Lacunes of presumed vascular origin (lacunes) are associated with an increased risk of stroke, gait impairment, and dementia and are a primary imaging feature of the small vessel disease. Quantification of lacunes may be of great importance to elucidate the mechanisms behind neuro-degenerative disorders and is recommended as part of study standards for small vessel disease research. However, due to the different appearance of lacunes in various brain regions and the existence of other similar-looking structures, such as perivascular spaces, manual annotation is a difficult, elaborative and subjective task, which can potentially be greatly improved by reliable and consistent computer-aided detection (CAD) routines.In this paper, we propose an automated two-stage method using deep convolutional neural networks (CNN). We show that this method has good performance and can considerably benefit readers. We first use a fully convolutional neural network to detect initial candidates. In the second step, we employ a 3D CNN as a false positive reduction tool. As the location information is important to the analysis of candidate structures, we further equip the network with contextual information using multi-scale analysis and integration of explicit location features. We trained, validated and tested our networks on a large dataset of 1075 cases obtained from two different studies. Subsequently, we conducted an observer study with four trained observers and compared our method with them using a free-response operating characteristic analysis. Shown on a test set of 111 cases, the resulting CAD system exhibits performance similar to the trained human observers and achieves a sensitivity of 0.974 with 0.13 false positives per slice. A feasibility study also showed that a trained human observer would considerably benefit once aided by the CAD system.

Purpose Quality control (QC) is a prerequisite for clinical MR spectroscopic imaging (MRSI) to avoid that bad spectra hamper data interpretation. The aim of this work was to present a simple automatic QC for prostate 1H MRSI that can handle data obtained with different commonly used pulse sequences, echo times, field strengths, and MR platforms. Methods A QC method was developed with a ratio (Qratio) where the numerator and the denominator are functions of several signal heights, logically combined for their positive or negative contribution to spectral quality. This Qratio was tested on 4 data sets obtained at 1.5, 3, and 7T, with and without endorectal coil and different localization sequences and echo times. Spectra of 25,248 voxels in 26 prostates were labeled as acceptable or unacceptable by MRS experts as gold standard. A threshold value was determined for Qratio from a subset of voxels, labeled in consensus by 4 experts, for an optimal accuracy to separate spectra. Results Applying this Qratio threshold to the remaining test voxels, an automatic separation of good and bad spectra was possible with an accuracy of 0.88, similar to manual separation between the 2 classes. Qratio values were used to generate maps representing spectral quality on a binary or continuous scale. Conclusion Automated QC of prostate 1H MRSI by Qratio is fast, simple, easily transferable and more practical than supervised feature extraction methods and therefore easy to integrate into different clinical MR systems. Moreover, quality maps can be generated to read the reliability of spectra in each voxel.