Machine and Deep Learning Approaches Applied to Classify Gougerot-Sjögren Syndrome and Jointly Segment Salivary Glands.

To diagnose Gougerot-Sjögren syndrome (GSS), ultrasound imaging (US) is a promising tool for helping physicians and experts. Our project focuses on the automatic detection of the presence of GSS using US. Ultrasound imaging suffers from a weak signal-to-noise ratio. Therefore, any classification or segmentation task based on these images becomes a difficult challenge. To address these two tasks, we evaluate different approaches: a classification using a machine learning method along with feature extraction based on a set of measurements following the radiomics guidance and a deep-learning-based classification. We propose, therefore, an innovative method to enhance the training of a deep neural network with a two phases: multiple supervision using joint classification and a segmentation implemented as pretraining. We highlight the fact that our learning methods provide segmentation results similar to those performed by human experts. We obtain proficient segmentation results for salivary glands and promising detection results for Gougerot-Sjögren syndrome; we observe maximal accuracy with the model trained in two phases. Our experimental results corroborate the fact that deep learning and radiomics combined with ultrasound imaging can be a promising tool for the above-mentioned problems.

Active learning strategy and hybrid training for infarct segmentation on diffusion MRI with a U-shaped network.

Automatic and reliable stroke lesion segmentation from diffusion magnetic resonance imaging (MRI) is critical for patient care. Methods using neural networks have been developed, but the rate of false positives limits their use in clinical practice. A training strategy applied to three-dimensional deconvolutional neural networks for stroke lesion segmentation on diffusion MRI was proposed. Infarcts were segmented by experts on diffusion MRI for 929 patients. We divided each database as follows: 60% for a training set, 20% for validation, and 20% for testing. Our hypothesis was a two-phase hybrid learning scheme, in which the network was first trained with whole MRI (regular phase) and then, in a second phase (hybrid phase), alternately with whole MRI and patches. Patches were actively selected from the discrepancy between expert and model segmentation at the beginning of each batch. On the test population, the performances after the regular and hybrid phases were compared. A statistically significant Dice improvement with hybrid training compared with regular training was demonstrated ( p < 0.01 ). The mean Dice reached 0.711 ± 0.199 . False positives were reduced by almost 30% with hybrid training ( p < 0.01 ). Our hybrid training strategy empowered deep neural networks for more accurate infarct segmentations on diffusion MRI.

Bottom-up fabrication of InAs-on-nothing MOSFET using selective area molecular beam epitaxy.

In this paper we report on the fabrication and electrical characterization of InAs-on-nothing metal-oxide-semiconductor field-effect transistor composed of a suspended InAs channel and raised InAs n+ contacts. This architecture is obtained using 3D selective and localized molecular beam epitaxy on a lattice mismatched InP substrate. The suspended InAs channel and InAs n+ contacts feature a reproducible and uniform shape with well-defined 3D sidewalls. Devices with 1 µm gate length present a saturation drain current (I Dsat) of 300 mA mm-1 at V DS = 0.8 V and a trans-conductance (GM ) of 120 mS mm-1 at V DS = 0.5 V. In terms of electrostatic control, the devices display a minimal subthreshold swing of 110 mV dec-1 at V DS = 0.5 V and a small drain induced barrier lowering of 50 mV V-1.

MOCVD Growth of High-Quality and Density-Tunable GaAs Nanowires on ITO Catalyzed by Au Nanoparticles Deposited by Centrifugation.

High-quality and density-tunable GaAs nanowires (NWs) are directly grown on indium tin oxide (ITO) using Au nanoparticles (NPs) as catalysts by metal organic chemical vapor deposition (MOCVD). Au catalysts were deposited on ITO glass substrate using a centrifugal method. Compared with the droplet-only method, high-area density Au NPs were uniformly distributed on ITO. Tunable area density was realized through variation of the centrifugation time, and the highest area densities were obtained as high as 490 and 120 NP/µm(2) for 10- and 20-nm diameters of Au NPs, respectively. Based on the vapor-liquid-solid growth mechanism, the growth rates of GaAs NWs at 430 °C were 18.2 and 21.5 nm/s for the highest area density obtained of 10- and 20-nm Au NP-catalyzed NWs. The growth rate of the GaAs NWs was reduced with the increase of the NW density due to the competition of precursor materials. High crystal quality of the NWs was also obtained with no observable planar defects. 10-nm Au NP-induced NWs exhibit wurtzite structure whereas zinc-blende is observed for 20-nm NW samples. Controllable density and high crystal quality of the GaAs NWs on ITO demonstrate their potential application in hybrid a solar cell.