Affiliations: [a] Department of Neuroradiology, Klinikum rechts der Isar, Technical University of Munich, Munich, Germany
| [b] Department of Radiology, University Hospital, LMU Munich, Munich, Germany
| [c] Department of Psychiatry and Psychotherapy, Klinikum rechts der Isar, Technical University of Munich, Munich, Germany
| [d] Department of Neurology, Inselspital Bern, University of Bern, Bern, Switzerland
| [e] Department of Nuclear Medicine, Klinikum rechts der Isar, Technical University of Munich, Munich, Germany
Correspondence:
[*]
Correspondence to: Dr. Dennis M. Hedderich, MD, Department of Neuroradiology, Klinikum rechts der Isar, Ismaninger Str. 22, D-81675 Munich, Germany. E-mail: dennis.hedderich@tum.de.
Abstract: Background:Volumetric quantification of structural MRI has been shown to increase the diagnostic accuracy of patients with mild cognitive impairment (MCI); however, its implementation in clinical routine is usually technically difficult and time-consuming. Objective:The purpose of this study was to investigate whether volumetric information obtained from the free and easy-to-use online tool volBrain can improve correct identification of MCI patients with Alzheimer’s disease (AD) compared to visual reading. Methods:The study cohort consisted of 27 patients with MCI due to AD (AD positive) as determined by biomarker information and 26 cognitively normal controls (CN). Three blinded readers, 2 radiologists and 1 clinical dementia expert, assessed the patients’ MRI regarding brain atrophy and probability of underlying AD two times, without and with supporting volumetric information from volBrain. To assess diagnostic accuracy of volBrain measures alone, a simple sum score based on basic volumetric measures was developed and tested. Results:Correct patient classification by readers 1, 2, and 3 without a volumetric report was 73.6%, 77.4%, and 83.0%. With a volumetric report, correct classification increased for the radiological readers to 77.4% and 81.1%, respectively and decreased to 77.4% for reader 3. Usage of the volumetric report alone yielded the highest diagnostic accuracy of 84.9%. Diagnostic confidence increased significantly for radiological readers. Conclusion:Volumetric information from volBrain increases the radiologist’s diagnostic performance and confidence in identifying MCI patients with AD. We propose that such tools may be implemented in the routine diagnostic work-up of patients with suspected AD.
