Authors (including presenting author) :
LI YHS(1), YIP KW(1), LUI HK(1) LEUNG HW(1), CHAN CF(1)
Affiliation :
(1)Department of Old Age Psychiatry, Castle Peak Hospital
Introduction :
Fall risk is one of the major risk areas in psychogeriatric ward, as the elderly patients are predisposed to multiple fall risk factors including physical deterioration, cognitive impairment and medication side effects. Following a fall incident, elderly patient is more likely to suffer from serious physical and psychological complications, leading to higher level of dependence and longer duration of hospitalization. Healthcare professionals often encounter significant challenges in preventing fall incidents.
In response to these issues, the Department of Old Age Psychiatry collaborates with the Hospital Authority Head Office Innovation Laboratory to apply the first AI Fall Detection Visual Sensor in psychogeriatric in-patient setting to enhance fall prevention strategies.
Objectives :
To reduce the occurrence of fall incident in psychogeriatric ward by using AI Fall Detection Visual Sensor
To raise awareness of staff towards fall prevention.
Methodology :
Selected patients in a male psychogeriatric admission ward were monitored by the AI Fall Detection Visual Sensor. Selection criteria were: (1) high fall risk; (2) able to walk with assistance / aid. Outcome was evaluated by: (1) numbers of fall incident; (2) staff feedback.
Result & Outcome :
The deployment of the AI Fall Detection Visual Sensor has demonstrated remarkable outcome in terms of the reduction of fall incidents. Fall incidents dropped significantly from 6 to 10 per year in 2020 and 2021, to 0 to 3 per year from 2022 to 2024. Notably, no fall incident occurred in the ward area monitored by the AI Fall Detection Visual Sensor.
Positive feedback was received from the frontline staff, especially those working night shift. Timely alarm triggered by the AI Fall Detection Visual Sensor helped to alert staff to attend to patients with high fall risk promptly when they tried to leave the bed, thus preventing the potential occurrence of fall.
The AI Fall Detection Visual Sensor is capable of detecting patient movement that might lead to fall. By triggering early alarm to staff when patient with high fall risk attempts to leave the bed, staff could attend and offer help to the patient immediately. Nevertheless, further enhancement is required to decrease frequency of false alarm and the response time of the system (i.e. the time gap between detection of potential fall and the sounding of alarm). Furthermore, addition of automatic calibration function would enhance user-experience by reducing the time needed for system setup.
This initiative highlights the potential of AI technologies in reducing fall incidents and promoting safety culture in healthcare settings.
