AI Assessment and Prediction Models for Geriatric Trauma

Establishing AI Assessment and Prediction Models for Geriatric Trauma and Effects of AI Incorporated Into an Innovative Health Service Program in Geriatric Trauma Patients

To fill up the research gaps and build up a healthcare service model for geriatric trauma patients, this two-stage study aims to (a) develop geriatric trauma Artificial Intelligence (AI) prediction models by comparing the outcomes from clinical assessment and AI joint images; and (b) examine the long-term effects of an innovative health service program in geriatric trauma patients.

Background: About 40% trauma patients will be over age 65 years by 2050, thus geriatric trauma constitutes an increasingly recognized problem that should be concerned. However, little is known about complicated causes of injury and long-term outcomes in older injured patients. Previous study strategies to promote health outcomes for geriatric trauma still are inconclusive. Due to the limited healthcare resources and workforce, health service for geriatric trauma care should be refined in the new era. Expected outcomes/impacts: This study attempts to build up a modern healthcare model that may influence geriatric trauma care as well geriatric assessment. The AI prediction model can be used to quickly assess "degree of balance" for geriatric trauma patients. This study will involve a nursing student and healthcare professionals, and help them to learn how AI connects to geriatric trauma care. This would be an opportunity to cultivate potential experts for AI and healthcare. Last, the outcomes are highly associated with the government's policies related to smart health, smart healthcare, and long-term care workforce.

The program will be implemented into the experimental group, the program has been designed as a three-stage model including hospital management, transition care, and post-acute stage care.

The NPRS is a one-dimensional measure of pain intensity in adults. A NPRS has similar anchors at the extremes but offers numbers from 0 through 10. It is well known and broadly used scale in clinical settings. No further reliability and validity tests are needed.

The NPRS is a one-dimensional measure of pain intensity in adults. A NPRS has similar anchors at the extremes but offers numbers from 0 through 10. It is well known and broadly used scale in clinical settings. No further reliability and validity tests are needed.

The GDS-SF is a 15-item scale used to identify depression in older adults. It is from the original GDS developed by Sheikh and Yesavage to examine the self-perceived emotions of older adults. The version will be used in this study is a validated Chinese that has been tested its reliability in geriatric older population in Taiwan. The GDS-SF features binary scores for a total score of 0-15. A higher score indicates a more severe of depression. In the GDS-SF, a score of 0 to 5 is normal, a score > 5 suggests depression, and a score ≥ 10 is almost indicative of depression. The Cronbach's alpha for this study will be retested.

The WHOQoOL-BREF will be used to measure quality of life. Six popular generic instruments of HRQOL are commonly used as disability outcome measures, some of them were developed later were tested trans-culturally in extensive research programs such as the WHOQoL-BREF. The WHOQoL-BREF is a 26-item scale that is measure along with four subscales including physical, psychological, social relationship and environment domains. Scores range from 1 to 5 with higher scores indicating higher quality of life. The reliability and validity of the Chinese WHOQoL-BREF have been tested in patients with traumatic injuries, but not in geriatric trauma. Thus, the Cronbach's alpha for this study will be retested.

The BBS is a 14-item scale used to assess balance of the older adult in various clinical settings. Each item of the BBS is scored from 0 to 4, which are added to make a total score from 0 to 56; a higher score indicates better balance.

Although the BBS has been widely used and has good psychometric properties, the scale has its limitation in detecting detailed change of balance in older adults. Neuls et al. suggested that clinicians could use the BBS combined with other measures considering unique patient factors to quantify the chances of older adults. Therefore, the key points for 17 human parts and body joints matching the 14 items of the BBS will be recorded as images. Video captured by front and side cameras will be made by each item with 5 levels. Because of page limits, we present 7 items to demonstrate how those key points compared to BBS items captured by cameras.

Inclusion Criteria: admitted to the hospital due to injury first time able to communicate in Mandarin agree to participate in this study from hospital discharge to one month after hospital discharge patients and their primary caregivers have smart mobile phones and internet access available can move independently before trauma Exclusion Criteria: not frequent user for smart phones severe cognitive impairment victims of hangings and poisonings