The world’s population is living longer, but that doesn’t mean it’s any healthier.
Millions of people are plagued by chronic health conditions such as respiratory illness, arthritis, and back pain. Medical advances can mitigate these conditions, but unless people change their diet and exercise habits, things won’t change.
While poor health can motivate people to change their ways and exercise more, failing to exercise correctly can lead to even more health problems.
But there is good news for those who want a healthier lifestyle.
Researchers have developed one of a new generation of health recommender systems called the Context-Aware Recommender System (CARS) that offers advice and recommended fitness routes using crowdsourcing and smart city sensors.
“Healthcare models are steadily shifting toward patient-centric approaches in which patients are not only passive elements but also proactive contributors to their own and their peers’ health. From a much wider perspective, the sharing of information has led to recommender systems, which have evolved and integrated transparently into our daily lives. Such systems exploit collaboration among users and help them make better decisions,” write Fran Casino, Constantinos Patsakis, Edgar Batista, Frederic Borràs, and Antoni Martínez-Ballesté, co-authors of “Healthy Routes in the Smart City: A Context-Aware Mobile Recommender,” which appears in the November/December 2017 issue of IEEE Software.
So, what makes the CARS recommender system unique?
Smart cities have sensors and communication infrastructures that provide huge amounts of data. People can access and contribute to the flow of data, getting advice from seasoned runners and offering help and information to others. Because the information can be uploaded in real time, users get local and timely data that would otherwise be unavailable.
“Our system uses a memory-based method that employs neighborhood search (to determine groups of similar users) and recommendation or prediction computation, using various techniques,” say the authors.
The CARS system offers personalized recommendations to people according to their preferences, medical condition, and real-time information from the smart city.
For example, someone with respiratory problems would want to go where the air quality is better. In response, CARS can provide information about air pollution and possible allergens.
“Users can interact with CARS through the SmartRoute mobile app (available through Google Play). The app uses real weather data and air quality measurements—for example, ozone, nitrogen dioxide, or particulate matter,” say the authors.
In addition, users would want to know if dangerous conditions exist on their chosen route. To help, other users can enter data about traffic accidents, bad weather, landslides, and other dangers.
“When a route session starts, SmartRoute shows information such as distance, approximate duration, speed, and possible alerts or changes in the route in real time. In addition, users can send notifications to warn others of issues with that route,” the authors say.
After running real and simulated experiments, the researchers found that CARS provided recommendations with more than 80% accuracy. They plan to conduct more experiments and collaborate further with users and health experts to refine the system.
“The wide deployment of smart sensors in cities will pave the way for the success of tools like SmartRoute and will help promote active lifestyles without compromising people’s health,” they say.
Research related to fitness in the Computer Society Digital Library:
- Superhuman Sports: Applying Human Augmentation to Physical Exercise
- Sonic trampoline: the effect of audio feedback on the user experience during an exercise of jumping
- On-Line Analysis of Exercise Electrocardiograms
- Cycling through Cyberspace
- Topic-Aware Physical Activity Propagation in a Health Social Network
- Fitness Applications for Home-Based Training
- Can Fitness Trackers Help Diabetic and Obese Users Make and Sustain Lifestyle Changes?