Fall detection from a manual wheelchair: preliminary findings based on accelerometers using machine learning techniques

Automated fall detection devices for individuals who use wheelchairs to minimize the consequences of falls are lacking. This study aimed to develop and train a fall detection algorithm to differentiate falls from wheelchair mobility activities using machine learning techniques. Thirty, healthy, ambulatory, young adults simulated falls from a wheelchair and performed other wheelchair-related mobility activities in a laboratory. Neural Network classifiers were used to train the algorithm developed based on data retrieved from accelerometers mounted at the participant's wrist, chest, and head. Results indicate excellent accuracy to differentiate between falls and wheelchair mobility activities. The sensors mounted at the wrist, chest, and head presented with an accuracy of 100%, 96.9%, and 94.8%, respectively, using data from 258 falls and 220 wheelchair mobility activities. This pilot study indicates that a fall detection algorithm developed in a laboratory setting based on fall accelerometer patterns can accurately differentiate wheelchair-related falls and wheelchair mobility activities. This algorithm should be integrated into a wrist-worn devices and tested among individuals who use a wheelchair in the community.

SMARTBATHROOM 102: STUDYING BATHING TRANSFER PERFORMANCE

Age-related changes in functional ability, particularly among people aging with long-term disabilities, can impact safe performance of toilet and bathtub transfers and severely limit their ability to successfully age-in-place. The SmartBathroom is a state-of-the-art bathroom laboratory that features mechanically adjustable features and an array of sensors that can measure walking performance, location of feet and hands, and forces applied to the bathroom surfaces and fixtures. To complement the SmartToilet system, TechSAge recently developed the SmartBathing Transfer Testbed prototype that will enable us to study how different bathing environment configurations can impact transfer performance. The testbed consists of a height adjustable tub wall and three-wall, height and angle adjustable grab bars with integrated sensors to measure hand location and grip and load forces. This session will describe the SmartBathing Testbed and present results of studies with individuals both with and without ambulatory impairments performing simulated transfers within the testbed.

Insights on an automated fall detection device designed for older adult wheelchair and scooter users: A qualitative study

BACKGROUND

Falls are a concern for older adults who use wheelchairs and scooters. Many wheelchair and scooter users require assistance to recover from a fall and often lie on the ground waiting for assistance for 10 min or more. An automated fall detection device may facilitate communication with care partners and expedite recovery; however, there is limited research on the specifications and features of an automated fall detection device preferred by older adults who use wheelchair and scooter.

OBJECTIVE

To examine the desired specifications, perceived ease of use and perceived usefulness of an automated fall detection device desired by older adults who use a wheelchair or scooter through semi-structured interviews.

METHODS

Fifteen full-time wheelchair and scooter users (9 females; age: 68 ± 5 years) were interviewed from July to November 2020. Interviews were transcribed, coded, and analyzed.

RESULTS

Preferred features include wireless charging, a watch form, ability to change the individual who is contacted in the event of a fall, and the ability to disable a notification in the event of a false alarm. Participants felt that an automated fall detection device would be useful and easy to use.

CONCLUSIONS

Older adults who use a wheelchair or scooter indicated the need for an automated fall detection device to facilitate recovery from a fall. Participants reported challenges with previous fall detection devices and the need for specific design requirements to facilitate ongoing use. Participants' insights inform the design of a fall detection device to maximize usability and prevent technology abandonment.

Sensitivity of Apple Watch fall detection feature among wheelchair users

A reliable fall detection device is crucial to minimize long-term consequences of falls among wheelchair users. This study examines the sensitivity of Apple Watch to detect intentional falls from a wheelchair. Twenty-five able bodied (age: 21.9 ± 2.5 years) participated in a protocol in which they intentionally fell out of a wheelchair in a laboratory setting. Each participant wore an Apple Watch Series 5 and performed 3 falls in the forward, right and left sideways, and backward directions onto a crash pad totaling 12 falls each. The Apple Watch was manually checked after each fall to determine if the device registered a fall. From 300 fall trials captured, the Apple Watch detected 14 falls showing a sensitivity of 4.7%, a false negative rate of 95.3%. Logistic regression showed that participant's height, impact force, lower limb functioning, and fall direction are parameters that may influence the ability of the Apple Watch to detect falls from a wheelchair. The Apple Watch fall detection feature presented with a very poor sensitivity to detect intentional falls from a wheelchair among able bodied young adults. Due to the high incidence and consequences of falls, a reliable fall detection device specific for wheelchair users is warranted.

Laser treatment of diabetic macular edema: comparison of ETDRS-level treatment with threshold-level treatment by using high-contrast discriminant central visual field testing

Grid laser is recognized as an extremely effective treatment of diabetic macular edema, although it causes significant chorioretinal damage when applied and scars that expand with time. The purpose of this study is to compare the effects of two methods of grid laser photocoagulation for diabetic macular edema on high-contrast target discrimination in the central visual field. Grid laser photocoagulation with the Early Treatment Diabetic Retinopathy Study intensity burns has previously been shown to cause full retinal thickness burns. In this study, it produced severe destruction of paraxial vision, most marked at 2 degrees to 10 degrees from fixation. Grid laser using threshold-level burns, in contrast, appeared to produce some improvement in thresholded high-contrast vision at eccentricities from 2 degrees to 3 degrees outward, but failed to normalize visual parameters at these intercepts or at intercepts closer to fixation. Therefore, the recommendations are made (1) to use screening modalities other than biomicroscopic perception of retinal swelling to define earlier opportunities for intervention in the diabetic maculopathic process and (2) to use threshold or sub-threshold methods of laser grid photocoagulation for treating leakage and/or edema.