Generating diagnostic report for medical image by high-middle-level visual information incorporation on double deep learning models.

BACKGROUND AND OBJECTIVES: Writing diagnostic reports for medical images is a heavy and tedious work. The automatic generation of medical image diagnostic reports can assist doctors to reduce their workload and improve diagnosis efficiency. It is of great significance to introduce image caption algorithm into medical image processing. Existing approaches attempt to generate medical image diagnostic reports using image caption algorithms but without taking the accuracy of pathological information in generated diagnostic reports into account. METHODS: To solve the mentioned problem, we propose a Semantic Fusion Network (SFNet) including a lesion area detection model and a diagnostic generation model. The lesion area detection model can extract visual and pathological information from medical image, and the diagnostic report generation model can learn to fuse the two kinds of information to generate reports. Thus, the pathological information in the generated diagnostic reports can be more accurate. RESULTS: Experimental results have verified the performance of our model (Accuracy increases 1.2% on the Ultrasound Image Dataset and 2.4% on the Open-i X-ray Image Dataset), compared with the model only using visual feature to generate diagnostic reports. CONCLUSIONS: This work utilizes computer algorithms to generate the more accurate diagnostic reports for medical images automatically, which expands the application of computer-aided diagnosis and promotes the implementation of deep learning in the medical image analysis field.

Candidate chemosensory genes identified from the greater wax moth, Galleria mellonella, through a transcriptomic analysis.

The greater wax moth, Galleria mellonella Linnaeus (Lepidoptera: Galleriinae), is a ubiquitous pest of the honeybee, and poses a serious threat to the global honeybee industry. G. mellonella pheromone system is unusual compared to other lepidopterans and provides a unique olfactory model for pheromone perception. To better understand the olfactory mechanisms in G. mellonella, we conducted a transcriptomic analysis on the antennae of both male and female adults of G. mellonella using high-throughput sequencing and annotated gene families potentially involved in chemoreception. We annotated 46 unigenes coding for odorant receptors, 25 for ionotropic receptors, two for sensory neuron membrane proteins, 22 for odorant binding proteins and 20 for chemosensory proteins. Expressed primarily in antennae were all the 46 odorant receptor unigenes, nine of the 14 ionotropic receptor unigenes, and two of the 22 unigenes coding for odorant binding proteins, suggesting their putative roles in olfaction. The expression of some of the identified unigenes were sex-specific, suggesting that they may have important functions in the reproductive behavior of the insect. Identification of the candidate unigenes and initial analyses on their expression profiles should facilitate functional studies in the future on chemoreception mechanisms in this species and related lepidopteran moths.

Prevalence and genome characteristics of canine astrovirus in southwest China.

The aim of this study was to investigate canine astrovirus (CaAstV) infection in southwest China. We collected 107 faecal samples from domestic dogs with obvious diarrhoea. Forty-two diarrhoeic samples (39.3 %) were positive for CaAstV by RT-PCR, and 41/42 samples showed co-infection with canine coronavirus (CCoV), canine parvovirus-2 (CPV-2) and canine distemper virus (CDV). Phylogenetic analysis based on 26 CaAstV partial ORF1a and ORF1b sequences revealed that most CaAstV strains showed unique evolutionary features. Interestingly, putative recombination events were observed among four of the five complete ORF2 sequences cloned in this study, and three of the five complete ORF2 sequences formed a single unique group, suggesting that these strains could be a novel genotype. We successfully sequenced the complete genome of one CaAstV strain (designated 2017/44/CHN), which was 6628 nt in length. The features of this genome include putative recombination events in the ORF1a, ORF1b and ORF2 genes, while the ORF2 gene had a continuous insertion of 7 aa in region II compared with the other complete ORF2 sequences available in GenBank. Phylogenetic analysis showed that 2017/44/CHN formed a single group based on genome sequences, suggesting that this strain might be a novel genotype. The results of this study revealed that CaAstV circulates widely in diarrhoeic dogs in southwest China and exhibits unique evolutionary events. To the best of our knowledge, this is the first report of recombination events in CaAstV, and it contributes to further understanding of the genetic evolution of CaAstV.

Design, synthesis and biological evaluation of 1H-pyrrolo[2,3-b]pyridine and 1H-pyrazolo[3,4-b]pyridine derivatives as c-Met inhibitors.

Five novel 1H-pyrrolo[2,3-b]pyridine or 1H-pyrazolo[3,4-b]pyridine derivatives, with a methylene, sulfur, sulfoxide or cyclopropyl group as a linker, were designed, synthesized and biologically evaluated against c-Met and ALK. The development of these methods of compound synthesis may provide an important reference for the construction of novel 7-azaindole and 7-azaindazole derivatives with a single atom linker. The enzyme assay and cell assay in vitro showed that compound 9 displayed strong c-Met kinase inhibition with IC50 of 22.8nM, moderate ALK kinase inhibition, and strong cell inhibition with MKN-45 IC50 of 329nM and EBC-1 IC50 of 479nM. In order to find the better candidate compounds, compounds 8, 9 and 10 have been selected as tool compounds for further optimization.

Synthesis and anticancer activities of novel 1,2,4-triazolo[3,4-a]phthalazine derivatives.

Trying to develop potent and selective anticancer agents, two series of novel 1,2,4-triazolo[3,4-a]phthalazine derivatives were designed and synthesized. Their antitumor activities were evaluated by MTT method against four selected human cancer cell lines (MGC-803, EC-9706, HeLa and MCF-7). Our results showed that compound 11h exhibited good anticancer activities compared to 5-fluorouracil against the four tested cell lines, with IC50 values ranging from 2.0 to 4.5 µM. Flow cytometry analysis indicated that compound 11h induced the cellular early apoptosis and cell cycle arrest at G2/M phase in EC-9706.