No abstract

Characterizing the subtle changes of functional brain networks associated with the pathological cascade of Alzheimer's disease (AD) is important for early diagnosis and prediction of disease progression prior to clinical symptoms. We developed a new deep learning method, termed multiple graph Gaussian embedding model (MG2G), which can learn highly informative network features by mapping high-dimensional resting-state brain networks into a lowdimensional latent space. These latent distribution-based embeddings enable a quantitative characterization of subtle and heterogeneous brain connectivity patterns at different regions, and can be used as input to traditional classifiers for various downstream graph analytic tasks, such as AD early stage prediction, and statistical evaluation of between-group significant alterations across brain regions. We used MG2G to detect the intrinsic latent dimensionality of MEG brain networks, predict the progression of patients with mild cognitive impairment (MCI) to AD, and identify brain regions with network alterations related to MCI.

New architectures of transparent conductive electrodes (TCEs) incorporating graphene monolayers in different configurations have been explored with the aim to improve the performance of silicon-heterojunction (SHJ) cell front transparent contacts. In SHJ technology, front electrodes play an important additional role as anti-reflectance (AR) coatings. In this work, different transparent-conductive-oxide (TCO) thin films have been combined with graphene monolayers in different configurations, yielding advanced transparent electrodes specifically designed to minimize surface reflection over a wide range of wavelengths and angles of incidence and to improve electrical performance. A preliminary analysis reveals a strong dependence of the optoelectronic properties of the TCEs on (i) the order in which the different thin films are deposited or the graphene is transferred and (ii) the specific TCO material used. The results shows a clear electrical improvement when three graphene monolayers are placed on top on 80-nm-thick ITO thin film. This optimum TCE presents sheet resistances as low as 55 Ω/sq and an average conductance as high as 13.12 mS. In addition, the spectral reflectance of this TCE also shows an important reduction in its weighted reflectance value of 2–3%. Hence, the work undergone so far clearly suggests the possibility to noticeably improve transparent electrodes with this approach and therefore to further enhance silicon-heterojunction cell performance. These results achieved so far clearly open the possibility to noticeably improve TCEs and therefore to further enhance SHJ contact-technology performance.

Hybrid transparent contacts based on combinations of a transparent conductive oxide and a few graphene monolayers were developed in order to evaluate their optical and electrical performance with the main aim to use them as front contacts in optoelectronic devices. The assessment of the most suitable strategies for their fabrication was performed by testing different protocols addressing such issues as the protection of the device structure underneath, the limitation of sample temperature during the graphene-monolayer transfer process and the determination of the most suitable stacking structure. Suitable metal ohmic electrodes were also evaluated. Among a number of options tested, the metal contact based on Ti + Ag showed the highest reproducibility and the lowest contact resistivity. Finally, with the objective of extracting the current generated from optoelectronic devices to the output pins of an external package, focusing on a near future commercial application, the electrical properties of the connections made with an ultrasonic bonding machine (sonic welding) between the optimized Ti + Ag metal contacts and Al or Au micro-wires were also evaluated. All these results have an enormous potential as hybrid electrodes based on graphene to be used in novel designs of a future generation of optoelectronic devices, such as solar cells.