Deep learning methods for medical image fusion: A review.

The image fusion methods based on deep learning has become a research hotspot in the field of computer vision in recent years. This paper reviews these methods from five aspects: Firstly, the principle and advantages of image fusion methods based on deep learning are expounded; Secondly, the image fusion methods are summarized in two aspects: End-to-End and Non-End-to-End, according to the different tasks of deep learning in the feature processing stage, the non-end-to-end image fusion methods are divided into two categories: deep learning for decision mapping and deep learning for feature extraction. According to the different types of the networks, the end-to-end image fusion methods are divided into three categories: image fusion methods based on Convolutional Neural Network, Generative Adversarial Network, and Encoder-Decoder Network; Thirdly, the application of the image fusion methods based on deep learning in medical image field is summarized from two aspects: method and data set; Fourthly, evaluation metrics commonly used in the field of medical image fusion are sorted out from 14 aspects; Fifthly, the main challenges faced by the medical image fusion are discussed from two aspects: data sets and fusion methods. And the future development direction is prospected. This paper systematically summarizes the image fusion methods based on the deep learning, which has a positive guiding significance for the in-depth study of multi modal medical images.

What caused the formation of the absorption maximum at 421 nm in vivo spectra of Rhodopseudomonas palustris.

A spectral peak at ~421 nm appeared in vivo spectrum of Rhodopseudomonas palustris CQV97 cultured in acetate-glutamate medium (M1) but not in acetate-ammonium sulfate medium (M2). However, the spectral origin of 421 nm peak was not clear and frequently attributed to carotenoid component(s). In this study, comparative analysis of the extracted components showed that magnesium protoporphyrin IX monomethylester (MPE) was accumulated as one of the predominate components in M1 culture. The amounts of bacteriochlorophyll a in M1 culture were higher than that in M2, whereas the amounts of carotenoids were nearly identical in both cultures. A simple, rapid and minimum interference with carotenoid and bacteriochlorophyll method to efficiently extract the compounds involving in the formation of 421 nm peak was developed in this study. Assembly of purified MPE with protein components from R. palustris in vitro demonstrated that MPE caused the formation of 421 nm peak. The localization analysis in vivo demonstrated it is MPE associating to protein components and accounting for the peak at ~421 nm. This work clarified the 421 nm peak in vivo mainly originated from MPE accumulation, and will be very helpful to further explore the physiological roles of MPE or its derivatives in photosynthesis.

[Fingerprinting analysis of photopigments in purple bacteria].

OBJECTIVE: Photopigments, including carotenoid and bacteriochlorophyll a, are the most important functional units of photosynthesis in purple bacteria. We developed rapid qualitative and quantitative methods to determine photopigments. METHODS: Using Rhodopesudomonas palustris CQV97 as a reference, we used image gray intensity analysis, absorption spectrophotometry, thin layer chromatography (TLC), HPLC and mass spectrometry (MS) for photopigment analysis. RESULTS: The total amount of photopigments increased by 13.5% by using modified acetone-methanol extraction. We developed two types of photopigment fingerprintings by TLC and HPLC, estimated the apparent relative content of each photopigment of fingerprintings, and determined the corresponding relationships between R(f) value of each photopigment on TLC fingerprinting and retention time of each photopigment elution in HPLC fingerprinting. Based on the data from the absorption spectra, MS and related photopigment biosynthetic pathway analysis, we identified 11 photopigments in CQV97 strain. Using this strain as a standard, we analyzed photopigments of the tested samples by TLC or HPLC. It was shown that (1) the relative standard deviation (RSD) of the two methods was less than 5%; (2) the compositions and contents of the theory sample were consistent with that of the standard sample; (3) the photopigment compositions of the real sample was the same as the standard sample, but the photopigment content was different. CONCLUSION: Both of TLC and HPLC analyses for photopigment determination have good stability and repeatability. The fingerprintings analyses are suitable for rapid determination of photopigments of purple bacteria and have important application in control of regulation mechanism for photopigment synthesis.