The current state of virtual healthcare

The burgeoning number of elderly people presents societies with both economic and social challenges.

During the past decade technologies have been developed to improve the quality of life for the elderly and to extend their ability to live independently.  The Active and Assisted Living (AAL) systems industry is booming. inlisol is part of this growing trend and aims to accelerate the development of the whole AAL system.

Active Assisted Living (AAL) systems

AAL systems have been developed to create better living conditions for older and disabled people, and to support caregivers and medical staff with behavioural and alerting information that can be useful for prevention or timely intervention.

AAL systems are being used to monitor simple daily habits, measuring, for example, movement or rest periods, verifying food and medicine intake, etc, and also to highlight social components that effect psychological well-being.  Many aspects of the daily activities of elderly people can be automatically checked with AAL systems, which reduces the costs and increases the monitoring capabilities for healthcare professionals.

Design of AAL systems

When designing an AAL system, the first step should be sensors selection for data acquisition and functionalities.

The resulting data must be collected and evaluated to extract meaningful information, such as recognising changes in habits.  Heterogeneous sensors are often used, and the data must be collected over a long period of time in order to learn the “normal” behaviour of each individual.

When choosing which sensors to use the most important aspect to consider is the acceptability of the sensors to the people being monitored.

The design of AAL systems must consider the needs of the people being monitored.  A human-centred approach, where the end users are involved and participate in all stages of the design process is best practice.

Detecting anomalies

To detect anomalies and abnormal changes in behaviour, first, a systematic model from long-term observations of an individual’s normal activities must be built. Then, the normal pattern is compared with new observations and the deviation is estimated.

The concept of normal is not general and it is not the same for different subjects. It is closely related to each individual person, so it must be learned from the prolonged observation of each person.  Recent technological developments have made it possible to collect and store a huge amount of data on people’s habits.

Processing constraints

Many AAL systems have been devised to provide alarms when dangerous situations are detected. In these cases, real-time processing of data is necessary to provide prompt interventions, but, at the same time, robust processing techniques are necessary to avoid false alarms.

When detecting changes in habits, long periods of observation are required. In this case, AAL systems must collect data regarding all aspects of a person’s daily life, and the results can only be evaluated when models of normal behaviours have been built. In these cases, offline processing can be done on a large sample of data collected in a cloud computing framework. 


Over the next decade, there is likely to be a considerable spread of technologies in people’s homes, and huge sets of rea-time data will be generated, the challenge is to provide these systems with intelligence. Companies like inlisol are focusing on developing and incorporating intelligence through machine learning. 

The future of AAL

In recent years the development of AAL systems has been receiving a lot of attention from the scientific community for several reasons:

  • they can reduce the costs of daily life assistance of elderly people; 
  • they can monitor the physical and psychological wellbeing of people who live alone; 
  • and, they can detect behavioural changes that could be a sign of early-stage neuro-degenerative diseases.