Early classification of brain tumors from magnetic resonance imaging (MRI) plays an important role in the diagnosis of such diseases. There are many diagnostic imaging methods used to identify tumors in the brain. MRI is commonly used for such tasks because of its unmatched image quality. The relevance of artificial intelligence (AI) in the form of deep learning (DL) has revolutionized new methods of automated medical image diagnosis. This study aimed to develop a robust and efficient method based on transfer learning technique for classifying brain tumors using MRI. In this article, the popular deep learning architectures are utilized to develop brain tumor diagnostic system. The pre-trained models such as Xception, NasNet Large, DenseNet121 and InceptionResNetV2 are used to extract the deep features from brain MRI. The experiment was performed using two benchmark datasets that are openly accessible from the web. Images from the dataset were first cropped, preprocessed, and augmented for accurate and fast training. Deep transfer learning models are trained and tested on a brain MRI dataset using three different optimization algorithms (ADAM, SGD, and RMSprop). The performance of the transfer learning models is evaluated using performance metrics such as accuracy, sensitivity, precision, specificity and F1-score. From the experimental results, our proposed CNN model based on the Xception architecture using ADAM optimizer is better than the other three proposed models. The Xception model achieved accuracy, sensitivity, precision specificity, and F1-score values of 99.67%, 99.68%, 99.68%, 99.66%, and 99.68% on the MRI-large dataset, and 91.94%, 96.55%, 87.50%, 87.88%, and 91.80% on the MRI-small dataset, respectively. The proposed method is superior to the existing literature, indicating that it can be used to quickly and accurately classify brain tumors.

We propose and experimentally demonstrate a compact bending sensor. The head of the sensor is only 0.8 mm in length, and consists of an abrupt biconical fiber taper formed using a conventional fusion splicer, in which a fiber Bragg grating (FBG) is inscribed using a femtosecond laser. The biconical taper incorporating the FBG can couple light from the cladding to the backward-propagating core mode, which realizes an interferometer in reflection-mode. Bending of the structure can be detected from the contrast change of interference fringes. A configuration to measure curvature is investigated to demonstrate the sensing characteristics. The temperature cross-sensitivity of the sensor is studied, and the results demonstrate that it is insensitive to temperature.

We report the observation of memorized polarization-dependent light scattering in a ${\mathrm{Eu}}^{\mathrm{2+}}$ -doped fluoroaluminate glass sample. Anisotropic light scattering along the plane of the light polarization was observed in the glass sample after the excitation of a focused 800 nm, 150 fs laser beam at a repetition rate of 200 kHz. When we changed the direction of the light polarization and irradiated the same location, we observed an anisotropic light-scattering pattern identical to the original one at the beginning, but then observed a new pattern along the new direction of the light polarization while the original light-scattering pattern disappeared gradually with the passage of time. This observed phenomenon was considered to be due to the light scattering of the polarization-dependent permanent microstructure induced by the polarized ultrashort pulsed laser itself.

Reduced graphene oxide (rGO) decorated with silver nanoparticles (Ag NPs) is synthesized by femtosecond laser ablation in solution method. The nonlinear optical properties of both rGO and Ag NPs/rGO are measured using a femtosecond laser Z-scan technique. The results reveal that both the nonlinear absorption and nonlinear refraction in the hybrid are enhanced due to the interaction of the energy state of Ag NPs. The composite shows a saturation intensity 18.5 MW/cm2 and boosts the nonlinear refraction as large as 2~3 times of that of rGO reaching to −1.1 × 10⁻¹² m²/W. The enhancement of the saturable absorption might be caused by the further bleaching the valence band of rGO due to the transfer of the light excited carriers from graphene to the metal state of the NPs. The slow relaxation of the excited carriers to the ground state of rGO will also cause the increase of the carrier density and thereby result in the enhancement of the nonlinear refractive index of the material.

Supercontinuum generation and its application have become one of the attractive research fields in ultrafast optics. We have acquired a series of narrow bandwidth and symmetric gated spectra continuously from the chirped supercontinuum generated in a sapphire plate with a femtosecond laser, using an ultrafast optical Kerr gate (OKG) with ${\mathrm{Bi}}_2{\mathrm{O}}_3–{\mathrm{B}}_2{\mathrm{O}}_3–\mathrm{Si}{\mathrm{O}}_2$ oxide glass as the Kerr material. Experimental results show that the gated spectra obtained using OKG of the glass have more superiorities comparing with $\mathrm{C}{\mathrm{S}}_2$ .

We developed a new method to fabricate tilted fiber Bragg gratings (TFBGs) by using a femtosecond laser with a phase mask. During the laser processing, the fiber was obliquely moved at a tilt angle, in which the laser beam and the phase mask were fixed. The peak loss of the cladding modes with a tilt angle of 4.9° reaches to ~-8 dB, and the insertion loss is less than −0.2 dB. The TFBG was stable at temperature up to 700°C and slight degraded at 800 °C. The temperature sensing of the TFBG was demonstrated at a high temperature up to 800°C. The temperature sensitivities of the Bragg mode, the ghost mode, and the cladding mode were measured to be 15.72 pm/°C, 15.56 pm/°C, and 15.52 pm/°C, respectively. The refractive index response of the TFBGs was also measured.