public concern regarding the need for cardiovascular healthcare system s. On the other hand, online telemedicine systems are being considered useful and necessary for providing efficient healthcare services . The rapid growth in wireless and communication technolog ies has remarkably enhanced the scope of health monitoring systems . Vital sign monitoring systems are rapidly becoming the core of today’s healthcare deliveries. The paradigm has shifted from tradi tional and manual recording to computer based electronic records and further to handheld devices as versatile and innovative healthcare monitoring systems. In the present study, a real - time heart monitoring system has been developed for collect ing real - time cardia c data from remote patients, evaluation of m ultiple vital signs, long - term monitoring , and a wireless connection to the doctor at a super specialty hospital in an emergency . The main aim of this research is to facilitate the connection bet ween remote patient and doctor which might not be possible otherwise due to low doctor - to - patient ratio . The developed monitoring system can continuously acquire multiple vital signs and further relayed those to an intelligent data analysis scheme to diagn ose anomalies for diagnosing potential chronic diseases. Furthermore, this study analyzed the acquisition of ECG (electrocardiogram) signal from the wearable sensor by detecting noise from the electrocardiogram waveform. Acquired ECG signals are then compa red with available ECG data bases (MIT - BIH database). Arduino UNO has been used for visualizing and storing the signal in real - t ime. It stores the extracted signal from the sensor and then sends to doctor’s interface to calculate R - R peak detection and QRS detection. This ECG signal processing helps the medical experts to analyze the status of the patient at the initial stage. The developed system has a user friendly web interface for medical experts to observe immediate pulse signals for monitoring patients remotely. Once abnormal event happened in vital signs , data will immediately transmit to remote medical ser ver through the internet (Wi - Fi and 3G /4G ) . An alert mechanism has been also developed for each vital parameter to give warning messages to the patients in case his/ her heart rate is going critical. In Alert mechanism, threshold values for each vital parameter ha ve been designed after referrin g the medical research reports l ike European Society of Hypertension, World Congress of Cardiovascular Health and also discussed with medical experts inside Thammasat Hospital Thailand. When the measured data exceeds the threshold range, the system sends an alarm message to the doctor’s web server. The key contributions of the dissertation include: 1. The protocol of developed system has been successfully developed and implemented, which offered a high standard of healthcare with a major reduction in medical cost. More broadly, the system is conceptualized to provide an interphase between doctor and the patients for two way communications by considering cost, ease of application, accuracy and data security. 2. Android listening port has been created to receive and store medical information of th e patient which is then transmitted to the web interface using wireless communication. The w eb interface iv has been designed to be on the doctors’ side so as to inform them of the medical status along with the location of the patient in real - time . 3. An alarmin g system based on threshold values has also been designed which sends an alert message to the doctor in case of a nomalies such as arrhythmia, hypotension, hypertension, fever, and hypothermia. It has been found that the time taken for sending alarm from patient to doctor is 30 secs under Wi - Fi protocol and under 3G it is approximately 56 secs between sending and receiving the alarm . It has been found that the alarm sending time for both of thes e wireless protocols is under an acceptable range of medical standards (4 to 6 minutes as per American Heart Association ). 4. In order to evaluate and show the practical implementation, the developed system has been used to monitor sixty five (65) cardiac pat ients located far away from the designed web application. The data obtained through the developed system is found to be significantly acceptable when compared with the hospital’s conventional machines.5. This study provides a framework of Real - time low cost ECG monitoring system which performs the real - time ECG signal acquisition from patient’s body via wearable sensor. The acquired real - time data from the sensor is then transferred to windows based app lication which is installed on d octor’s computer. Furthe rmore, the developed ECG system provides the noise detection al gorithm based on Wavelet based t ransformations which includes the R - R detection, QRS and HR (heart rate) calculation. The proposed algorithm also provides the noise detection in acquired ECG si gnal from the wearable sensor. The performance of the proposed system has been evaluated against the MIT - BIH normal arrhythmia database . ECG signals from the electrodes are amplified with a gain of 300 and filtered with the cut - off frequencies of 0.5Hz in the high pass filter and 100Hz in the low pass filter . The ECG recordings are digitized at 360 samples per second per channel over a 10 mV range . It has been reported that the system is able to detect at an R - peak with a success rate of 97.8% and it is als o able to detect abnormal heartbeat condition with a success rate of 78.9%. The total delay in real - time ECG monitoring system is 94ms. On comparing wit h the existing algorithms, Pan - T ompkins algorithm causes around 148ms. So, developed ECG monitoring algo rithm given less time delay during signal analysis . 6. The p resent study demonstrates the applicability of remote care monitoring and it assures the 24/7 continuous screening to patients residing at distance from hospitals. The experimental validation proces s has been done for remote patients residing 10 - 100 km away from the hospital server. From this study , it can be stated that the proposed system is convenient, reliable, low cost, accurate, real time and ensures data security. The developed system would in form the doctor in case of emergency through alarms; however, delay in alarms might occur due to weak signals of 3G networks in some remote areas. Though the delayed alarming time is still within the golden period of time (time period is given by medical e xperts in World Heart Federation) yet improvement in alarming lead time should be considered in future research. As wireless technology is emerging day by day, the u se of latest wireless technologies may overcome these issues which ultimately increase the applicability and usefulness of the proposed remote monitoring system. Furthermore, the false alarms can be generated due to the battery issues of sensors and smartphone. The research can be extended to overcome these battery and false alarm limitatio ns. This study could be continued for the improvement of the ECG algorithm to detect further abnormalities in different types of heart diseases such as Hypertension and Heart stroke. |
| Year | 2016 |
| Corresponding Series Added Entry | Asian Institute of Technology. Dissertation ; no. RS-16-04 |
| Type | Dissertation |
| School | School of Engineering and Technology (SET) |
| Department | Department of Information and Communications Technologies (DICT) |
| Academic Program/FoS | Remote Sensing (RS) |
| Chairperson(s) | Nitin K. Tripathi |
| Examination Committee(s) | Sarawut Ninsawat ;Peeyush Soni ;Peerapong Kitipawong |
| Scholarship Donor(s) | Asian Institute of Technology Fellowship |
| Degree | Thesis (Ph. D.) - Asian Institute of Technology, 2016 |