Feasibility and Diagnostic Accuracy of Ischemic Stroke Territory Recognition Based on Two-Dimensional Projections of Three-Dimensional Diffusion MRI Data

No fucking idea what use could be made of this for your stroke rehab. You better hope your doctor is a genius and can explain this to you at a 10th grade level.
http://journal.frontiersin.org/article/10.3389/fneur.2015.00239/full?

Jana Katharina Wrosch¹*, Bastian Volbers^1,2, Philipp Gölitz³, Daniel Frederic Gilbert⁴, Stefan Schwab², Arnd Dörfler³, Johannes Kornhuber¹ and Teja Wolfgang Groemer^1,5

• ¹Department of Psychiatry and Psychotherapy, Friedrich-Alexander University of Erlangen-Nuremberg, Erlangen, Germany
• ²Department of Neurology, Friedrich-Alexander University of Erlangen-Nuremberg, Erlangen, Germany
• ³Department of Neuroradiology, Friedrich-Alexander University of Erlangen-Nuremberg, Erlangen, Germany
• ⁴Institute of Medical Biotechnology, Friedrich-Alexander University of Erlangen-Nuremberg, Erlangen, Germany
• ⁵Psychiatric and Neurological Ambulatory Care Office, Bamberg, Germany

This study was conducted to assess the feasibility and diagnostic accuracy of brain artery territory recognition based on geoprojected two-dimensional maps of diffusion MRI data in stroke patients. In this retrospective study, multiplanar diffusion MRI data of ischemic stroke patients was used to create a two-dimensional map of the entire brain. To guarantee correct representation of the stroke, a computer-aided brain artery territory diagnosis was developed and tested for its diagnostic accuracy. The test recognized the stroke-affected brain artery territory based on the position of the stroke in the map. The performance of the test was evaluated by comparing it to the reference standard of each patient’s diagnosed stroke territory on record. This study was designed and conducted according to Standards for Reporting of Diagnostic Accuracy (STARD). The statistical analysis included diagnostic accuracy parameters, cross-validation, and Youden Index optimization. After cross-validation on a cohort of 91 patients, the sensitivity of this territory diagnosis was 81% with a specificity of 87%. With this, the projection of strokes onto a two-dimensional map is accurate for representing the affected stroke territory and can be used to provide a static and printable overview of the diffusion MRI data. The projected map is compatible with other two-dimensional data such as EEG and will serve as a useful visualization tool.

Introduction

Worldwide, ischemic stroke is the third-most common cause of years of live lost (1). In developed countries, stroke even ranks second (1). Routine patients presenting with stroke symptoms are typically evaluated by assessing diffusion MRI data, which is the most accurate method of illustrating the affected brain area in the acute phase of a stroke (2). In diffusion MRI, the degree to which water molecules can diffuse freely in the tissue is measured. In brain regions affected by stroke, the diffusion of water is impaired and can be detected as a reduced darkening of the MRI image over a recording time of (typically) 20 ms per layer (3). To evaluate the results, the two-dimensional images are stacked to form a three-dimensional dataset. Through this method of representing the imaging data, an observer can only assess the full information through visual inspection and steady rotation or direct manipulation of the images (e.g., scrolling through the layers). This approach is not always possible and limits the applicability of the method in printed reports.

Computer-aided diagnosis of stroke has gained in importance over the last decade. There is a multitude of approaches (4–7). Most groups use CT images to make early developments of stroke visible and detectable (8–10). Tyan et al. (11) considered diffusion MRI data as well as CT images for unsupervised computer-aided diagnosis of stroke-affected brain volume. The focus of all these approaches lies in enhancing the imaging data so that search algorithms can detect stroke symptoms. With this work, we aim to add to the visualization possibilities, which are not addressed by most computer-aided diagnosis approaches. The dimensionality reduction to a two-dimensional depiction of the entire brain provides advantages in print, in compatibility with brain surface measurements such as EEG and in tracking changes of the stroke lesion over time. We here present a new approach of converting the three-dimensional imaging data set to a two-dimensional map by geoprojection. To ensure the correct representation of strokes in the two-dimensional map, we developed a computer-aided artery territory recognition algorithm, using the projected maps as input.

The projection of three-dimensional data into a two-dimensional map via Mollweide geoprojection is a common method in other fields (12–14). We applied this mathematical method to diffusion MRI data to create a static, two-dimensional representation of the entire brain providing accurate information on stroke position and size.

The purpose of this retrospective study was to develop an intuitive two-dimensional representation of three-dimensional brain imaging data and to probe it for its diagnostic validity. To validate the projection outcome, we assessed the diagnostic accuracy of computer-aided stroke territory recognition based on geoprojected two-dimensional maps of diffusion MRI data.

More confusion at the link.