Search

Purpose: In recent years, deep learning–based image processing has emerged as a valuable tool for medical imaging owing to its high performance. However, the quality of deep learning–based methods heavily relies on the amount of training data; the high cost of acquiring a large data set is a limitation to their utilization in medical fields. Herein, based on deep learning, we developed a computed tomography (CT) modality conversion method requiring only a few unsupervised images. Methods: The proposed method is based on cycle‐consistency generative adversarial network (CycleGAN) with several extensions tailored for CT images, which aims at preserving the structure in the processed images and reducing the amount of training data. This method was applied to realize the conversion of megavoltage computed tomography (MVCT) to kilovoltage computed tomography (kVCT) images. Training was conducted using several data sets acquired from patients with head and neck cancer. The size of the data sets ranged from 16 slices (two patients) to 2745 slices (137 patients) for MVCT and 2824 slices (98 patients) for kVCT. Results: The required size of the training data was found to be as small as a few hundred slices. By statistical and visual evaluations, the quality improvement and structure preservation of the MVCT images converted by the proposed model were investigated. As a clinical benefit, it was observed by medical doctors that the converted images enhanced the precision of contouring. Conclusions: We developed an MVCT to kVCT conversion model based on deep learning, which can be trained using only a few hundred unpaired images. The stability of the model against changes in data size was demonstrated. This study promotes the reliable use of deep learning in clinical medicine by partially answering commonly asked questions, such as "Is our data sufficient?" and "How much data should we acquire?" [ABSTRACT FROM AUTHOR]

Polyethylene glycol dehydrogenase (PEGDH) from Sphingopyxis terrae is composed of 535 amino acid residues and one flavin adenine dinucleotide per monomer protein in a homodimeric structure. Based on three-dimensional molecular modeling and kinetic characterization of wild-type PEGDH and mutant PEGDHs constructed by site-directed mutagenesis, residues potentially involved in catalysis and substrate binding were found in the vicinity of the flavin ring.

Japan is experiencing the best economic performance among the major industrial countries. The effects of the Gulf Crisis on Japan may be less severe than in the 1970s. After considering [...]

The sphingosine‐1‐phosphate (S1P) receptor modulator regulates lymphocyte trafficking, resulting in its depletion from circulation, which ultimately causes immunosuppression. In this study, we investigated the preventive effect of fingolimod (FTY720) in the non‐obese type 2 diabetic model, Spontaneously Diabetic Torii (SDT) rats. The S1P receptor modulator, FTY720 (0.3 mg/kg p.o.), was administered for 12 weeks to SDT rats from 5 to 17 weeks of age. Based on our findings, FTY720 could suppress the incidence of diabetes in SDT rats. Further, glucose intolerance was improved in FTY720‐treated SDT rats at 14 weeks of age. Based on the haematological and histological analyses performed at 17 to 18 weeks of age, a decrease in lymphocytes and monocytes in the peripheral blood and a decrease in lymphocyte and atrophy in spleen occurred in the FTY720‐treated SDT rats. Furthermore, the pancreatic changes, such as inflammation, atrophy, and fibrosis in islets observed in SDT rats were improved by FTY720 treatment. These findings suggest that the immunomodulatory effects of FTY720 reduced the pancreatic lesion in SDT rats, thereby demonstrating its preventive effect against diabetes. The development of diabetes in SDT rats is related to disorders of the immune system. However, the S1P receptor modulator may be useful for treating type 2 diabetes. [ABSTRACT FROM AUTHOR]

Since 2018, the Japanese Ministry of the Environment (MoE) has worked on the "Recycling Demonstration Project for the Soil Generated from Decontamination Activities" (the Project), which aims to offer empirical evidence for the final disposal of radiation‐contaminated substances. To offer practical suggestions for the Project, this study aimed to explore determinants of its support. More specifically, building on previous studies on trust, this study proposed a hypothetical model in which support for the Project was determined by trust in the MoE, which in turn was determined by salient value similarity, ability, and motivation. Additionally, it was hypothesised that the relationships between trust and its different determinants differed: Salient value similarity showed a stronger relationship with trust in those who lived near the place where the Project is taking place whereas ability and motivation showed stronger relationships with trust in those who did not. Questionnaire‐based research was conducted in both Iitate village (n = 106), Fukushima, where the Project had been taking place, and Tokyo (n = 163), located approximately 220 km from Fukushima. Multilevel structural equation modelling supported the hypothesis. These results suggest that emphasising the similarity of values between the MoE and the public would help obtain the public's support for the Project. [ABSTRACT FROM AUTHOR]

Abstract Background Marbling defined by the amount and distribution of intramuscular fat, so-called Shimofuri, is an economically important trait of beef cattle in Japan. Our previous study detected 3 single nucleotide polymorphisms (SNPs), g.231054C > T, g.3109537C > T and c.*188G > A, respectively, in the 5' flanking region of the titin (TTN), the 5' flanking region of the ribosomal protein L27a (RPL27A) and the 3' untranslated region of the akirin 2 genes (AKIRIN2), which have been considered as positional functional candidates for the genes responsible for marbling, and showed association of these SNPs with marbling in Japanese Black beef cattle. In the present study, we investigated the allele frequency distribution of the 3 SNPs among the 5 cattle breeds, Japanese Black, Japanese Brown, Japanese Shorthorn, Holstein and Brown Swiss breeds. Findings We genotyped the TTN g.231054C > T, RPL27A g.3109537C > T and AKIRIN2 c.*188G > A SNPs by polymerase chain reaction-restriction fragment length polymorphism method, using 101 sires and 1,705 paternal half sib progeny steers from 8 sires for Japanese Black, 86 sires and 27 paternal half sib progeny steers from 3 sires for Japanese Brown, 79 sires and 264 paternal half sib progeny steers from 14 sires for Japanese Shorthorn, 119 unrelated cows for Holstein, and 118 unrelated cows for Brown Swiss breeds. As compared to the frequencies of the g.231054C > T T, g.3109537C > T T and c.*188G > A A alleles, associated with high marbling, in Japanese Black breed that has been subjected to a strong selection for high marbling, those in the breeds, Japanese Shorthorn, Holstein and Brown Swiss breeds, that have not been selected for high marbling were null or lower. The Japanese Brown breed selected slightly for high marbling showed lower frequency than Japanese Black breed in the g.3109537C > T T allele, whereas no differences were detected between the 2 breeds in the frequencies of the g.231054C > T T and c.*188G > A A alleles. Conclusions Based on this finding, we hypothesized that the pressure of the strong selection for high marbling in Japanese Black breed has increased the frequencies of the T, T and A alleles at the TTN g.231054C > T, RPL27A g.3109537C > T and AKIRIN2 c.*188G > A SNPs, respectively. This study, together with the previous association studies, suggested that the 3 SNPs may be useful for effective marker-assisted selection to increase the levels of marbling.