Optical spectrum augmentation for machine learning powered spectroscopic ellipsometry.

As structures of semiconductors become more complex and finer, the importance of an accurate measurement system has emerged. Previous studies have suggested various methodologies to improve the accuracy. However, since multiple measuring instruments are used in mass production, repeatability and reproducibility are as important as the accuracy of the values produced by predictive models. In this study, we adopted a data augmentation approach that minimizes the physical difference between multiple measuring instruments by using the domain knowledge of the spectroscopic ellipsometry (SE) field. By modeling the photodetector misalignment as polynomials and taking into account random noise, we proposed stochastic polynomial wavelength calibration (s-PWC) which can improve the percentage of the gage repeatability and reproducibility (Gage R&R) value. In experiments, the proposed methodology was applied to train the nanostructure prediction model of a three-dimensional vertical NAND Flash memories with industrial data sets. The performance improvements before and after applying the method were evaluated. Gaussian noise augmentation (GNA) and polynomial wavelength calibration (PWC) methodologies devised based on previous studies were also evaluated for relative comparison. As a result of conducting the experiments under conditions similar to the actual production environment, the average value of the percentage of Gage R&R decreased from 10.23% to 6.3% when applying the proposed method, while the GNA and PWC methodologies reduced the values to 10.01% and 7.62%, respectively. There were no significant changes in the values of coefficient of determination (R2) and root mean square error (RMSE) when applying the three methods based on the data augmentation approach. In other words, applying s-PWC ensures that the predictive model produces consistent values for the same sample when it needs to infer data obtained from multiple measuring instruments, while maintaining R2 and RMSE. Future research on data augmentation techniques by modeling differences between other physical components might extend the explanations of the methodologies to improve R2 and RMSE of predictive models. We expect this study could provide guidelines for improving the performance of inferential models based on machine learning and SE in mass production environments.

Optimum phase shift for quantitative phase microscopy in volume measurement.

Volume measurement of a phase object is one of the most distinctive capabilities of quantitative phase microscopy (QPM). However, the accuracy of a measured volume is limited by the different noises of a measurement system and the finite bandpass filter used in the phase extraction algorithm. In this paper, we analyze the inherent errors in volume measurement with QPM and propose the optimum condition that can minimize these errors. We find that phase information of a sample in the frequency domain nonlinearly oscillates as a function of the phase shift corresponding to the sample and its medium, and that the phase information of a sample inside the bandpass filter can be maximized by a proper phase shift. Through numerical simulations and actual experiments, we demonstrate that the error in phase volume measurement can be effectively reduced by the enhancement of the phase signal inside the bandpass region using an optimum amount of phase, which can be controlled by changing either the medium index or the wavelength of illumination.

Usefulness of Ki-67 (MIB-1) immunostaining in the diagnosis of pulmonary sclerosing hemangiomas.

Pulmonary sclerosing hemangioma (PSH) is an uncommon lung neoplasm with a clinical outcome that is generally benign. However, differentiating PSH from pulmonary carcinoma is sometimes difficult as both lesions share similar histopathologic and immunohistochemical features. In this study, we investigated the usefulness of Ki-67 (MIB-1) immunostaining in the diagnosis of PSH. We compared the staining pattern for Ki-67 (MIB-1) in 29 cases of typical PSH and 79 cases of pulmonary non-small cell carcinoma (NSCLC) using an immunohistochemical method on formalin-fixed paraffin-embedded tissues. In all studied PSH cases, we noted cell membrane and cytoplasmic staining for Ki-67 (MIB-1), but this was not observed in any of the NSCLC cases. The Ki-67 proliferation index was lower in PSH than in the NSCLC cases (mean, 1.1% vs mean, 5.5%; p < 0.001). These findings suggest that cell membrane and cytoplasmic staining for Ki-67 (MIB-1), as well as the Ki-67 proliferation index, may be useful for distinguishing PSH from pulmonary carcinoma.

Contrast-enhanced fusion of multi-sensor images using subband-decomposed multiscale retinex.

In this paper, we propose a novel pixel-level multi-sensor image fusion algorithm with simultaneous contrast enhancement. In order to accomplish both image fusion and contrast enhancement simultaneously, we suggest a modified framework of the subband-decomposed multiscale retinex (SDMSR), our previous contrast enhancement algorithm. This framework is based on a fusion strategy that reflects the multiscale characteristics of the SDMSR well. We first apply two complementary intensity transfer functions to source images in order to effectively utilize hidden information in both shadows and highlights in the fusion process. We then decompose retinex outputs into nearly nonoverlapping spectral subbands. The decomposed retinex outputs are then fused subband-by-subband, by using global weighting as well as local weighting to overcome the limitations of the pixel-based fusion approach. After the fusion process, we apply a space-varying subband gain to each fused subband-decomposed retinex output according to the subband characteristic so that the contrast of the fused image can be effectively enhanced. In addition, in order to effectively manage artifacts and noise, we make the degree of enhancement of fused details adjustable by improving a detail adjustment function. From experiments with various multi-sensor image pairs, the results clearly demonstrate that even if source images have poor contrast, the proposed algorithm makes it possible to generate a fused image with highly enhanced contrast while preserving visually salient information contained in the source images.

New classification of focal cortical dysplasia: application to practical diagnosis.

BACKGROUND AND PURPOSE: Malformation of cortical development (MCD) is a well-known cause of drug-resistant epilepsy and focal cortical dysplasia (FCD) is the most common neuropathological finding in surgical specimens from drug-resistant epilepsy patients. Palmini's classification proposed in 2004 is now widely used to categorize FCD. Recently, however, Blumcke et al. recommended a new system for classifying FCD in 2011. METHODS: We applied the new classification system in practical diagnosis of a sample of 117 patients who underwent neurosurgical operations due to drug-resistant epilepsy at Severance Hospital in Seoul, Korea. RESULTS: Among 117 cases, a total of 16 cases were shifted to other FCD subtypes under the new classification system. Five cases were reclassified to type IIIa and five cases were categorized as dual pathology. The other six cases were changed within the type I category. CONCLUSIONS: The most remarkable changes in the new classification system are the advent of dual pathology and FCD type III. Thus, it will be very important for pathologists and clinicians to discriminate between these new categories. More large-scale research needs to be conducted to elucidate the clinical influence of the alterations within the classification of type I disease. Although the new FCD classification system has several advantages compared to the former, the correlation with clinical characteristics is not yet clear.

Silver nanoparticle-induced degranulation observed with quantitative phase microscopy.

Monitoring a degranulation process in a live mast cell is a quite important issue in immunology and pharmacology. Because the size of a granule is normally much smaller than the resolution limit of an optical microscope system, there is no direct real-time live cell imaging technique for observing degranulation processes except for fluorescence imaging techniques. In this research, we propose optical quantitative phase microscopy (QPM) as a new observation tool to study degranulation processes in a live mast cell without any fluorescence labeling. We measure the cell volumes and the cross sectional profiles (x-z plane) of an RBL-2H3 cell and a HeLa cell, before and after they are exposed to calcium ionophore A23187 and silver nanoparticles (AgNPs). We verify that the volume and the cross sectional line profile of the RBL-2H3 cell were changed significantly when it was exposed to A23187. When 50 microg/mL of AgNP is used instead of A23187, the measurements of cell volume and cross sectional profiles indicate that RBL-2H3 cells also follow degranulation processes. Degranulation processes for these cells are verified by monitoring the increase of intracellular calcium ([Ca(2+)](i)) and histamine with fluorescent methods.