This Article Full Text Full Text (PDF) All Versions of this Article: 40/9/3006    most recent STROKEAHA.109.552216v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Google Scholar Articles by Jonsdottir, K. Y. Articles by Mouridsen, K. PubMed PubMed Citation Articles by Jonsdottir, K. Y. Articles by Mouridsen, K. Pubmed/NCBI databases Medline Plus Health Information MRI Scans Stroke Related Collections Acute Cerebral Infarction Computerized tomography and Magnetic Resonance Imaging

(Stroke. 2009;40:3006.)
© 2009 American Heart Association, Inc.

Original Contributions

Predicting Tissue Outcome From Acute Stroke Magnetic Resonance Imaging

Improving Model Performance by Optimal Sampling of Training Data

Kristjana Yr Jonsdottir, PhD, MSc; Leif Østergaard, DMSc, PhD, MD, MSc Kim Mouridsen, PhD, MSc

From the Center of Functionally Integrative Neuroscience (CFIN), Department of Neuroradiology, Århus University Hospital, Denmark.

Correspondence to Kristjana Yr Jonsdottir, PhD, MSc, Department of Neuroradiology, Center of Functionally Integrative Neuroscience, Århus University Hospital, Nørrebrogade 44, 8000 Århus C, Denmark. E-mail krissa{at}pet.auh.dk

Background and Purpose— It has been hypothesized that algorithms predicting the final outcome in acute ischemic stroke may provide future tools for identifying salvageable tissue and hence guide individualized therapy. We developed means of quantifying predictive model performance to identify model training strategies that optimize performance and reduce bias in predicted lesion volumes.

Methods— We optimized predictive performance based on the area under the receiver operating curve for logistic regression and used simulated data to illustrate the effect of an unbalanced (unequal number of infarcting and surviving voxels) training set on predicted infarct risk. We then tested the performance and optimality of models based on perfusion-weighted, diffusion-weighted, and structural MRI modalities by changing the proportion of mismatch voxels in balanced training material.

Results— Predictive performance (area under the receiver operating curve) based on all brain voxels is excessively optimistic and lacks sensitivity in performance in mismatch tissue. The ratio of infarcting and noninfarcting voxels used for training predictive algorithms significantly biases tissue infarct risk estimates. Optimal training strategy is obtained using a balanced training set. We show that 60% of noninfarcted voxels consists of mismatch voxels in an optimal balanced training set for the patient data presented.

Conclusions— An equal number of infarcting and noninfarcting voxels should be used when training predictive models. The choice of test and training sets critically affects predictive model performance and should be closely evaluated before comparisons across patient cohorts.

Key Words: acute stroke • diagnostic imaging • magnetic resonance imaging • tissue outcome prediction