A pulse-coupled neural network (PCNN) is a promising image segmentation approach that requires no training. However, it is challenging to successfully apply a PCNN to medical image segmentation due to common but difficult scenarios such as irregular object shapes, blurred boundaries, and intensity inhomogeneity. To improve this situation, a novel framework incorporating fuzzy connectedness (FC) is proposed. First, a comparative study of the traditional PCNN models is carried out to analyze the framework and firing mechanism. Then, the characteristic matrix of fuzzy connectedness (CMFC) is presented for the first time. The CMFC can provide more intensity information and spatial relationships at the pixel level, which is helpful for producing a more reasonable firing mechanism in the PCNN models. Third, by integrating the CMFC into the PCNN framework models, a construction scheme of FC-PCNN models is designed. To illustrate this concept, a general solution that can be applied to different PCNN models is developed. Next, the segmentation performances of the proposed FC-PCNN models are evaluated by comparison with the traditional PCNN models, the traditional segmentation methods, and deep learning methods. The test images include synthetic and real medical images from the Internet and three public medical image datasets. The quantitative and visual comparative analysis demonstrates that the proposed FC-PCNN models outperform the traditional PCNN models and the traditional segmentation methods and achieve competitive performance to the deep learning methods. In addition, the proposed FC-PCNN models have favorable capability to eliminate inference from surrounding artifacts.

In this paper, we build four-part cone models to explore the coupling effect of seven cone fiber couplers. Moreover, this is the first study of the coupling effect of four layers of biological couplers in animals and other biological lives. We simulate the four layers cone couplers by using the beam propagation method, and we assume the input beam is located at the outer fiber of the central cone. Our simulation results showed that there are two wavelength regions (short and long wavelength regions) with the strongest coupling, where the most power of input optical powers of the central cones will transfer to the six surrounding cones after transmitting through the four layers of cone couplers. However, within a wavelength region of ±75 nm near to the peak wavelengths, located in the yellow–green wavelength range, the splitting ratios at the output of the outer segment of the central cone are always greater than the sum of the splitting ratios of the six surrounding cones. These cone couplers may play an important role in color preprocessing (e.g., doing opponent color processing partially). The cone fiber coupler effect and light absorption of cones are considered separately in our models. By taking account of both the cone fiber coupling effect and absorption of outer segment of L cone, we find the multiplication of the relative optical power of cone couplers, the spectral sensitivity data of the L cone, and a normalized coefficient that matches with the photopic luminous efficiency of the human eye well. This is the attempt to use both the cone fiber coupling effect and the absorption of L cones to explain the photopic luminous efficiency. The splitting ratios of the central cones are greater than 80% at peak wavelengths located in the yellow-green wavelength range, and this can help to explain why the human eye is more sensitive to green light.

As one of the main routes of vaccine production, cell suspension culture technology has become an inevitable trend in the development of biopharmaceutical industry. Its main advantage is that it can maximize the product quality while achieving high yield through accurate and effective process control means. As the basic material needed in the process of cell suspension culture, cell culture medium mainly provides energy through substances such as glucose. In order to obtain high-quality and high-performance cell products, it is necessary to continuously monitor the culture process. Traditional biochemical method can achieve the purpose of monitoring, but the detection cycle is long which making it hard to meet the needs of real-time monitoring. In contrast, Near-infrared (NIR) spectroscopy can detect glucose concentration in cell culture medium in real time without loss and it has the advantages of fast, accurate, and high sensitivity. In this study, a partial least squares regression (PLSR) model was established based on the near infrared spectrum of cell medium containing a certain amount of glucose to verify its validity and feasibility. Efforts have been made on the processing of sample outlier, spectrum preprocessing and wavelength optimization method. And good prediction results were obtained with the coefficient of determination (R2) of the model was 0.998, the cross-validation mean square error (RMSECV) was 0.0036 and the residual prediction deviation (RPD) value was 4.15. On this basis, in order to further improve the prediction accuracy of the model, this paper introduced glutamine variables to establish a multi-component fusion model and further optimized the combined wavelength selection method. The results showed that the model could achieve better prediction accuracy when using our proposed optimization method which indicated that the method of real-time detection of glucose concentration in cell medium based on near infrared spectroscopy was feasible.