Purpose
While early and accurate diagnosis of dementia is challenging,
computer-aided diagnosis methods based on quantitative biomarkers such as structural MRI can aid this (Klöppel et al.,
2012).
Thesemethods use machine learning techniques that train a model to differentiate groups and make the diagnosis.Such algorithms for computer-aided diagnosis of dementia have shown very promising results for diagnosing Alzheimer’s disease (AD) and mild cognitive impairment (MCI) (Falahati et al.,
2014).
However,
comparison of these algorithms is difficult as for evaluation often different data sets and methodology are...
Methods and materials
A total of 384 clinically representative T1-weighted MRI scans were acquired at the VU Medical Center (Amsterdam,
Netherlands),
Erasmus MC (Rotterdam,
Netherlands),
and University of Porto (Portugal).
Details on the data can be found in Fig. 2.
Fig. 3 shows an example scan for an AD patient.
The diagnostic label (AD,
MCI,
CN) was based on clinical criteria as reference standard.
The algorithms were trained on a small training set (n=30) and optionally on other data (e.g.,
Alzheimer’s Disease Neuroimaging Initiative (ADNI; Jack et al.,...
Results
As shown in Fig. 4 and Fig. 5,
the algorithms used a wide range of approaches.
Different feature extraction methods and classifiers were used.
Most of the algorithms were trained on the 30 provided training sets and on data from ADNI and AIBL.
The performances for the five best performing methods are shown in Fig. 6 (accuracy) and Fig. 7 (AUC).
Results for the other methods can be found on http://caddementia.grand-challenge.org.
The best performing algorithm (Sørensen et al.,
2014) yielded an AUC of 78.8% (CI:...
Conclusion
Public large-scale validation studies,
such as this work,
are an important step towards the implementation of high-potential algorithms for computer-aided diagnosis of dementia into clinical practice.
The web-based frameworkremains open for new algorithms to be compared.
For more details on this study,
please see Bron et al.
(2015).
Personal information
Esther E.
Bron (
[email protected]),
Wiro Niessen,
and Stefan Klein are with the Biomedical Imaging Group Rotterdam,
Departments of Medical Informatics and Radiology,
Erasmus MC,
Rotterdam,
the Netherlands.
Marion Smits is with the Department of Radiology,
Erasmus MC,
Rotterdam,
the Netherlands.
Frederik Barkhof is with the Department of Radiology & Nuclear Medicine,
VU University Medical Center,
Neuroscience Campus Amsterdam,
the Netherlands.
António J.
Bastos-Leite is with the Department of Medical Imaging,
Faculty of Medicine,
University of Porto,
Porto,
Portugal.
John C.
van Swieten is with the...
References
Bron EE,
Smits M,
van Swieten JC,
Niessen WJ,
Klein S (2014): Proc MICCAI workshop Challenge on Computer-Aided Diagnosis of Dementia Based on Structural MRI Data.Download.
Bron EE,
Smits M,
Van der Flier WM,
et al.
(2015): Standardized evaluation of algorithms for computer-aided diagnosis of dementia based on strucural MRI: the CADDementia challenge.
Neuroimage.
Ellis KA,
Bush AI,
Darby D,
De Fazio D,
Foster J,
Hudson P,
Lautenschlager NT,
Lenzo N,
Martins RN,
Maruff P,
Masters C,
Milner A,
Pike K,
Rowe C,
Savage G,...