Performance of a deep convolutional neural network to classify crystal structures using selected area electron beam diffraction patterns containing lattice defect information.

A deep convolutional neural network (DCNN) architecture ResNet has been tested to verify its ability to handle selected area electron diffraction pattern (SADP) datasets carrying information on lattice defects including strains, thermal lattice vibrations, point defects, dislocations, and twin boundaries. The disordered states of the crystal lattices in the presence of these defects were predicted by ab initio molecular dynamics simulations, first principles geometry optimizations, and lattice manipulation operations in an effort to establish a possible dataset augmentation strategy for the improvement of classification performance of the ResNet. Using the disordered lattice information originating from the defects, test dataset SADPs were generated by simulating electron diffraction in transmission electron microscopy. The ResNet, pre-trained using SADPs from defect-free materials, showed decreasing but acceptable classification accuracies with increasing degrees of lattice disorder regarding the lattice vibrations and point defects. When tested using the diffraction patterns for strained lattices, the ResNet responded to the changing lattice symmetry when strain levels are relatively high suggesting that it has capability to discern different symmetries induced by large strains. However, the ResNet failed to recognize lattice structure when dislocations and twin boundaries were considered. It is suggested that DCNN architectures be trained over various scenarios including changes in the image feature characteristics in the diffraction patterns related to defects in future developments for improved general classification performances.

Specular Detection on Glossy Surface Using Geometric Characteristics of Specularity in Top-View Images.

In an autonomous driving assistance system (ADAS), top views effectively represent objects around the vehicle on a 2D plane. Top-view images are therefore widely used to detect lines in ADAS applications such as lane-keeping assistance and parking assistance. Because line detection is a crucial step for these applications, the false positive detection of lines can lead to failure of the system. Specular reflections from a glossy surface are often the cause of false positives, and since certain specular patterns resemble actual lines in the top-view image, their presence induces false positive lines. Incorrect positions of the lines or parking stalls can thus be obtained. To alleviate this problem, we propose two methods to estimate specular pixels in the top-view image. The methods use a geometric property of the specular region: the shape of the specular region is stretched long in the direction of the camera as the distance between the camera and the light source becomes distant, resulting in a straight line. This property can be used to distinguish the specular region in images. One estimates the pixel-wise probability of the specularity using gradient vectors obtained from an edge detector and the other estimates specularity using the line equation of each line segment obtained by line detection. To evaluate the performance of the proposed method, we added our methods as a pre-processing step to existing parking stall detection methods and investigated changes in their performance. The proposed methods improved line detection performance by accurately estimating specular components in the top-view images.

Classification of crystal structures using electron diffraction patterns with a deep convolutional neural network.

Investigations have been made to explore the applicability of an off-the-shelf deep convolutional neural network (DCNN) architecture, residual neural network (ResNet), to the classification of the crystal structure of materials using electron diffraction patterns without prior knowledge of the material systems under consideration. The dataset required for training and validating the ResNet architectures was obtained by the computer simulation of the selected area electron diffraction (SAD) in transmission electron microscopy. Acceleration voltages, zone axes, and camera lengths were used as variables and crystal information format (CIF) files obtained from open crystal data repositories were used as inputs. The cubic crystal system was chosen as a model system and five space groups of 213, 221, 225, 227, and 229 in the cubic system were selected for the test and validation, based on the distinguishability of the SAD patterns. The simulated diffraction patterns were regrouped and labeled from the viewpoint of computer vision, i.e., the way how the neural network recognizes the two-dimensional representation of three-dimensional lattice structure of crystals, for improved training and classification efficiency. Comparison of the various ResNet architectures with varying number of layers demonstrated that the ResNet101 architecture could classify the space groups with the validation accuracy of 92.607%.

Continuous Separation of Circulating Tumor Cells from Whole Blood Using a Slanted Weir Microfluidic Device.

The separation of circulating tumor cells (CTCs) from the peripheral blood is an important issue that has been highlighted because of their high clinical potential. However, techniques that depend solely on tumor-specific surface molecules or just the larger size of CTCs are limited by tumor heterogeneity. Here, we present a slanted weir microfluidic device that utilizes the size and deformability of CTCs to separate them from the unprocessed whole blood. By testing its ability using a highly invasive breast cancer cell line, our device achieved a 97% separation efficiency, while showing an 8-log depletion of erythrocytes and 5.6-log depletion of leukocytes. We also developed an image analysis tool that was able to characterize the various morphologies and differing deformability of the separating cells. From the results, we believe our system possesses a high potential for liquid biopsy, aiding future cancer research.

Smartphone-Based Point-of-Care Urinalysis Under Variable Illumination.

Urine tests are performed by using an off-the-shelf reference sheet to compare the color of test strips. However, the tabular representation is difficult to use and more prone to visual errors, especially when the reference color-swatches to be compared are spatially apart. Thus, making it is difficult to distinguish between the subtle differences of shades on the reagent pads. This manuscript represents a new arrangement of reference arrays for urine test strips (urinalysis). Reference color swatches are grouped in a doughnut chart, surrounding each reagent pad on the strip. The urine test can be evaluated using naked eye by referring to the strip with no additional sheet necessary. Along with this new strip, an algorithm for smartphone based application is also proposed as an alternative to deliver diagnostic results. The proposed colorimetric detection method evaluates the captured image of the strip, under various color spaces and evaluates ten different tests for urine. Thus, the proposed system can deliver results on the spot using both naked eye and smartphone. The proposed scheme delivered accurate results under various environmental illumination conditions without any calibration requirements, exhibiting performances suitable for real-life applications and an ease for a common user.

Hydrochromic conjugated polymers for human sweat pore mapping.

Hydrochromic materials have been actively investigated in the context of humidity sensing and measuring water contents in organic solvents. Here we report a sensor system that undergoes a brilliant blue-to-red colour transition as well as 'Turn-On' fluorescence upon exposure to water. Introduction of a hygroscopic element into a supramolecularly assembled polydiacetylene results in a hydrochromic conjugated polymer that is rapidly responsive (<20 µs), spin-coatable and inkjet-compatible. Importantly, the hydrochromic sensor is found to be suitable for mapping human sweat pores. The exceedingly small quantities (sub-nanolitre) of water secreted from sweat pores are sufficient to promote an instantaneous colorimetric transition of the polymer. As a result, the sensor can be used to construct a precise map of active sweat pores on fingertips. The sensor technology, developed in this study, has the potential of serving as new method for fingerprint analysis and for the clinical diagnosis of malfunctioning sweat pores.