PTOTSLPOtherNeurorehabilitation and neural repair2023

Predicting Motor Outcomes Using Atlas-Based Voxel Features of Post-Stroke Neuroimaging: A Scoping Review.

Ji-Hun Yoo, Benjamin Chong, Peter Alan Barber and 2 others

PMID 37191349

WHAT IT FOUND

Atlas-based brain imaging methods for predicting stroke motor recovery are currently unfit for clinical practice.

The 17 studies reviewed used widely varying methods, small samples, and lacked validation, making results impossible to compare or apply to patient care.

Key findings

01The reviewed prediction models are not ready for clinical use due to methodological variability and lack of validation.

02Only one of the included studies performed out-of-sample validation, a key step for ensuring a model works on new patients.

03Reporting of imaging details was poor, with no study meeting all mandatory standards for describing how scans were taken.

STILL TO COME

How it was doneWhat they foundWhat it means for PTsWhat it means for OTsWhat it means for SLPs

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.

Already have one?

What it does not show

The search was limited to English-language studies and routine clinical imaging modalities, excluding advanced research techniques. Only one author screened the studies, increasing the risk of selection bias. The included studies were highly heterogeneous in methods and outcomes, preventing meta-analysis or direct comparison. Sample sizes in the reviewed studies were small (30 to 207 participants), limiting the generalizability of their findings.

Declared interests

The authors declared no potential conflicts of interest. The work was partially supported by the Health Research Council of New Zealand.

The easy way to misread this

Do not interpret the reported prediction accuracies (e.g., AUCs or R-squared values in the tables) as evidence that these tools are ready for use. The review explicitly states these methods are currently unfit for clinical practice due to a lack of validation and methodological inconsistency.

Read it on PubMed →