Software that keeps an eye on Grandma

Networked sensors and machine learning make it easy to see when things are out of the ordinary.

Much of health care — particularly for the elderly — is about detecting change, and, as the mobile health movement would have it, computers are very good at that. Given enough sensors, software can model an individual’s behavior patterns and then figure out when things are out of the ordinary — when gait slows, posture stoops or bedtime moves earlier.

Technology already exists that lets users set parameters for households they’re monitoring. Systems are available that send an alert if someone leaves the house in the middle of the night or sleeps past a preset time. Those systems involve context-specific hardware (i.e., a bed-pressure sensor) and conscientious modeling (you have to know what time your grandmother usually wakes up).

The next step would be a generic system. One that, following simple setup, would learn the habits of the people it monitors and then detect the sorts of problems that beset elderly people living alone — falls, disorientation, and so forth — as well as more subtle changes in behavior that could signal other health problems.

A group of researchers from Austria and Turkey has developed just such a system, which they presented at the IEEE’s Industrial Electronics Society meeting in Montreal in October.*

Activity as surmised in different rooms by the researchers' machine-learning algorithms
Activity as surmised in different rooms by the researchers’ machine-learning algorithms. Source: “Activity Recognition Using a Hierarchical Model.”

In their approach, the researchers train a machine-learning algorithm with several days of routine household activity using door and motion sensors distributed through the living space. The sensors aren’t associated with any particular room at the outset: their software algorithmically determines the relative positions of the sensors, then classifies the rooms that they’re in based on activity patterns over the course of the day.

From there, it’s easy to train software with habits — when bedtime typically occurs, how long an occupant usually spends in the kitchen — though these are handled generically (you don’t need to label the bedroom as the bedroom in order for the algorithm to detect that something is amiss when the occupant spends too long there).

The result is somewhat more subtle in its understanding of how a household works and when something might be out of order: if movement in the bedroom between 7 and 8 A.M. is usually followed by the opening of the bedroom door, then the same movement pattern without the door opening might suggest that someone has fallen while getting out of bed.

The researchers found that, compared to activity manually labeled by test users, their system was accurate at 81% to 87% depending on the type of algorithm used (SVM, CVS, or Hierarchical).

Networks of devices can bring intelligence out of individual machines and into centralized software that can understand an environment in its totality. That’s a central part of the philosophy of the industrial Internet, in which networked machines feed data into sophisticated software that can solve complex optimization problems that take large systems into account.

Dietmar Bruckner, a professor at Vienna University of Technology and an author of the paper, says his software (known by the tortured acronym ATTEND — AdapTive scenario recogniTion for Emergency and Need Detection) is tailored to the home-monitoring case outlined in his paper, but it could eventually be generalized to other types of building-monitoring applications.

Asked about bringing the technology to market, Bruckner said his research was being discontinued under funding cutbacks at his university. That’s unfortunate given the technology industry’s interest in using machine intelligence to deliver better health care. Might this be an opportunity for a startup to pick up where Bruckner et al. leave off?

*Available for a fee from IEEE: C. Tirkaz, D. Bruckner, G. Yin, J. Haase, “Activity Recognition Using a Hierarchical Model,” Proceedings of the 38th Annual Conference of the IEEE Industrial Electronics Society, pp. 2802-2808, 2012.

This is a post in our industrial Internet series, an ongoing exploration of big machines and big data. The series is produced as part of a collaboration between O’Reilly and GE.

tags: , , ,