Hospital-acquired bedsores avoidable with AI

AI has shown strong potential for predicting which recently hospitalized patients will develop pressure injuries (PIs), also known as pressure ulcers or bedsores, if they aren’t treated early with preventive medicine.

The capability was developed at the University of Michigan’s school of nursing and is detailed in a study published Aug. 30 in BMC Medical Informatics and Decision Making.

Co-senior authors Ivo Dinov, PhD, and Dana Tschannen, PhD, RN, and colleagues began with established and validated preprocessing steps for detecting various PI biomarkers.

To bring AI into their prediction model, they used EHR data from more than 23,000 inpatients to train machine learning classifiers on numerous PI features.

They based the training on “large, incongruent, incomplete, heterogeneous and time-varying data[sets] of hospitalized patients” who’d had any of numerous surgeries.

To forecast PI outcomes and stratify PI risk factors by predictive power, the team used both model-based statistical methods and model-free AI strategies.

The model-free techniques proved the better of the two, and a Random Forest classification algorithm consistently came through with the best PI forecasts, the authors report.

“AI techniques show promise to automatically identify patients at risk for hospital acquired PIs in different surgical services,” the authors comment. “Our PI prediction models provide a first generation of AI guidance to prescreen patients at risk for developing PIs.”

More:

Accurate prediction of PI is critical to assure that patients with risk are receiving the nursing care needed to prevent PI development. To date, our understanding of risk has been limited due to limitations in sampling (e.g., one surgical service) and/or methodology (e.g., failure to include all factors predictive of risk). This study is one of the first to use AI techniques with a large, general sample of surgical patients. Findings from this study have identified risk profiles for various surgical services that must be considered when determining prevention intervention strategies to employ.”

The study is posted in full for free.

Around the web

The Palo Alto giant used exams from nearly 250,000 patients to upgrade its already robust algorithm.

Exams performed using the deep learning-based reconstruction tool also maintained high image quality, experts reported Wednesday.

Stratifying exams according to risk can reduce unnecessary imaging and downstream costs of care, Hawaiian researchers reported in Radiology.

Trimed Popup
Trimed Popup