Identification of Skin Lesions by Snapshot Hyperspectral Imaging.

This study pioneers the application of artificial intelligence (AI) and hyperspectral imaging (HSI) in the diagnosis of skin cancer lesions, particularly focusing on Mycosis fungoides (MF) and its differentiation from psoriasis (PsO) and atopic dermatitis (AD). By utilizing a comprehensive dataset of 1659 skin images, including cases of MF, PsO, AD, and normal skin, a novel multi-frame AI algorithm was used for computer-aided diagnosis. The automatic segmentation and classification of skin lesions were further explored using advanced techniques, such as U-Net Attention models and XGBoost algorithms, transforming images from the color space to the spectral domain. The potential of AI and HSI in dermatological diagnostics was underscored, offering a noninvasive, efficient, and accurate alternative to traditional methods. The findings are particularly crucial for early-stage invasive lesion detection in MF, showcasing the model's robust performance in segmenting and classifying lesions and its superior predictive accuracy validated through k-fold cross-validation. The model attained its optimal performance with a k-fold cross-validation value of 7, achieving a sensitivity of 90.72%, a specificity of 96.76%, an F1-score of 90.08%, and an ROC-AUC of 0.9351. This study marks a substantial advancement in dermatological diagnostics, thereby contributing significantly to the early and precise identification of skin malignancies and inflammatory conditions.

Preparing Well for Esophageal Endoscopic Detection Using a Hybrid Model and Transfer Learning.

Early detection of esophageal cancer through endoscopic imaging is pivotal for effective treatment. However, the intricacies of endoscopic diagnosis, contingent on the physician's expertise, pose challenges. Esophageal cancer features often manifest ambiguously, leading to potential confusions with other inflammatory esophageal conditions, thereby complicating diagnostic accuracy. In recent times, computer-aided diagnosis has emerged as a promising solution in medical imaging, particularly within the domain of endoscopy. Nonetheless, contemporary AI-based diagnostic models heavily rely on voluminous data sources, limiting their applicability, especially in scenarios with scarce datasets. To address this limitation, our study introduces novel data training strategies based on transfer learning, tailored to optimize performance with limited data. Additionally, we propose a hybrid model integrating EfficientNet and Vision Transformer networks to enhance prediction accuracy. Conducting rigorous evaluations on a carefully curated dataset comprising 1002 endoscopic images (comprising 650 white-light images and 352 narrow-band images), our model achieved exceptional outcomes. Our combined model achieved an accuracy of 96.32%, precision of 96.44%, recall of 95.70%, and f1-score of 96.04%, surpassing state-of-the-art models and individual components, substantiating its potential for precise medical image classification. The AI-based medical image prediction platform presents several advantageous characteristics, encompassing superior prediction accuracy, a compact model size, and adaptability to low-data scenarios. This research heralds a significant stride in the advancement of computer-aided endoscopic imaging for improved esophageal cancer diagnosis.

Detection of Hydroxychloroquine Retinopathy via Hyperspectral and Deep Learning through Ophthalmoscope Images.

Hydroxychloroquine, also known as quinine, is primarily utilized to manage various autoimmune diseases, such as systemic lupus erythematosus, rheumatoid arthritis, and Sjogren's syndrome. However, this drug has side effects, including diarrhea, blurred vision, headache, skin itching, poor appetite, and gastrointestinal discomfort. Blurred vision is caused by irreversible retinal damages and can only be mitigated by reducing hydroxychloroquine dosage or discontinuing the drug under a physician's supervision. In this study, color fundus images were utilized to identify differences in lesions caused by hydroxychloroquine. A total of 176 color fundus images were captured from a cohort of 91 participants, comprising 25 patients diagnosed with hydroxychloroquine retinopathy and 66 individuals without any retinopathy. The mean age of the participants was 75.67 ± 7.76. Following the selection of a specific region of interest within each image, hyperspectral conversion technology was employed to obtain the spectrum of the sampled image. Spectral analysis was then conducted to discern differences between normal and hydroxychloroquine-induced lesions that are imperceptible to the human eye on the color fundus images. We implemented a deep learning model to detect lesions, leveraging four artificial neural networks (ResNet50, Inception_v3, GoogLeNet, and EfficientNet). The overall accuracy of ResNet50 reached 93% for the original images (ORIs) and 96% for the hyperspectral images (HSIs). The overall accuracy of Inception_v3 was 87% for ORIs and 91% for HSI, and that of GoogLeNet was 88% for ORIs and 91% for HSIs. Finally, EfficientNet achieved an overall accuracy of 94% for ORIs and 97% for HSIs.

Large area MoS2 thin film growth by direct sulfurization.

In this study, we present the growth of monolayer MoS2 (molybdenum disulfide) film. Mo (molybdenum) film was formed on a sapphire substrate through e-beam evaporation, and triangular MoS2 film was grown by direct sulfurization. First, the growth of MoS2 was observed under an optical microscope. The number of MoS2 layers was analyzed by Raman spectrum, atomic force microscope (AFM), and photoluminescence spectroscopy (PL) measurement. Different sapphire substrate regions have different growth conditions of MoS2. The growth of MoS2 is optimized by controlling the amount and location of precursors, adjusting the appropriate growing temperature and time, and establishing proper ventilation. Experimental results show the successful growth of a large-area single-layer MoS2 on a sapphire substrate through direct sulfurization under a suitable environment. The thickness of the MoS2 film determined by AFM measurement is about 0.73 nm. The peak difference between the Raman measurement shift of 386 and 405 cm-1 is 19.1 cm-1, and the peak of PL measurement is about 677 nm, which is converted into energy of 1.83 eV, which is the size of the direct energy gap of the MoS2 thin film. The results verify the distribution of the number of grown layers. Based on the observation of the optical microscope (OM) images, MoS2 continuously grows from a single layer of discretely distributed triangular single-crystal grains into a single-layer large-area MoS2 film. This work provides a reference for growing MoS2 in a large area. We expect to apply this structure to various heterojunctions, sensors, solar cells, and thin-film transistors.

Detection of weak micro-scratches on aspherical lenses using a Gabor neural network and transfer learning.

Surface defect detection is a crucial step in ensuring the quality of lenses. One method to check for surface defects is to use an optical system integrated with an industrial camera to magnify and highlight the position of a defect on the surface of a lens. Therefore, automatic optical inspection systems are applied to detect micro-defects. In this study, we propose an automatic inspection platform based on a deep neural network for automatically imaging and examining the surface of a lens. High-resolution images of 2448×2048 pixels are acquired using a hybrid lighting system. A convolutional neural network integrated with a trainable Gabor filter is used as a machine vision algorithm to perform image classification and defect segmentation tasks. The experimental results show that the proposed method effectively performed with noise in the background, achieving a segmentation accuracy of 98%.

Design of a Functional Eye Dressing for Treatment of the Vitreous Floater.

With the rapid development of display technology, related diseases of the human eye are also increasing day by day. Eye floaters are one of the diseases that affect humans. Herein, we present a functional ophthalmic dressing that can permeate the skin tissues of the eyes through oxygen and hydrogen to improve the symptoms of floaters. In clinical tests, the symptoms of sensory floaters improved in 28 patients, and the recovery rates of mild, moderate, and severe floaters were about 70%, 66.7%, and 83.3%, respectively.

Characteristics of P-Type and N-Type Photoelectrochemical Biosensors: A Case Study for Esophageal Cancer Detection.

P-type and N-type photoelectrochemical (PEC) biosensors were established in the laboratory to discuss the correlation between characteristic substances and photoactive material properties through the photogenerated charge carrier transport mechanism. Four types of human esophageal cancer cells (ECCs) were analyzed without requiring additional bias voltage. Photoelectrical characteristics were examined by scanning electron microscopy (SEM), X-ray diffraction (XRD), UV-vis reflectance spectroscopy, and photocurrent response analyses. Results showed that smaller photocurrent was measured in cases with advanced cancer stages. Glutathione (L-glutathione reduced, GSH) and Glutathione disulfide (GSSG) in cancer cells carry out redox reactions during carrier separation, which changes the photocurrent. The sensor can identify ECC stages with a certain level of photoelectrochemical response. The detection error can be optimized by adjusting the number of cells, and the detection time of about 5 min allowed repeated measurement.

Identification of Skin Lesions by Using Single-Step Multiframe Detector.

An artificial intelligence algorithm to detect mycosis fungoides (MF), psoriasis (PSO), and atopic dermatitis (AD) is demonstrated. Results showed that 10 s was consumed by the single shot multibox detector (SSD) model to analyze 292 test images, among which 273 images were correctly detected. Verification of ground truth samples of this research come from pathological tissue slices and OCT analysis. The SSD diagnosis accuracy rate was 93%. The sensitivity values of the SSD model in diagnosing the skin lesions according to the symptoms of PSO, AD, MF, and normal were 96%, 80%, 94%, and 95%, and the corresponding precision were 96%, 86%, 98%, and 90%. The highest sensitivity rate was found in MF probably because of the spread of cancer cells in the skin and relatively large lesions of MF. Many differences were found in the accuracy between AD and the other diseases. The collected AD images were all in the elbow or arm and other joints, the area with AD was small, and the features were not obvious. Hence, the proposed SSD could be used to identify the four diseases by using skin image detection, but the diagnosis of AD was relatively poor.

Intelligent Identification of MoS2 Nanostructures with Hyperspectral Imaging by 3D-CNN.

Increasing attention has been paid to two-dimensional (2D) materials because of their superior performance and wafer-level synthesis methods. However, the large-area characterization, precision, intelligent automation, and high-efficiency detection of nanostructures for 2D materials have not yet reached an industrial level. Therefore, we use big data analysis and deep learning methods to develop a set of visible-light hyperspectral imaging technologies successfully for the automatic identification of few-layers MoS2. For the classification algorithm, we propose deep neural network, one-dimensional (1D) convolutional neural network, and three-dimensional (3D) convolutional neural network (3D-CNN) models to explore the correlation between the accuracy of model recognition and the optical characteristics of few-layers MoS2. The experimental results show that the 3D-CNN has better generalization capability than other classification models, and this model is applicable to the feature input of the spatial and spectral domains. Such a difference consists in previous versions of the present study without specific substrate, and images of different dynamic ranges on a section of the sample may be administered via the automatic shutter aperture. Therefore, adjusting the imaging quality under the same color contrast conditions is unnecessary, and the process of the conventional image is not used to achieve the maximum field of view recognition range of ~1.92 mm2. The image resolution can reach ~100 nm and the detection time is 3 min per one image.

Hyperspectral Ophthalmoscope Images for the Diagnosis of Diabetic Retinopathy Stage.

A methodology that applies hyperspectral imaging (HSI) on ophthalmoscope images to identify diabetic retinopathy (DR) stage is demonstrated. First, an algorithm for HSI image analysis is applied to the average reflectance spectra of simulated arteries and veins in ophthalmoscope images. Second, the average simulated spectra are categorized by using a principal component analysis (PCA) score plot. Third, Beer-Lambert law is applied to calculate vessel oxygen saturation in the ophthalmoscope images, and oxygenation maps are obtained. The average reflectance spectra and PCA results indicate that average reflectance changes with the deterioration of DR. The G-channel gradually decreases because of vascular disease, whereas the R-channel gradually increases with oxygen saturation in the vessels. As DR deteriorates, the oxygen utilization of retinal tissues gradually decreases, and thus oxygen saturation in the veins gradually increases. The sensitivity of diagnosis is based on the severity of retinopathy due to diabetes. Normal, background DR (BDR), pre-proliferative DR (PPDR), and proliferative DR (PDR) are arranged in order of 90.00%, 81.13%, 87.75%, and 93.75%, respectively; the accuracy is 90%, 86%, 86%, 90%, respectively. The F1-scores are 90% (Normal), 83.49% (BDR), 86.86% (PPDR), and 91.83% (PDR), and the accuracy rates are 95%, 91.5%, 93.5%, and 96%, respectively.

Nanoparticle-Doped Polydimethylsiloxane Fluid Enhances the Optical Performance of AlGaN-Based Deep-Ultraviolet Light-Emitting Diodes.

This paper proposes a new encapsulation structure for aluminum nitride-based deep UV light-emitting diodes (DUV-LEDs) and eutectic flip chips containing polydimethylsiloxane (PDMS) fluid doped with SiO2 nanoparticles (NPs) with a UV-transparent quartz hemispherical glass cover. Experimental results reveal that the proposed encapsulation structure has considerably higher light output power than the traditional one. The light extraction efficiency was increased by 66.49% when the forward current of the DUV-LED was 200 mA. Doping the PDMS fluid with SiO2 NPs resulted in higher light output power than that of undoped fluid. The maximum efficiency was achieved at a doping concentration of 0.2 wt%. The optical output power at 200 mA forward current of the encapsulation structure with NP doping of the fluid was 15% higher than that without NP doping. The optical output power of the proposed encapsulation structure was 81.49% higher than that of the traditional encapsulation structure. The enhanced light output power was due to light scattering caused by the SiO2 NPs and the increased average refractive index. The encapsulation temperature can be reduced by 4 °C at a driving current of 200 mA by using the proposed encapsulation structure.