In this paper, we present a novel automated method for White Matter (WM) lesion segmentation of Multiple Sclerosis (MS) patient images. Our approach is based on a cascade of two 3D patch-wise convolutional neural networks (CNN). The first network is trained to be more sensitive revealing possible candidate lesion voxels while the second network is trained to reduce the number of misclassified voxels coming from the first network. This cascaded CNN architecture tends to learn well from small sets of training data, which can be very interesting in practice, given the difficulty to obtain manual label annotations and the large amount of available unlabeled Magnetic Resonance Imaging (MRI) data. We evaluate the accuracy of the proposed method on the public MS lesion segmentation challenge MICCAI2008 dataset, comparing it with respect to other state-of-the-art MS lesion segmentation tools. Furthermore, the proposed method is also evaluated on two private MS clinical datasets, where the performance of our method is also compared with different recent public available state-of-the-art MS lesion segmentation methods. At the time of writing this paper, our method is the best ranked approach on the MICCAI2008 challenge, outperforming the rest of 60 participant methods when using all the available input modalities (T1-w, T2-w and FLAIR), while still in the top-rank (3rd position) when using only T1-w and FLAIR modalities. On clinical MS data, our approach exhibits a significant increase in the accuracy segmenting of WM lesions when compared with the rest of evaluated methods, highly correlating (r ≥ 0.97) also with the expected lesion volume.

Although not designed as a formal interobserver study, the current study suggests that comparing available literature data on cortical lesions may be problematic, and increased consistency in acquisition protocols may improve scoring agreement. Sensitivity and specificity of the proposed recommendations should now be studied in a more formal, prospective, multicenter setting using similar DIR protocols.

Early evaluation of treatment response and prediction of disease evolution are key issues in the management of people with multiple sclerosis (MS). In the past 20 years, MRI has become the most useful paraclinical tool in both situations and is used clinically to assess the inflammatory component of the disease, particularly the presence and evolution of focal lesions — the pathological hallmark of MS. However, diffuse neurodegenerative processes that are at least partly independent of inflammatory mechanisms can develop early in people with MS and are closely related to disability. The effects of these neurodegenerative processes at a macroscopic level can be quantified by estimation of brain and spinal cord atrophy with MRI. MRI measurements of atrophy in MS have also been proposed as a complementary approach to lesion assessment to facilitate the prediction of clinical outcomes and to assess treatment responses. In this Consensus statement, the Magnetic Resonance Imaging in MS (MAGNIMS) study group critically review the application of brain and spinal cord atrophy in clinical practice in the management of MS, considering the role of atrophy measures in prognosis and treatment monitoring and the barriers to clinical use of these measures. On the basis of this review, the group makes consensus statements and recommendations for future research.

In recent years, several convolutional neural network (CNN) methods have been proposed for the automated white matter lesion segmentation of multiple sclerosis (MS) patient images, due to their superior performance compared with those of other state-of-the-art methods. However, the accuracies of CNN methods tend to decrease significantly when evaluated on different image domains compared with those used for training, which demonstrates the lack of adaptability of CNNs to unseen imaging data. In this study, we analyzed the effect of intensity domain adaptation on our recently proposed CNN-based MS lesion segmentation method. Given a source model trained on two public MS datasets, we investigated the transferability of the CNN model when applied to other MRI scanners and protocols, evaluating the minimum number of annotated images needed from the new domain and the minimum number of layers needed to re-train to obtain comparable accuracy. Our analysis comprised MS patient data from both a clinical center and the public ISBI2015 challenge database, which permitted us to compare the domain adaptation capability of our model to that of other state-of-the-art methods. In both datasets, our results showed the effectiveness of the proposed model in adapting previously acquired knowledge to new image domains, even when a reduced number of training samples was available in the target dataset. For the ISBI2015 challenge, our one-shot domain adaptation model trained using only a single case showed a performance similar to that of other CNN methods that were fully trained using the entire available training set, yielding a comparable human expert rater performance. We believe that our experiments will encourage the MS community to incorporate its use in different clinical settings with reduced amounts of annotated data. This approach could be meaningful not only in terms of the accuracy in delineating MS lesions but also in the related reductions in time and economic costs derived from manual lesion labeling.

[ 11 C]PK11195 is used in positron emission tomography (PET) studies for imaging brain inflammation in vivo as it binds to the peripheral-type benzodiazepine receptor (PBR) expressed by reactive glia and macrophages. However, features of the cellular reaction required to induce a positive [ 11 C]PK11195 signal are not well characterized. We performed [ 11 C]PK11195 PET and autoradiography in rats after transient focal cerebral ischemia. We determined [ 3 H]PK11195 binding and PBR expression in brain tissue and examined the lesion with several markers. [ 11 C]PK11195 standard uptake value increased at day 4 and grew further at day 7 within the ischemic core. Accordingly, ex vivo [ 3 H]PK11195 binding increased at day 4, and increases further at day 7. The PET signal also augmented in peripheral regions, but to a lesser extent than in the core. Binding in the region surrounding infarction was supported by [ 11 C]PK11195 autoradiography at day 7 showing that the radioactive signal extended beyond the infarcted core. Enhanced binding was preceded by increases in PBR mRNA expression in the ipsilateral hemisphere, and a 18-kDa band corresponding to PBR protein was detected. Peripheraltype benzodiazepine receptor immunohistochemistry showed subsets of ameboid microglia/macrophages within the infarcted core showing a distinctive strong PBR expression from day 4. These cells were often located surrounding microhemorrhages. Reactive astrocytes forming a rim surrounding infarction at day 7 also showed some PBR immunostaining. These results show cellular heterogeneity in the level of PBR expression, supporting that PBR is not a simple marker of inflammation, and that the extent of [ 11 C]PK11195 binding depends on intrinsic features of the inflammatory cells.

Aromatase catalyzes the last step in estrogen biosynthesis. Brain aromatase is involved in diverse neurophysiological and behavioral functions including sexual behavior, aggression, cognition and neuroprotection. Using positron emission tomography (PET) with the radiolabeled aromatase inhibitor [N-methyl-11 C]vorozole, we characterized the tracer distribution and kinetics in the living human brain. Six young, healthy subjects, 3 men and 3 women, were administered the radiotracer alone on two separate occasions. Women were scanned in distinct phases of the menstrual cycle. Specificity was confirmed by pretreatment with a pharmacological (2.5mg) dose of the aromatase inhibitor letrozole. PET data were acquired over a 90 min period and regions of interest placed over selected brain regions. Brain and plasma time activity curves, corrected for metabolites, were used to derive kinetic parameters. Distribution volume (VT) values in both men and women followed the rank order: thalamus>amygdala=preoptic area>medulla(inferior olive) > accumbens, pons, occipital and temporal cortex, putamen, cerebellum and white matter. Pretreatment with letrozole reduced VT in all regions, though the size of the reduction was region dependent; ranging from ~70% blocking in thalamus and preoptic area to ~10% in cerebellum. The high levels of aromatase in thalamus and medulla (inferior olive) appear to be unique to humans. These studies set the stage for the non-invasive assessment of aromatase involvement in various physiological and pathological processes affecting the human brain.

Nanoparticles have been proposed for several biomedical applications; however, in vivo biodistribution studies to confirm their potential are scarce. Nanodiamonds are carbon nanoparticles that have been recently proposed as a promising biomaterial. In this study, we labeled nanodiamonds with (18)F to study their in vivo biodistribution by positron emission tomography. Moreover, the impact on the biodistribution of their kinetic particle size and of the surfactant agents has been evaluated. Radiolabeled diamond nanoparticles accumulated mainly in the lung, spleen, and liver and were excreted into the urinary tract. The addition of surfactant agents did not lead to significant changes in this pattern, with the exception of a slight reduction in the urinary excretion rate. On the other hand, after filtration of the radiolabeled diamond nanoparticles to remove those with a larger kinetic size, the uptake in the lung and spleen was completely inhibited and significantly reduced in the liver.

In a multicenter setting, we applied voxel-based methods to different structural MR imaging modalities to define the relative contributions of focal lesions, normal-appearing white matter (NAWM), and gray matter (GM) damage and their regional distribution to cognitive deficits as well as impairment of specific cognitive domains in multiple sclerosis (MS) patients. Approval of the institutional review boards was obtained, together with written informed consent from all participants. Standardized neuropsychological assessment and conventional, diffusion tensor and volumetric brain MRI sequences were collected from 61 relapsing-remitting MS patients and 61 healthy controls (HC) from seven centers. Patients with ≥2 abnormal tests were considered cognitively impaired (CI). The distribution of focal lesions, GM and WM atrophy, and microstructural WM damage were assessed using voxel-wise approaches. A random forest analysis identified the best imaging predictors of global cognitive impairment and deficits of specific cognitive domains. Twenty-three (38%) MS patients were CI. Compared with cognitively preserved (CP), CI MS patients had GM atrophy of the left thalamus, right hippocampus and parietal regions. They also showed atrophy of several WM tracts, mainly located in posterior brain regions and widespread WM diffusivity abnormalities. WM diffusivity abnormalities in cognitive-relevant WM tracts followed by atrophy of cognitive-relevant GM regions explained global cognitive impairment. Variable patterns of NAWM and GM damage were associated with deficits in selected cognitive domains. Structural, multiparametric, voxel-wise MRI approaches are feasible in a multicenter setting. The combination of different imaging modalities is needed to assess and monitor cognitive impairment in MS.