Using U-Net convolutional neural network to model pixel-based electrostatic potential distributions in GaN power MIS-HEMTs.

This study demonstrates a novel use of the U-Net convolutional neural network (CNN) for modeling pixel-based electrostatic potential distributions in GaN metal-insulator-semiconductor high-electron mobility transistors (MIS-HEMTs) with various gate and source field plate designs and drain voltages. The pixel-based images of the potential distribution are successfully modeled from the developed U-Net CNN with an error of less than 1% error relative to a TCAD simulated reference of a 500-V electrostatic potential distribution in the AlGaN/GaN interface. Furthermore, the modeling time of potential distributions by U-Net takes about 80 ms. Therefore, the U-Net CNN is a promising approach to efficiently model the pixel-based distributions characteristics in GaN power devices.

Prediction of Bone Healing around Dental Implants in Various Boundary Conditions by Deep Learning Network.

Tissue differentiation varies based on patients' conditions, such as occlusal force and bone properties. Thus, the design of the implants needs to take these conditions into account to improve osseointegration. However, the efficiency of the design procedure is typically not satisfactory and needs to be significantly improved. Thus, a deep learning network (DLN) is proposed in this study. A data-driven DLN consisting of U-net, ANN, and random forest models was implemented. It serves as a surrogate for finite element analysis and the mechano-regulation algorithm. The datasets include the history of tissue differentiation throughout 35 days with various levels of occlusal force and bone properties. The accuracy of day-by-day tissue differentiation prediction in the testing dataset was 82%, and the AUC value of the five tissue phenotypes (fibrous tissue, cartilage, immature bone, mature bone, and resorption) was above 0.86, showing a high prediction accuracy. The proposed DLN model showed the robustness for surrogating the complex, time-dependent calculations. The results can serve as a design guideline for dental implants.

Device simulations with A U-Net model predicting physical quantities in two-dimensional landscapes.

Although Technology Computer-Aided Design (TCAD) simulation has paved a successful and efficient way to significantly reduce the cost of experiments under the device design, it still encounters many challenges as the semiconductor industry goes through rapid development in recent years, i.e. Complex 3D device structures, power devices. Recently, although machine learning has been proposed to enable the simulation acceleration and inverse­design of devices, which can quickly and accurately predict device performance, up to now physical quantities (such as electric field, potential energy, quantum-mechanically confined carrier distributions, and so on) being essential for understanding device physics can still only be obtained by traditional time-consuming self-consistent calculation. In this work, we employ a modified U-Net and train the models to predict the physical quantities of a MOSFET in two-dimensional landscapes for the first time. Errors in predictions by the two models have been analyzed, which shows the importance of a sufficient amount of data to prediction accuracy. The computation time for one landscape prediction with high accuracy by our well-trained U-Net model is much faster than the traditional approach. This work paves the way for interpretable predictions of device simulations based on convolutional neural networks.

Artificial intelligence deep learning for 3D IC reliability prediction.

Three-dimensional integrated circuit (3D IC) technologies have been receiving much attention recently due to the near-ending of Moore's law of minimization in 2D IC. However, the reliability of 3D IC, which is greatly influenced by voids and failure in interconnects during the fabrication processes, typically requires slow testing and relies on human's judgement. Thus, the growing demand for 3D IC has generated considerable attention on the importance of reliability analysis and failure prediction. This research conducts 3D X-ray tomographic images combining with AI deep learning based on a convolutional neural network (CNN) for non-destructive analysis of solder interconnects. By training the AI machine using a reliable database of collected images, the AI can quickly detect and predict the interconnect operational faults of solder joints with an accuracy of up to 89.9% based on non-destructive 3D X-ray tomographic images. The important features which determine the "Good" or "Failure" condition for a reflowed microbump, such as area loss percentage at the middle cross-section, are also revealed.

Engineer design process assisted by explainable deep learning network.

Engineering simulation accelerates the development of reliable and repeatable design processes in various domains. However, the computing resource consumption is dramatically raised in the whole development processes. Making the most of these simulation data becomes more and more important in modern industrial product design. In the present study, we proposed a workflow comprised of a series of machine learning algorithms (mainly deep neuron networks) to be an alternative to the numerical simulation. We have applied the workflow to the field of dental implant design process. The process is based on a complex, time-dependent, multi-physical biomechanical theory, known as mechano-regulatory method. It has been used to evaluate the performance of dental implants and to assess the tissue recovery after the oral surgery procedures. We provided a deep learning network (DLN) with calibrated simulation data that came from different simulation conditions with experimental verification. The DLN achieves nearly exact result of simulated bone healing history around implants. The correlation of the predicted essential physical properties of surrounding bones (e.g. strain and fluid velocity) and performance indexes of implants (e.g. bone area and bone-implant contact) were greater than 0.980 and 0.947, respectively. The testing AUC values for the classification of each tissue phenotype were ranging from 0.90 to 0.99. The DLN reduced hours of simulation time to seconds. Moreover, our DLN is explainable via Deep Taylor decomposition, suggesting that the transverse fluid velocity, upper and lower parts of dental implants are the keys that influence bone healing and the distribution of tissue phenotypes the most. Many examples of commercial dental implants with designs which follow these design strategies can be found. This work demonstrates that DLN with proper network design is capable to replace complex, time-dependent, multi-physical models/theories, as well as to reveal the underlying features without prior professional knowledge.

NeuroRetriever: Automatic Neuron Segmentation for Connectome Assembly.

Segmenting individual neurons from a large number of noisy raw images is the first step in building a comprehensive map of neuron-to-neuron connections for predicting information flow in the brain. Thousands of fluorescence-labeled brain neurons have been imaged. However, mapping a complete connectome remains challenging because imaged neurons are often entangled and manual segmentation of a large population of single neurons is laborious and prone to bias. In this study, we report an automatic algorithm, NeuroRetriever, for unbiased large-scale segmentation of confocal fluorescence images of single neurons in the adult Drosophila brain. NeuroRetriever uses a high-dynamic-range thresholding method to segment three-dimensional morphology of single neurons based on branch-specific structural features. Applying NeuroRetriever to automatically segment single neurons in 22,037 raw brain images, we successfully retrieved 28,125 individual neurons validated by human segmentation. Thus, automated NeuroRetriever will greatly accelerate 3D reconstruction of the single neurons for constructing the complete connectomes.

Kaleido: Visualizing Big Brain Data with Automatic Color Assignment for Single-Neuron Images.

Effective 3D visualization is essential for connectomics analysis, where the number of neural images easily reaches over tens of thousands. A formidable challenge is to simultaneously visualize a large number of distinguishable single-neuron images, with reasonable processing time and memory for file management and 3D rendering. In the present study, we proposed an algorithm named "Kaleido" that can visualize up to at least ten thousand single neurons from the Drosophila brain using only a fraction of the memory traditionally required, without increasing computing time. Adding more brain neurons increases memory only nominally. Importantly, Kaleido maximizes color contrast between neighboring neurons so that individual neurons can be easily distinguished. Colors can also be assigned to neurons based on biological relevance, such as gene expression, neurotransmitters, and/or development history. For cross-lab examination, the identity of every neuron is retrievable from the displayed image. To demonstrate the effectiveness and tractability of the method, we applied Kaleido to visualize the 10,000 Drosophila brain neurons obtained from the FlyCircuit database ( http://www.flycircuit.tw/modules.php?name=kaleido ). Thus, Kaleido visualization requires only sensible computer memory for manual examination of big connectomics data.

Differential synchrotron X-ray imaging markers based on the renal microvasculature for tubulointerstitial lesions and glomerulopathy.

High resolution synchrotron microtomography capable of revealing microvessels in three dimensional (3D) establishes distinct imaging markers of mouse kidney disease strongly associated to renal tubulointerstitial (TI) lesions and glomerulopathy. Two complementary mouse models of chronic kidney disease (CKD), unilateral ureteral obstruction (UUO) and focal segmental glomerulosclerosis (FSGS), were used and five candidates of unique 3D imaging markers were identified. Our characterization to differentially reflect the altered microvasculature of renal TI lesions and/or glomerulopathy demonstrated these image features can be used to differentiate the disease status and the possible cause therefore qualified as image markers. These 3D imaging markers were further correlated with the histopathology and renal microvessel-based molecular study using antibodies against vascular endothelial cells (CD31), the connective tissue growth factor or the vascular endothelial growth factor. We also found that these 3D imaging markers individually characterize the development of renal TI lesions or glomerulopathy, quantitative and integrated use of all of them provide more information for differentiating the two renal conditions. Our findings thus establish a practical strategy to characterize the CKD-associated renal injuries by the microangiography-based 3D imaging and highlight the impact of dysfunctional microvasculature as a whole on the pathogenesis of the renal lesions.

Folding a protein with equal probability of being helix or hairpin.

We explore the possibility for the native structure of a protein being inherently multiconformational in an ab initio coarse-grained model. Based on the Wang-Landau algorithm, the complete free energy landscape for the designed sequence 2DX4: INYWLAHAKAGYIVHWTA is constructed. It is shown that 2DX4 possesses two nearly degenerate native structures: one is a helix structure with the other a hairpinlike structure, and their free energy difference is <2% of that of local minima. Two degenerate native structures are stabilized by an energy barrier of ∼10 kcal/mol. Furthermore, the hydrogen-bond and dipole-dipole interactions are found to be two major competing interactions in transforming one conformation into the other. Our results indicate that two degenerate native structures are stabilized by subtle balance between different interactions in proteins. In particular, for small proteins, balance between the hydrogen-bond and dipole-dipole interactions happens for proteins of sizes being ∼18 amino acids and is shown to the main driving mechanism for the occurrence of degeneracy. These results provide important clues to the study of native structures of proteins.

Detecting small lung tumors in mouse models by refractive-index microradiology.

Refractive-index (phase-contrast) radiology was able to detect lung tumors less than 1 mm in live mice. Significant micromorphology differences were observed in the microradiographs between normal, inflamed, and lung cancer tissues. This was made possible by the high phase contrast and by the fast image taking that reduces the motion blur. The detection of cancer and inflammation areas by phase contrast microradiology and microtomography was validated by bioluminescence and histopathological analysis. The smallest tumor detected is less than 1 mm(3) with accuracy better than 1 × 10(-3) mm(3). This level of performance is currently suitable for animal studies, while further developments are required for clinical application.

Monte Carlo simulations of proteins in cages: influence of confinement on the stability of intermediate states.

We study the folding of small proteins inside confining potentials. Proteins are described using an effective potential model that contains the Ramachandran angles as degrees of freedom and does not need any a priori information about the native state. Hydrogen bonds, dipole-dipole-, and hydrophobic interactions are taken explicitly into account. An interesting feature displayed by this potential is the presence of metastable intermediates between the unfolded and native states. We consider different types of confining potentials to describe proteins folding inside cages with repulsive or attractive walls. Using the Wang-Landau algorithm, we determine the density of states and analyze in detail the thermodynamical properties of the confined proteins for different sizes of the cages. We show that confinement dramatically reduces the phase space available to the protein and that the presence of intermediate states can be controlled by varying the properties of the confining potential. Cages with strongly attractive walls destabilize the intermediate states and lead to a two-state folding into a configuration that is less stable than the native structure. However, cages with slightly attractive walls enhance the stability of native structure and induce a folding process, which occurs through intermediate configurations.

Effective potentials for folding proteins.

A coarse-grained off-lattice model that is not biased in any way to the native state is proposed to fold proteins. To predict the native structure in a reasonable time, the model has included the essential effects of water in an effective potential. Two new ingredients, the dipole-dipole interaction and the local hydrophobic interaction, are introduced and are shown to be as crucial as the hydrogen bonding. The model allows successful folding of the wild-type sequence of protein G and may have provided important hints to the study of protein folding.