Surveillance system and method for predicting patient falls using motion feature patterns

A method for detecting a fall risk condition, the method comprising:

• retrieving, from a storage device, video data that is associated with triggering false alarm cases of patient falls;
• determining motion feature patterns of the video data associated with the false alarm cases;
• storing the motion feature patterns of the video data;
• analyzing, using a classifier that is trained with a set of features associated with triggering alarms for patient falls, a plurality of frames generated from a surveillance camera;
• classifying one or more of the plurality of frames as causing an alarm;
• and
• verifying the alarm by extracting motion feature patterns of the one or more of the plurality of frames and comparing the extracted motion feature patterns with the stored motion feature patterns.

A system for detecting a fall risk condition, the system comprising:

• a computer system comprising memory and logic circuitry configured to:
• retrieve, from a storage device, video data that is associated with triggering false alarm cases of patient falls;
• determine motion feature patterns of the video data associated with the false alarm cases;
• and
• store the motion feature patterns of the video data in a database stored on a memory unit accessible to one or more classifier servers over a network, the one or more classifier servers verify one or more of a plurality of frames generated from a surveillance camera as causing an alarm based on the stored motion feature patterns.