Clinical Research Directory

Detecting Delayed Discharge in Acute Geriatric Unit Using Natural Language Processing

Delayed discharge in geriatric units is a health and economic issue. There is no algorithm to automatically measure the appropriateness of admissions or hospital days. 30% of the days of hospitalization in acute geriatric units (AGU) are not appropriate. Waiting for a transfer to a follow-up care and rehabilitation unit (SSR) is the main risk factor for inappropriate days. The purpose of this project is to develop an algorithm using natural language processing to predict the appropriateness of an admission to UGA, or a day at UGA.

Postural Control in Adults

The investigators propose a 2-week intervention involving adults over 65 years old. Participants will undergo dynamic balance training on an movable platform (stabilometer) with either restricted or full vision. The primary objective is to facilitate sensory re-weighting, hypothesizing that restricted vision training will increase somatosensory input utilization for postural control. Pre- and post-training assessments will measure postural control during gait initiation under a dual-task condition. The investigators hypothesize that the restricted vision group will demonstrate: 1. Improved postural control, indicated by increased center of pressure displacement during gait initiation. 2. Improved ability to utilize somatosensory input during a balance task.