Browse Prior Art Database

Behaviour Analytics for People with Mental Illnesses Disclosure Number: IPCOM000235471D
Publication Date: 2014-Mar-03
Document File: 2 page(s) / 37K

Publishing Venue

The Prior Art Database


There is a range of mental illnesses (schizophrenia, bi-polar disorder, depression) which can result in a person trying to do harm to themselves. Often, when such tendencies are noticed by a relative or a doctor - the person suffering from the disorder can be sectioned by a court. Whilst being cared for in a secure unit the person is constantly monitored and proper medication is being administered to them to ensure their well-being and safety. When they are brought back into a state when the medical personnel deems that they are no longer in danger of hurting themselves - the person is discharged. However, often the state of the discharged might start to deteriorate without medical supervision. This device allows to monitor a patient in a non-intrusive way while they are in a secure unit, gather data on their normal behaviour and formulate a baseline behaviour consisting of behavioural patterns that should be considered as normal for this person. Then, after the patient is discharged the non-intrusive monitoring continues in their house or apartment. The current behaviour is constantly compared to the baseline. The monitoring algorithms look for specific manifestations of illnesses that are characteristic to the person being monitored. Adjustments are being made to the baseline to account for the former patient being home, rather than in a hospital. When the monitor senses a dangerous change in the person's behaviour - an emergency notification is sent to the designated person and/or the patient's treating doctor.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 51% of the total text.

Page 01 of 2

Behaviour Analytics for People with Mental Illnesses

When a person with a mental illness gets sectioned, they are placed in an environment with sensors similar to Microsoft Kinect. Those sensors are able to distinguish between different people. As soon as the person arrives in the hospital - the sensors start recording their behavioural patterns. The patterns recorded in the beginning - are more representative of the behaviour that is typical for the person

when they are out of balance. When the medical personnel is confident that the patient has gotten into a stable safe state - the sensors are switched into the mode,

where they store the newly recorded behavioural patterns as baseline normal. Statistical analysis of pattern occurrence frequencies will be calculated and added to the baseline behaviour. For example: the person typically eats twice a day, once in the morning in the time period between 9 and 11; once in the evening in the time period between 17 and 20.

    Additional patterns may be added to the baseline record that the person is not demonstrating. For example, it is normal for a human being to spend 6 to 9 hours sleeping. Provided the person doesn't work on night-shifts - it is normal for them to be awake during the day, and sleep at night. It is also normal for them to eat at least once a day, attend a toilet, take a shower, etc.

    As well as that the sensors and analytics will also be able to pick out less obvious aspects of the patient's behaviour, like twitching, for example, which, based on the situation, can be a side-effect from medication and will be added to the baseline behaviour, or be caused by the patient not taking medication and will be considered a deviation from the norm.

The above examples represent a small fraction of types of behavioural analysis that

will be performed....