Histopathological image analysis is an examination of tissue under a light microscope for cancerous disease diagnosis. Computer-assisted diagnosis (CAD) systems work well by diagnosing cancer from histopathology images. However, stain variability in histopathology images is inevitable due to the use of different staining processes, operator ability, and scanner specifications. These stain variations present in histopathology images affect the accuracy of the CAD systems. Various stain normalization techniques have been developed to cope with inter-variability issues, allowing standardizing the appearance of images. However, in stain normalization, these methods rely on the single reference image rather than incorporate color distributions of the entire dataset. In this paper, we design a novel machine learning-based model that takes advantage of whole dataset distributions as well as color statistics of a single target image instead of relying only on a single target image. The proposed deep model, called stain acclimation generative adversarial network (SA-GAN), consists of one generator and two discriminators. The generator maps the input images from the source domain to the target domain. Among discriminators, the first discriminator forces the generated images to maintain the color patterns as of target domain. While second discriminator forces the generated images to preserve the structure contents as of source domain. The proposed model is trained using a color attribute metric, extracted from a selected template image. Therefore, the designed model not only learns dataset-specific staining properties but also image-specific textural contents. Evaluated results on four different histopathology datasets show the efficacy of SA-GAN to acclimate stain contents and enhance the quality of normalization by obtaining the highest values of performance metrics. Additionally, the proposed method is also evaluated for multiclass cancer type classification task, showing a 6.9% improvement in accuracy on ICIAR 2018 hidden test data.

In this paper, we address the performance optimization of Holmium doped fiber amplifier (HDFA) for optical communications in 2–2.15 μm wavelength range based on a single in-band forward pump source. The performance of the HDFA is analyzed with the help of theoretical simulations by considering an optimized length of Holmium doped fiber (HDF), doping concentration of Ho³⁺, and pump power. The impact of signal wavelength and power on gain, amplified spontaneous emission (ASE) noise, and noise figure (NF) of the amplifier is investigated. Furthermore, we investigate the variations in the gain of the amplifier, its output power, and NF by varying the power and wavelength of the pump source. After optimizing the parameters of the amplifier, the peak gain observed is around 56.5 dB, the 3 dB saturated output power obtained is 33.3 dBm, and the output power is 3 W at signal wavelength of 2.0321 μm for HDF having an optimized length of 12 m and pump power of 3.5 W. Minimum NF of around 8.2 dB is observed at 2.0321 μm for signal power of −5 dBm. The impact of ion-ion interaction on the performance of HDFA is also investigated. A reduction of 24.2 dB and 0.051 W is observed in peak gain and output power of HDFA, respectively by considering the ion-ion interaction.

Broadband amplification in the O+E-band is very desirable nowadays as a way of coping with increasing bandwidth demands. The main issue with doped fiber amplifiers working in this band such as the bismuth-doped fiber amplifier is that they are costly and not widely available. Therefore, a wideband and flat-gain hybrid optical amplifier (HOA) covering the O+E-band based on a parallel combination of a praseodymium-doped fiber amplifier (PDFA) and a semiconductor optical amplifier (SOA) is proposed and demonstrated through numerical simulations. The praseodymium-doped fiber (PDF) core is pumped using a laser diode with a power of 500 mW that is centered at a wavelength of 1030 nm. The SOA is driven by an injection current of 60 mA. The performance of the HOA is analyzed by the optimization of various parameters such as the PDF length, Pr${}^{3+}$ concentration, pump wavelength, and injection current. A flat average gain of 24 dB with a flatness of 1 dB and an output power of 9.6 dBm is observed over a wavelength range of 1270–1450 nm. The noise figure (NF) varies from a minimum of 4 dB to a maximum of 5.9 dB for a signal power of 0 dBm. A gain reduction of around 4 dB is observed for an O-band signal at a wavelength of 1290 nm by considering the up-conversion effect. The transmission performance of the designed HOA as a pre-amplifier is evaluated based on the bit-error rate (BER) analysis for a coarse wavelength-division multiplexing (CWDM) system of eight on-off keying (OOK)-modulated channels, each having a data rate of 10 Gbps. An error-free transmission over 60 km of standard single-mode fiber (SMF) is achieved for different data rates of 5 Gbps, 7.5 Gbps, and 10 Gbps.