Refine
Document Type
- Article (1)
- Doctoral Thesis (1)
Language
- English (2)
Has Fulltext
- yes (2)
Is part of the Bibliography
- no (2)
Keywords
- - (1)
- Alzheimer's disease (1)
- FreeSurfer (1)
- Kernspintomographie (1)
- Magnetic Resonance Imaging (1)
- Mustererkennung (1)
- Pattern Recognition (1)
- Prostata (1)
- Prostate (1)
- dementia (1)
Institute
Publisher
Introduction: It has been shown that Alzheimer’s disease (AD) is accompanied by marked structural brain changes that can be detected several years before clinical diagnosis via structural magnetic resonance (MR) imaging. In this study, we developed a structural MR-based biomarker for in vivo detection of AD using a supervised machine learning approach. Based on an individual’s pattern of brain atrophy a continuous AD score is assigned which measures the similarity with brain atrophy patterns seen in clinical cases of AD.
Methods: The underlying statistical model was trained with MR scans of patients and healthy controls from the Alzheimer’s Disease Neuroimaging Initiative (ADNI-1 screening). Validation was performed within ADNI-1 and in an independent patient sample from the Open Access Series of Imaging Studies (OASIS-1). In addition, our analyses included data from a large general population sample of the Study of Health in Pomerania (SHIP-Trend).
Results: Based on the proposed AD score we were able to differentiate patients from healthy controls in ADNI-1 and OASIS-1 with an accuracy of 89% (AUC = 95%) and 87% (AUC = 93%), respectively. Moreover, we found the AD score to be significantly associated with cognitive functioning as assessed by the Mini-Mental State Examination in the OASIS-1 sample after correcting for diagnosis, age, sex, age·sex, and total intracranial volume (Cohen’s f2 = 0.13). Additional analyses showed that the prediction accuracy of AD status based on both the AD score and the MMSE score is significantly higher than when using just one of them. In SHIP-Trend we found the AD score to be weakly but significantly associated with a test of verbal memory consisting of an immediate and a delayed word list recall (again after correcting for age, sex, age·sex, and total intracranial volume, Cohen’s f2 = 0.009). This association was mainly driven by the immediate recall performance.
Discussion: In summary, our proposed biomarker well differentiated between patients and healthy controls in an independent test sample. It was associated with measures of cognitive functioning both in a patient sample and a general population sample. Our approach might be useful for defining robust MR-based biomarkers for other neurodegenerative diseases, too.
Prostate volume estimation in MR images for epidemiological and clinical studies – new methods
(2014)
Benign prostatic hyperplasia (BPH) is one of the most widespread diseases among men older than 50 years. The literature provides various cut-off values for pathological enlargement of the prostate. Prostate volume (PV) measurement in large population-based studies would allow deriving more objective reference values and a more valid early BPH diagnosis. A fully automated method is therefore required. In the clinical context, the measurement of the PV is important for treatment response monitoring in the clinical applications for BPH management research, and an accurate method for PV is essential. Magnetic Resonance Imaging was used for PV estimation. Two methods based on the Support Vector Machines (SVM) were developed: the binary Support Vector Machines (C SVM)-based method for epidemiological studies and the single-class Support Vector Machines (S SVM)-based method for clinical studies. The second method was additionally compared to the ellipsoid formula for PV estimation, which is widespread in the clinic. The comparison between volume measurement of the C SVM-based method and manual delineation of observers A and B yielded a strong correlation (Spearmans rank correlation coefficients ñ of 0.936 [p < 0.001] and 0.859 [p < 0.001], respectively). Comparing the C SVM-based method and the two manual delineations by observers A and B shows an agreement with a mean difference of 3.0 ml (95% confidence interval of -3.1 to +9.2 ml) and 1.9 ml (95% confidence interval of −7.1 to +10.8 ml), respectively. The S SVM-based method and the reference PV (manual delineation of observer A) show excellent correlation (Spearmans rank correlation coefficient ñ = 0.965, p < 0.001), while the ellipsoid formula is less well correlated with the reference PV (Spearmans rank correlation coefficient ñ = 0.873, p < 0.001). The mean difference between S SVM and the reference PV was −0.05 ml (95% confidence interval of −3.8 to +3.7 ml); on the other hand, the mean difference between the ellipsoid formula and the reference PV was much greater, with 8.6 ml (95% confidence interval of +1 to +16.2 ml). The C SVM-based method has considerable potential for integration in epidemiological studies. The prostate volumes obtained by the S SVM-based method agreed excellently with the reference and would be clinically useful for urologists in prostate volumetric analysis.