Use of a multilayer perceptron to create a prediction model for dressing independence in a small sample at a single facility.
Takaaki Fujita, Atsushi Sato, Akira Narita and 7 others
PMID 30774208WHAT IT FOUND
A neural-network model predicted which stroke patients would dress independently at discharge better than conventional models in 82 patients, but it was tested only at a single facility and not externally.
Key findings
01On the full dataset, the multilayer perceptron model had the highest AUROC, 0.937, and classification accuracy, 86.8%.
02In 10-fold cross-validation, multilayer perceptron had the highest AUROC, classification accuracy, specificity, positive-predictive value, and negative-predictive value, while logistic regression had the highest sensitivity.
STILL TO COME
How it was doneWhat they foundWhat it means for OTs
Read the rest of this summary
You get three full summaries a month, free, and we do not ask for a card. Search, the TL;DRs and your library stay unlimited either way.
What it does not show
Only 82 patients from a single facility were studied, so the results may not apply to other facilities. Validation used the same dataset with 10-fold cross-validation, not unseen patients. The authors state the models need further validity testing. Patients were included only if they could not dress independently at admission, so it does not address patients who already dressed independently. The study compared model accuracy only; it did not test whether using a model improves discharge planning or patient outcomes.
Declared interests
No conflicts of interest were declared. The work was supported by a grant from the Japanese Society for the Promotion of Science Grants-in-Aid for Scientific Research, grant number 18K17728.
The easy way to misread this
Do not assume this neural-network model can predict dressing independence for your patients. It was built and tested only on 82 patients from a single facility, and the authors state external validation with unknown data is needed.