Resource-Aware Mobile-Based Health Monitoring
Abstract:
Monitoring heart diseases often requires frequent measurements of Electrocardiogram (ECG) signals at different period of the day, and at different situations (e.g. traveling, and exercising). This can only be implemented using mobile devices in order to cope with mobility of patients under monitoring, thus supporting continuous monitoring practices. However, these devices are energy-aware, have limited computing resources (e.g. CPU speed and memory), and might lose network connectivity, which makes it very challenging to maintain a continuity of the monitoring episode. In this paper, we propose a mobile monitoring solution to cope with these challenges by compromising on the fly resources availability, battery level, and network intermittence. In order to solve this problem, first we divide the whole process into several subtasks such that each subtask can be executed sequentially either in the server or in the mobile or in parallel in both devices. Then we developed a mathematical model that considers all the constraints and finds a dynamic programming solution to obtain the best execution path (i.e., which substep should be done where). The solution guarantees an optimum execution time while considering device battery availability, execution and transmission time, and network availability. We conducted a series of experiments to evaluate our proposed approach using some key monitoring tasks starting from pre-processing, to classification and prediction. The results we have obtained proved that our approach gives the best (lowest) running time for any combination of factors including processing speed, input size and network bandwidth. Compared to several greedy but non optimal solutions, the execution time of our approach was at least 10 times faster and consumed 90% less energy.
Existing System:
The ever evolving need for independent healthcare management has shifted healthcare services from hospital centric to patient centric where patients health can be observed while they are home, accomplishing their normal daily activities, and free move between different locations. This transformation encounters varying challenges related to the limited resources in terms of processing power, battery, data size, and network bandwidth, making continuous monitoring very hard to maintain. Therefore, many research initiatives have been developed to address some of these challenges and proposed mobile and pervasive solutions to evaluate these influencing factors [1]–[4]. However, they fail to provide an optimum solution that adapts on the fly to the available resources (e.g. battery, network) and compromise between different execution alternatives, such as executing on the mobile or server or both. It is very hard to evaluate these dynamic, and unpredictable factors given the frequency of switching between various networks capabilities due to user’s mobility, the mobile devices capabilities, and also the nature of monitoring tasks (e.g. divisible task, parallel, or sequential) to be executed. A novel solution should evaluate these resources on the fly and decide whether to process monitoring tasks locally or remotely.
Proposed System:
We describe hereafter the key processes and the role of each component involved in the above architecture: Data Acquisition: we use non-invasive wearable sensors [39] to collect bio signals from a subject under observation. Signals are transmitted using Bluetooth to a mobile device. Smart Decision Support Module (SDSM): is deployed on the mobile device and implements a resource meter that measures a set of metrics to evaluate the mobile device’s resources availability (e.g. Memory, CPU, Battery). The SDSM uses the collected measures to decide where to execute each process: online (on the mobile device) or offline (e.g. delegate to a backend server). Data / signal processing and analytics: this module describes the set of processes involved in the signal processing and analytics. This includes preprocessing, feature extraction, and classification processes. These processes are executed in sequence and each one is used as input to the next process. The preprocessing consists of cleaning, removing noises, and filtering the signal in order to be used for feature extraction, selection, and classification.
Conclusion:
The advances in mobile device’s capabilities and the growth of the underlying network throughputs make Mobile-based health monitoring an attracting cost-effective solution for continuous monitoring of various diseases. However, using such solution reveals multitudes of challenges including the size of generated data, the resource scarcity of the mobile devices used to collect data and the capacity and availability of the network needed to transfer this data. In this paper, we proposed a mobile monitoring solution addressing these challenges and incorporating some smart features to encounter the energy insufficiency of mobile devices and network interruption. We developed a formal model that evaluated the best execution decision considering online, offline, or combined processing. The model used dynamic programming to determine the best execution path that guarantees an optimum execution time given the resources constraints mentioned above. We evaluated the applicability of our solution using ECG data set and we evaluated the key monitoring processes including pre-processing, feature extraction, and classification. Our approach outperformed other baseline approaches and provided the best running time for different computation configurations while considering key influencing factors: data size, processing speed and network bandwidth. We are planning to extend this work to cope with a large data set and a long term monitoring episodes and involving different diseases.
References:
[1] J. Chai, “The design of mobile ecg monitoring system,” in 4th ICEIEC, 2013, pp. 148–151.
[2] P. Wangyong, L. Bin, and H. Chen, “A system of portable ecg monitoring based on bluetooth mobile phone,” in ITME, 2011, pp. 309–312.
[3] U. Qidwai, J. Chaudhry, and M. Shakir, “Ubiquitous monitoring system for critical cardiac abnormalities,” in IEEE IECBES, 2012, pp. 872–877.
[4] A. M. Patel, P. K. Gakare, and A. N. Cheeran, “Real time ecg feature extraction and arrhythmia detection on a mobile platform,” Intl. J. of Comp. Apps, vol. 44, no. 23, 2012.
[5] E. Alzate and F. Martinez, “Ecg monitoring system based on arm9 and mobile phone technologies,” in ANDESCON conference, 2010, pp. 1–6.
[6] V. Bezerra, L. Leal, M. Lemos, C. Carvalho, J. Bringel Filho, and N. Agoulmine, “A pervasive energy-efficient ecg monitoring approach for detecting abnormal cardiac situations,” in IEEE Healthcom, 2013, pp. 340–345.
[7] S. Gradl, P. Kugler, C. Lohmuller, and B. Eskofier, “Realtime ecg monitoring and arrhythmia detection using androidbased mobile devices,” in IEEE conf. in Engg in Medicine and Biology, 2012.
[8] S. Pandey, W. Voorsluys, S. Niu, A. Khandoker, and R. Buyya, “An autonomic cloud environment for hosting ecg data analysis services,” J. of Future Generation Computer Systems, Elsevier, vol. 28, no. 1, pp. 147–154, 2012.
[9] C. Wen, M.-F. Yeh, K.-C. Chang, and R.-G. Lee, “Realtime ecg telemonitoring system design with mobile phone platform,” Els. Measurement, vol. 41, pp. 463–470, 2008.
[10] B.-S. Lin, W. Chou, H.-Y. Wang, Y.-J. Huang, and J.-S. Pan, “Development of novel non-contact electrodes for mobile electrocardiogram monitoring system,” IEEE J. of Transl. Engg in Hlth and Med., vol. 1, pp. 1–8, 2013, 10.