Raman spectroscopy with an improved support vector machine for discrimination of thyroid and parathyroid tissues.

The objective of this study was to discriminate thyroid and parathyroid tissues using Raman spectroscopy combined with an improved support vector machine (SVM) algorithm. In thyroid surgery, there is a risk of inadvertently removing the parathyroid glands. At present, there is a lack of research on using Raman spectroscopy to discriminate parathyroid and thyroid tissues. In this article, samples were obtained from 43 individuals with thyroid and parathyroid tissues for Raman spectroscopy analysis. This study employed partial least squares (PLS) to reduce dimensions of data, and three optimization algorithms are used to improve the classification accuracy of SVM algorithm model in spectral analysis. The results show that PLS-GA-SVM algorithm has higher diagnostic accuracy and better reliability. The sensitivity of this algorithm is 94.67% and the accuracy is 94.44%. It can be concluded that Raman spectroscopy combined with the PLS-GA-SVM diagnostic algorithm has significant potential for discriminating thyroid and parathyroid tissues.

Single Model Deep Learning on Imbalanced Small Datasets for Skin Lesion Classification.

Deep convolutional neural network (DCNN) models have been widely explored for skin disease diagnosis and some of them have achieved the diagnostic outcomes comparable or even superior to those of dermatologists. However, broad implementation of DCNN in skin disease detection is hindered by small size and data imbalance of the publically accessible skin lesion datasets. This paper proposes a novel single-model based strategy for classification of skin lesions on small and imbalanced datasets. First, various DCNNs are trained on different small and imbalanced datasets to verify that the models with moderate complexity outperform the larger models. Second, regularization DropOut and DropBlock are added to reduce overfitting and a Modified RandAugment augmentation strategy is proposed to deal with the defects of sample underrepresentation in the small dataset. Finally, a novel Multi-Weighted New Loss (MWNL) function and an end-to-end cumulative learning strategy (CLS) are introduced to overcome the challenge of uneven sample size and classification difficulty and to reduce the impact of abnormal samples on training. By combining Modified RandAugment, MWNL and CLS, our single DCNN model method achieved the classification accuracy comparable or superior to those of multiple ensembling models on different dermoscopic image datasets. Our study shows that this method is able to achieve a high classification performance at a low cost of computational resources and inference time, potentially suitable to implement in mobile devices for automated screening of skin lesions and many other malignancies in low resource settings.

Sensitivity Enhancement of Two-Dimensional Materials Based on Genetic Optimization in Surface Plasmon Resonance.

Sensitivity is an important performance index for evaluating surface plasmon resonance (SPR) biosensors. Sensitivity enhancement has always been a hot topic. It is found that the different refractive indices of samples require different combinations of prism and metal film for better sensitivity. Furthermore, the sensitivity can be enhanced by coating two-dimensional (2D) materials with appropriate layers on the metal film. At this time, it is necessary to choose the best film configuration to enhance sensitivity. With the emergence of more and more 2D materials, selecting the best configuration manually is becoming more complicated. Compared with the traditional manual method of selecting materials and layers, this paper proposes an optimization method based on a genetic algorithm to quickly and effectively find the optimal film configuration that enhances sensitivity. By using this method, not only can the optimal number of layers of 2D materials be determined quickly, but also the optimal configuration can be conveniently found when many materials are available. The maximum sensitivity can reach 400°/RIU after optimization. The method provided application value for the relevant researchers seeking to enhance sensitivity.

Astigmatism-free Czerny-Turner compact spectrometer with cylindrical mirrors.

A modified optical design for a broadband, high resolution, astigmatism-free Czerny-Turner spectrometer is proposed. Astigmatism is corrected by using cylindrical mirrors over a broad spectral range. The theory and method for astigmatism correction are thoroughly analyzed. The comparison between the modified Czerny-Turner spectrometer and the traditional Czerny-Turner spectrometer is also described in detail. The ray-tracing results show that the RMS spot radius has decreased to 4.2 µm at the central wavelength and 17 µm at the wedge wavelength.

Elimination of error induced by a beam splitter substrate for a dispersion-compensated polarization Sagnac interferometer.

A wire grating beam splitter (WGBS) substrate in a dispersion-compensated polarization Sagnac interferometer (DCPSI) may introduce an additional shear distance in the shear distance generated by the DCPSI, thereby causing poor adaptability of the DCPSI to white light. This work applies a compensation scheme of an optical flat with the same material and thickness as the WGBS and parallel to the WGBS introduced in the other arm of the DCPSI. Theoretically, this method can decrease the additional shear distance approaching 0. The ideal shear distance in the simulation experiment is 5.86 mm, and the shear distance before and after compensation is 5.40 and 5.86 mm, respectively. The theoretical value of the additional shear distance in this experiment is -0.6625 mm, and the average compensation value is 0.66 mm. Overall, experiment and simulation results indicate that the above method can effectively eliminate the additional shear distance.