The anatomical location of imaging features is of crucial importance for accurate diagnosis in many medical tasks. Convolutional neural networks (CNN) have had huge successes in computer vision, but they lack the natural ability to incorporate the anatomical location in their decision making process, hindering success in some medical image analysis tasks. In this paper, to integrate the anatomical location information into the network, we propose several deep CNN architectures that consider multi-scale patches or take explicit location features while training. We apply and compare the proposed architectures for segmentation of white matter hyperintensities in brain MR images on a large dataset. As a result, we observe that the CNNs that incorporate location information substantially outperform a conventional segmentation method with handcrafted features as well as CNNs that do not integrate location information. On a test set of 50 scans, the best configuration of our networks obtained a Dice score of 0.792, compared to 0.805 for an independent human observer. Performance levels of the machine and the independent human observer were not statistically significantly different (p-value = 0.06).

Objective:To investigate the temporal dynamics of cerebral small vessel disease (SVD) by 3 consecutive assessments over a period of 9 years, distinguishing progression from regression.Methods:Changes in SVD markers of 276 participants of the Radboud University Nijmegen Diffusion Tensor and Magnetic Resonance Imaging Cohort (RUN DMC) cohort were assessed at 3 time points over 9 years. We assessed white matter hyperintensities (WMH) volume by semiautomatic segmentation and rated lacunes and microbleeds manually. We categorized baseline WMH severity as mild, moderate, or severe according to the modified Fazekas scale. We performed mixed-effects regression analysis including a quadratic term for increasing age.Results:Mean WMH progression over 9 years was 4.7 mL (0.54 mL/y; interquartile range 0.95–5.5 mL), 20.3% of patients had incident lacunes (2.3%/y), and 18.9% had incident microbleeds (2.2%/y). WMH volume declined in 9.4% of the participants during the first follow-up interval, but only for 1 participant (0.4%) throughout the whole follow-up. Lacunes disappeared in 3.6% and microbleeds in 5.7% of the participants. WMH progression accelerated over time: including a quadratic term for increasing age during follow-up significantly improved the model (p < 0.001). SVD progression was predominantly seen in participants with moderate to severe WMH at baseline compared to those with mild WMH (odds ratio [OR] 35.5, 95% confidence interval [CI] 15.8–80.0, p < 0.001 for WMH progression; OR 5.7, 95% CI 2.8–11.2, p < 0.001 for incident lacunes; and OR 2.9, 95% CI 1.4–5.9, p = 0.003 for incident microbleeds).Conclusions:SVD progression is nonlinear, accelerating over time, and a highly dynamic process, with progression interrupted by reduction in some, in a population that on average shows progression.

Cigarette smoking doubles the risk of dementia and Alzheimer's disease. Various pathophysiological pathways have been proposed to cause such a cognitive decline, but the exact mechanisms remain unclear. Smoking may affect the microstructural integrity of cerebral white matter. Diffusion tensor imaging is known to be sensitive for microstructural changes in cerebral white matter. We therefore cross-sectionally studied the relation between smoking behaviour (never, former, current) and diffusion tensor imaging parameters in both normal-appearing white matter and white matter lesions as well as the relation between smoking behaviour and cognitive performance. A structured questionnaire was used to ascertain the amount and duration of smoking in 503 subjects with small-vessel disease, aged between 50 and 85 years. Cognitive function was assessed with a neuropsychological test battery. All subjects underwent 1.5 Tesla magnetic resonance imaging. Using diffusion tensor imaging, fractional anisotropy and mean diffusivity were calculated in both normal-appearing white matter and white matter lesions. A history of smoking was associated with significant higher values of mean diffusivity in normal-appearing white matter and white matter lesions (P-trend for smoking status = 0.02) and with poorer cognitive functioning compared with those who never smoked. Associations with smoking and loss of structural integrity appeared to be strongest in normal-appearing white matter. Furthermore, the duration of smoking cessation was positively related to lower values of mean diffusivity and higher values of fractional anisotropy in normal-appearing white matter [β = -0.004 (95% confidence interval -0.007 to 0.000; P = 0.03) and β = 0.019 (95% confidence interval 0.001-0.038; P = 0.04)]. Fractional anisotropy and mean diffusivity values in normal-appearing white matter of subjects who had quit smoking for >20 years were comparable with subjects who had never smoked. These data suggest that smoking affects the microstructural integrity of cerebral white matter and support previous data that smoking is associated with impaired cognition. Importantly, they suggest that quitting smoking may reverse the impaired structural integrity.

Background and Purpose-Gait disorders are common in the elderly and are related to loss of functional independence and death. White matter lesions (WMLs) may be related, but only a minority of individuals with WMLs has gait disorders. Probably other factors are involved, including location and the independent effect of frequently coinciding lacunar infarcts, the other aspect of cerebral small vessel disease. The aim of our study was to investigate the effect of both the severity and location of both WMLs and lacunar infarcts on gait. Methods-Four hundred thirty-one independently living, nondemented elderly aged between 50 and 85 years with cerebral small vessel disease were included in this analysis and underwent MRI scanning. The number and location of lacunar infarcts were rated and WML volume was assessed by manual segmentation with automated delineating of different regions. Gait was assessed quantitatively with an electronic walkway as well as the semiquantitatively Tinetti and Timed-Up-and-Go test. Results-WMLs and lacunar infarcts were both independently associated with most gait parameters with stride length as the most sensitive parameter related to WMLs. WMLs in the sublobar (basal ganglia/internal capsule) and limbic areas and lacunar infarcts in the frontal lobe and thalamus were related to a lower velocity.

These results support a role of network disruption playing a pivotal role in the genesis of dementia in SVD, and suggest network analysis of the connectivity of white matter has potential as a predictive marker in the disease.