Keywords

1 Introduction

Knowledge Based Cloud Engine performs a significant role in the development of smart vehicles, which offers smart transportation, cloud connectivity, vehicle-to-vehicle interaction, smartphone integration, safety, security, and e-healthcare services. Recent development trends show that auto industries are already paying attention to develop IoT cars that could integrate driver’s health status and driving safety. Both auto industry and key global original equipment manufacturers are integrating healthcare services into their next-generation products [1].

As health has become the major point of interest, many researches are focused on the development of smart health care system. Driving consumes a significant amount of time in our daily life. Sudden health issues like cardiac arrest, stroke etc., can happen during driving. In order to avoid those circumstances automotive manufacturers as well as users are interested to incorporate the real-time health monitoring system in the car [2, 3]. Driver’s health abnormality may also effect safety of other vehicles. So, automotive manufacturers and users are interested to include real-time health monitoring in car system. World population is aging rapidly and aged population is getting much more concern nowadays. Aging originates from increasing longevity and results in deteriorating fertility [4]. Population aging is taking place in nearly all the countries of the world. As age increases, older drivers become more conservative on the road. Age-related decline in cognitive function hampers safety and quality of life for an elder. As the aged population in the developed world is increasing, so the number of older drivers is becoming higher [5]. Research on age-related driving has shown that an increased risk of being involved in a vehicle crash is more at around the driver’s age of 65. Certain behavioral factors, in particular, may contribute to these statistics: drifts within the traffic lane, confusion in making left-hand turns, and decreased ability to adjust behavior in response to an unexpected or fast-changing situation [6].

Stroke is the second top reason of death above the age of 60 years, and its proportion is rising. Many health abnormality happens after stroke. Postural disorders is observed as one of the most common disabilities after stroke. [7]

Some developments in the wearables and embedded sensors to measure physiological and bio-signals during driving have been already done [8]. Faurecia developed an automotive seat which detects traveler’s heart rate and breathing rhythm through unique types of embedded sensors [9]. IPPOCRTE designed a steering wheel could measure vital physiological parameters including ECG, eye gaze, body temperature, and pulse rate [10].

This paper focused on briefly explaining the design and framework of the elderly drivers’ health monitoring services in connected car using IoT devices. The purpose of this study is to develop the knowledge based real-time health monitoring system for drivers during the sudden stroke onset using body balance air cushion and IoT devices.

2 Model and Methodology

2.1 Intelligent Car Seat Model

A model of automotive seat is designed for monitoring health status of driver. For monitoring body postural balance, an air cushion is designed that consists of four air chamber. Four chambers can indicate body inclination to four sides; front, rear, right and left side. Air chambers are pumped by a small air compressor to maintain inflation in order to detect symmetric body pressure. Air cushion is made of polyvinyl chloride, very common kind of synthetic plastic polymer.

For measuring body pressure over air cushion, each air chamber is equipped with air pressure sensor. This air cushion is inserted to inside of air seat and covered with seat cover. For better comfortability, air cushion top surface is placed in same level of car seat flat surface (Fig. 1). To add an air cushion in the seat cushion, small modifications have been done to make room for the air cushion. In order to get effective pressure response, air cushion has been placed in middle position of car seat. Identical air pressure in all chamber of air cushion indicate driver’s body balance over car seat. Large variation of pressure in air cushion chambers represents tilting of driver’s position in one side. Brain lesions may cause a difficulty of postural control, and postural disorder is found to be one of the most common disabilities after stroke onset [11].

Fig. 1.
figure 1

Intelligent car seat model embedded with IoT sensors.

Full size image

In back part of the seat, ECG sensor is attached and remains in contact with wearable clothes. Wearable clothes are made of woven conductive fabric. In front of seat, proximity heartbeat sensor is placed in order to measure driver’s heart rate. Face tracking and eye tracking camera will monitor face pattern and eye movement from the front side of driver.

2.2 Components of Driver’s Health Monitoring System

IoT sensors such as ECG/EEG sensor will monitor heart spectrum, pressure sensors will monitor body pressure balance, and heart rate will be measured using proximity sensors. Air cushion cell pressure can detect drivers’ postural balance in order to detect brain stroke onset of elderly drivers. Arduino Mega ADK is used as an interfacing platform with air cushion pressure sensors and Arduino platform is capable to feed data to Car control system and cloud server also (Fig. 1).

BIOPAC ECG sensor will monitor heart spectrum. For measuring heart rate, TI Launchpad based proximity radar sensor has been used. As victim lose conscience after stroke onset, postural position of drivers becomes unstable. Front & rear, right & left side inclination of postural position can be happen during stroke onset. Body unbalance can be happen for other reasons such as doing additional activities during driving. So, only one sensory system is not sufficient to detect stroke during driving for elderly people. ECG, Heart rate and air cushion pressure data together can detect any abnormalities during driving.

3 Framework of Knowledge Based Health Monitoring System

Knowledge based self-learning engine plays main role in this health monitoring system (Fig. 2). IoT sensors measures physiological data and data is transmitted to Service DB (Database) through network gateway. Service DB takes care data management and feed data for data processing. In Health status measurement model, algorithm extracts useful features and patterns of physiological parameter corresponding to specific target group. Then gathered knowledge stored in Knowledge DB for future use. These knowledge extraction trains up self-learning engine.

Fig. 2.
figure 2

Knowledge extraction process of physiological data.

Full size image

Framework of knowledge based health monitoring is showed in Fig. 3. During real-time health service, real-time biodata is feed to self-learning engine and compared with reference data knowledge. Health status/Stroke Expert intelligence model predicts and detects health abnormality if any abnormal pattern found in real-time data. Health Status alert service receives health updates and always ready to respond if any health risk, more specifically stroke happens. System will also generate an alert and deliver messages to emergency services, family of the victim, people around the victim, and hospitals in order to ensure the immediate medical assistance. Each sensor prediction result contributes in large set of IoT sensor network. The more sensor in health monitoring system, the more reliability of monitoring system.

Fig. 3.
figure 3

Framework of knowledge based health monitoring system.

Full size image

4 Conclusion

Knowledge based health monitoring system is expected to be promising service for real-time health monitoring. This study introduces an intelligent sensor based car seat model and also provides a basic framework Real-time knowledge based health monitoring; such as stroke detection system using body balance air cushion and IoT sensors for elderly drivers. Body balance system using air cushion, we can expect that the developed car seat is expected to identify stroke in tilted unbalanced postural position. In addition, ECG/EEG, heart rate sensor embedded in car seat can provide an intelligent in-car health monitoring platform and detect abnormality when health risk; such as brain stroke onset happens. Analysis of Stroke monitoring data has not been presented here. In future, health monitoring data analysis will be presented and study would consider a range of bio-sensors and technique in order to improve reliability of system for real-time health risk prediction during driving.