Automatic Detection of Chronic Insomnia from Polysomnographic and Clinical Variables Using Machine Learning.

Insomnia is defined subjectively by the presence and frequency of specific clinical symptoms and an association with distress. Although sleep study data has shown some weak associations, no objective test can currently be used to predict insomnia. The purpose of this study was to use previously reported and relatively crafted insomnia-related polysomnographic variables in machine learning models to classify groups with and without insomnia. Demographics, diagnosed depression, Epworth Sleepiness Scale (ESS), and features derived from electroencephalography (EEG), arousals, and sleep stages from 3,407 sleep clinic patients (2,617 without insomnia and 790 insomnia patients based on responses to a set of questions) were included in this analysis. The number of features were reduced using pair-wise correlation and recursive feature elimination. Predictive value of three machine learning models (logistic regression, neural network, and support vector machine) was investigated, and the best performance was achieved with logistic regression, yielding a balanced accuracy of 71%. The most important features in predicting insomnia were depression, age, sex, duration of longest arousal, ESS score, and EEG power in theta and sigma bands across all sleep stages. Results indicate potential of machine learning-based screening for insomnia using clinical variables and EEG.

Rapid microscopic 3D-diffusion tensor imaging fiber-tracking of mouse brain in vivo by super resolution reconstruction: validation on MAP6-KO mouse model.

OBJECT: Exploring mouse brains by rapid 3D-Diffusion Tensor Imaging (3D-DTI) of high spatial resolution (HSR) is challenging in vivo. Here we use the super resolution reconstruction (SRR) postprocessing method to demonstrate its performance on Microtubule-Associated-Protein6 Knock-Out (MAP6-KO) mice. MATERIALS AND METHODS: Two spin-echo DTI were acquired (9.4T, CryoProbe RF-coil): (i)-multislice 2D-DTI, (echo-planar integrating reversed-gradient) acquired in vivo in the three orthogonal orientations (360 µm slice-thickness, 120 × 120 µm in-plane resolution, 56 min scan duration); used in SRR software to reconstruct SRR 3D-DTI with HSR in slice-plane (120 × 120 × 120 µm) and (ii)-microscopic 3D-DTI (µ-3D-DTI), (100 × 100 × 100 µm; 8 h 6 min) on fixed-brains ex vivo, that were removed after paramagnetic contrast-agent injection to accelerate scan acquisition using short repetition-times without NMR-signal sensitivity loss. RESULTS: White-matter defects, quantified from both 3D-DTI fiber-tracking were found very similar. Indeed, as expected the fornix and cerebral-peduncle volume losses were - 39% and - 35% in vivo (SRR 3D-DTI) versus - 34% and - 32% ex vivo (µ-3D-DTI), respectively (p<0.001). This finding is robust since the µ-3D-DTI feasibility on MAP6-KO ex vivo was already validated by fluorescent-microscopy of cleared brains. DISCUSSION: First performance of the SRR to generate rapid HSR 3D-DTI of mouse brains in vivo is demonstrated. The method is suitable in neurosciences for longitudinal studies to identify molecular and genetic abnormalities in mouse models that are of growing developments.

3D imaging of the brain morphology and connectivity defects in a model of psychiatric disorders: MAP6-KO mice.

In the central nervous system, microtubule-associated protein 6 (MAP6) is expressed at high levels and is crucial for cognitive abilities. The large spectrum of social and cognitive impairments observed in MAP6-KO mice are reminiscent of the symptoms observed in psychiatric diseases, such as schizophrenia, and respond positively to long-term treatment with antipsychotics. MAP6-KO mice have therefore been proposed to be a useful animal model for these diseases. Here, we explored the brain anatomy in MAP6-KO mice using high spatial resolution 3D MRI, including a volumetric T1w method to image brain structures, and Diffusion Tensor Imaging (DTI) for white matter fiber tractography. 3D DTI imaging of neuronal tracts was validated by comparing results to optical images of cleared brains. Changes to brain architecture included reduced volume of the cerebellum and the thalamus and altered size, integrity and spatial orientation of some neuronal tracks such as the anterior commissure, the mammillary tract, the corpus callosum, the corticospinal tract, the fasciculus retroflexus and the fornix. Our results provide information on the neuroanatomical defects behind the neurological phenotype displayed in the MAP6-KO mice model and especially highlight a severe damage of the corticospinal tract with defasciculation at the location of the pontine nuclei.

In vivo quantification of magnetically labelled cells by MRI relaxometry.

Cellular MRI, which visualizes magnetically labelled cells (cells*), is an active research field for in vivo cell therapy and tracking. The simultaneous relaxation rate measurements (R2 *, R2 , R1 ) are the basis of a quantitative cellular MRI method proposed here. U937 cells were labelled with Molday ION Rhodamine B, a bi-functional superparamagnetic and fluorescent nanoparticle (U937*). U937* viability and proliferation were not affected in vitro. In vitro relaxometry was performed in a cell concentration range of [2.5 × 104 -108 ] cells/mL. These measurements show the existence of complementary cell concentration intervals where these rates vary linearly. The juxtaposition of these intervals delineates a wide cell concentration range over which one of the relaxation rates in a voxel of an in vivo image can be converted into an absolute cell concentration. The linear regime was found at high concentrations for R1 in the range of [106 - 2 × 108 ] cells/mL, at intermediate concentrations for R2 in [2.5 × 105 - 5 × 107 ] cells/mL and at low concentrations for R2 * in [8 × 104 - 5 × 106 ] cells/mL. In vivo relaxometry was performed in a longitudinal study, with labelled U937 cells injected into a U87 glioma mouse model. Using in vitro data, maps of in vivo U937* concentrations were obtained by converting one of the in vivo relaxation rates to cell concentration maps. MRI results were compared with the corresponding optical images of the same brains, showing the usefulness of our method to accurately follow therapeutic cell biodistribution in a longitudinal study. Results also demonstrate that the method quantifies a large range of magnetically labelled cells*. Copyright © 2016 John Wiley & Sons, Ltd.

Microscopic DTI accurately identifies early glioma cell migration: correlation with multimodal imaging in a new glioma stem cell model.

Monitoring glioma cell infiltration in the brain is critical for diagnosis and therapy. Using a new glioma Glio6 mouse model derived from human stem cells we show how diffusion tensor imaging (DTI) may predict glioma cell migration/invasion. In vivo multiparametric MRI was performed at one, two and three months of Glio6 glioma growth (Glio6 (n = 6), sham (n = 3)). This longitudinal study reveals the existence of a time window to study glioma cell/migration/invasion selectively. Indeed, at two months only Glio6 cell invasion was detected, while tumor mass formation, edema, blood-brain barrier leakage and tumor angiogenesis were detected later, at three months. To robustly confirm the potential of DTI for detecting glioma cell migration/invasion, a microscopic 3D-DTI (80 µm isotropic spatial resolution) technique was developed and applied to fixed mouse brains (Glio6 (n = 6), sham (n = 3)). DTI changes were predominant in the corpus callosum (CC), a known path of cell migration. Fractional anisotropy (FA) and perpendicular diffusivity (D⊥ ) changes derived from ex vivo microscopic 3D-DTI were significant at two months of tumor growth. In the caudate putamen an FA increase of +38% (p < 0.001) was observed, while in the CC a - 28% decrease in FA (p < 0.005) and a + 95% increase in D⊥ (p < 0.005) were observed. In the CC, DTI changes and fluorescent Glio6 cell density obtained by two-photon microscopy in the same brains were correlated (p < 0.001, r = 0.69), validating FA and D⊥ as early quantitative biomarkers to detect glioma cell migration/invasion. The origin of DTI changes was assessed by electron microscopy of the same tract, showing axon bundle disorganization. During the first two months, Glio6 cells display a migratory phenotype without being associated with the constitution of a brain tumor mass. This offers a unique opportunity to apply microscopic 3D-DTI and to validate DTI parameters FA and D⊥ as biomarkers for glioma cell invasion.