Anatomy of Inpatient Falls: Examining Fall Events Captured by Depth-Sensor Technology

Potter, Patricia; Allen, Kelly; Costantinou, Eileen; Klinkenberg, Dean; Malen, Jill; Norris, Traci; O’Connor, Elizabeth; Roney, Wilhelmina; Tymkew, Heidi Hahn

doi:10.1016/s1553-7250(16)42029-5

Cited by 9 publications

(6 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In a study that reviewed videos of actual falls, 16 falls were analyzed that occurred over a 6-month period in 2 hospital wards. That study found that the median time from the moment a patient starts to move to get out of bed to the time of the fall was 1 minute 46 seconds, ranging from 22 seconds to almost 15 minutes 9. This would mean that the median response time of 24 seconds that we found in the current study should be adequate to prevent the vast majority of falls.…”

Section: Discussionmentioning

confidence: 50%

“…That study found that the median time from the moment a patient starts to move to get out of bed to the time of the fall was 1 minute 46 seconds, ranging from 22 seconds to almost 15 minutes. 9 This would mean that the median response time of 24 seconds that we found in the current study should be adequate to prevent the vast majority of falls. This is confirmed by our result of 0 falls among patients wearing the Smart Socks during the 13-month study period.…”

Section: Discussionmentioning

confidence: 80%

“…Fall prevention measures usually focus on patient education, increasing nurse awareness, or preventive measures such as installing bed and chair pressure sensors. In hospitals, bed and chair pressure sensors are very common because most falls occur when patients try to get out of bed, often for toileting 9,10. In a study of 59 acute care units in 57 US hospitals, it was found that alarms were in use for 36% of the patients 11.…”

mentioning

confidence: 99%

See 2 more Smart Citations

Fall Prevention With the Smart Socks System Reduces Hospital Fall Rates

Moore

Kline

Palettas

et al. 2022

Journal of Nursing Care Quality

View full text Add to dashboard Cite

Background: Falls of inpatients are common in hospitals. Existing fall prevention measures do not work consistently. Purpose: To determine whether Smart Socks reduce fall rates in fall risk patients at a major academic health center's neurological and neurosurgical based units. Methods: A prospective study was conducted that provided fall risk patients with Smart Socks and no other fall prevention system. Data collected included duration of Smart Socks wearing, number of alarms, response times, and patient-days. Results: A total of 569 fall risk patients were included for 2211.6 patient-days. There were 4999 Smart Socks alarms, but none of the patients fell. We observed a lower fall rate, of 0 per 1000 patient-days, for patients wearing Smart Socks than the historical fall rate of 4 per 1000 patient-days. The median nurse response time was 24 seconds. Conclusions: The Smart Socks reduced fall rates of fall risk patients included in the study.

show abstract

Section: Discussionmentioning

confidence: 50%

Section: Discussionmentioning

confidence: 80%

mentioning

confidence: 99%

See 1 more Smart Citation

Fall Prevention With the Smart Socks System Reduces Hospital Fall Rates

Moore

Kline

Palettas

et al. 2022

Journal of Nursing Care Quality

View full text Add to dashboard Cite

show abstract

“…Falls were most commonly attributed to an imbalance or incorrect shift of body weight. Slips and trips occurred rarely in this sample, but have been reported in 1/3 of hospital in-patient falls (Potter et al, 2016). Because all apartment areas in view of the depth camera were carpeted, slips on slick flooring would not be possible.…”

Section: Discussionmentioning

confidence: 65%

Assessment of Fall Characteristics From Depth Sensor Videos

O'Connor¹,

Phillips²,

Folarinde³

et al. 2017

J Gerontol Nurs

View full text Add to dashboard Cite

Falls are a major source of death and disability in older adults; little data, however, are available about the etiology of falls in community-dwelling elders. Sensor systems installed in independent and assisted living residences of 105 older adults participating in an ongoing technology study were programmed to record live videos of probable fall events. Sixty-four fall video segments from 19 individuals were viewed and rated using the Falls Video Assessment Questionnaire. Raters identified that 56% (n=36) of falls were due to an incorrect shift of body weight and 27% (n=17) from losing support of an external object, such as an unlocked wheelchair or rolling walker. In 60% of falls, mobility aids were present in the room or in use at the time of the fall. Use of environmentally-embedded sensors provides a mechanism for real-time fall detection and, ultimately, may supply information to clinicians for fall prevention interventions.

show abstract

“…In [5], Y. Wang et al proposed a WiFall system, which utilized the channel status information (CSI) in Wi-Fi signals as an indicator for different motions, and then used SVMs to classify them into resting, falling, sitting down, walking and standing up. In [6], doctors presented experimental finds to show RGB-D sensor's potential benefits for fall detection. By using a commercially available RGB-D camera, the Microsoft Kinect v2, A. Amini et al in [7] implemented both heuristic based and machine learning based algorithms to detect fall.…”

Section: Introductionmentioning

confidence: 99%

Multiple Patients Behavior Detection in Real-time using mmWave Radar and Deep CNNs

Jin

Zhang

Sengupta

et al. 2019

2019 IEEE Radar Conference (RadarConf)

View full text Add to dashboard Cite

To address potential gaps noted in patient monitoring in the hospital, a novel patient behavior detection system using mmWave radar and deep convolution neural network (CNN), which supports the simultaneous recognition of multiple patients' behaviors in real-time, is proposed. In this study, we use an mmWave radar to track multiple patients and detect the scattering point cloud of each one. For each patient, the Doppler pattern of the point cloud over a time period is collected as the behavior signature. A three-layer CNN model is created to classify the behavior for each patient. The tracking and point clouds detection algorithm was also implemented on an mmWave radar hardware platform with an embedded graphics processing unit (GPU) board to collect Doppler pattern and run the CNN model. A training dataset of six types of behavior were collected, over a long duration, to train the model using Adam optimizer with an objective to minimize cross-entropy loss function. Lastly, the system was tested for real-time operation and obtained a very good inference accuracy when predicting each patient's behavior in a two-patient scenario.

show abstract

Anatomy of Inpatient Falls: Examining Fall Events Captured by Depth-Sensor Technology

Cited by 9 publications

References 12 publications

Fall Prevention With the Smart Socks System Reduces Hospital Fall Rates

Fall Prevention With the Smart Socks System Reduces Hospital Fall Rates

Assessment of Fall Characteristics From Depth Sensor Videos

Multiple Patients Behavior Detection in Real-time using mmWave Radar and Deep CNNs

Contact Info

Product

Resources

About