In current breast ultrasound Computer Aided Diagnosis systems, the radiologist preselects a region of interest (ROI) as an input for computerized breast ultrasound image analysis. This task is time consuming and there is inconsistency among human experts. Researchers attempting to automate the process of obtaining the ROIs have been relying on image processing and conventional machine learning methods. We propose the use of a deep learning method for breast ultrasound ROI detection and lesion localisation. We use the most accurate object detection deep learning framework -Faster-RCNN with Inception-ResNet-v2 -as our deep learning network. Due to the lack of datasets, we use transfer learning and propose a new 3-channel artificial RGB method to improve the overall performance. We evaluate and compare the performance of our proposed methods on two datasets (namely, Dataset A and Dataset B), i.e. within individual datasets and composite dataset. We report the lesion detection results with two types of analysis: 1) detected point (centre of the segmented region or the detected bounding box) and 2) Intersection over Union (IoU ). Our results demonstrate that the proposed methods achieved comparable results on detected point but with notable improvement on IoU. In addition, our proposed 3-channel artificial RGB method improves the recall of Dataset A. Finally, we outline some
The sequence of the genome of a Rupestris stem pitting-associated virus (RSPaV) isolated from a declining Syrah grapevine in California, designated the Syrah strain (RSPaV-SY) was determined. The genome of this strain had an overall nucleotide identity of 77% in comparison with RSPaV sequences in GenBank; the coat protein was the most conserved gene among RSPaV sequences and the replicase was the least conserved gene. Phylogenetic analysis of partial coat protein and replicase gene sequences showed RSPaV-SY clustered independently from the majority of RSPaV isolates.
A single real-time multiplex reverse transcription quantitative polymerase chain reaction (RT-qPCR) assay for the simultaneous detection of Citrus tristeza virus (CTV), Citrus psorosis virus (CPsV), and Citrus leaf blotch virus (CLBV) was developed and validated using three different fluorescently labeled minor groove binding qPCR probes. To increase the detection reliability, coat protein (CP) genes from large number of different isolates of CTV, CPsV and CLBV were sequenced and a multiple sequence alignment was generated with corresponding CP sequences from the GenBank and a robust multiplex RT-qPCR assay was designed. The capacity of the multiplex RT-qPCR assay in detecting the viruses was compared to singleplex RT-qPCR designed specifically for each virus and was assessed using multiple virus isolates from diverse geographical regions and citrus species as well as graft-inoculated citrus plants infected with various combination of the three viruses. No significant difference in detection limits was found and specificity was not affected by the inclusion of the three assays in a multiplex RT-qPCR reaction. Comparison of the viral load for each virus using singleplex and multiplex RT-qPCR assays, revealed no significant differences between the two assays in virus detection. No significant difference in Cq values was detected when using one-step and two-step multiplex RT-qPCR detection formats. Optimizing the RNA extraction technique for citrus tissues and testing the quality of the extracted RNA using RT-qPCR targeting the cytochrome oxidase citrus gene as an RNA specific internal control proved to generate better diagnostic assays. Results showed that the developed multiplex RT-qPCR can streamline viruses testing of citrus nursery stock by replacing three separate singleplex assays, thus reducing time and labor while retaining the same sensitivity and specificity. The three targeted RNA viruses are regulated pathogens for California's mandatory "Section 3701: Citrus Nursery Stock Pest Cleanliness Program". Adopting a compatible multiplex RT-qPCR testing protocol for these viruses as well as other RNA and DNA regulated pathogens will provide a valuable alternative tool for virus detection and efficient program implementation.
Citrus yellow-vein disease (CYVD) was first reported in California in 1957. We now report that CYVD is associated with a virus-like agent, provisionally named citrus yellow-vein associated virus (CYVaV). The CYVaV RNA genome has 2,692 nucleotides and codes for two discernable open reading frames (ORFs). ORF1 encodes a protein of 190 amino acid (aa) whereas ORF2 is presumably generated by a −1 ribosomal frameshifting event just upstream of the ORF1 termination signal. The frameshift product (717 aa) encodes the RNA-dependent RNA polymerase (RdRp). Phylogenetic analyses suggest that CYVaV is closely related to unclassified virus-like RNAs in the family Tombusviridae. Bio-indexing and RNA-seq experiments indicate that CYVaV can induce yellow vein symptoms independently of known citrus viruses or viroids.
Multistage processing of automated breast ultrasound lesions recognition is dependent on the performance of prior stages. To improve the current state of the art, we propose the use of end-to-end deep learning approaches using fully convolutional networks (FCNs), namely FCN-AlexNet, FCN-32s, FCN-16s, and FCN-8s for semantic segmentation of breast lesions. We use pretrained models based on ImageNet and transfer learning to overcome the issue of data deficiency. We evaluate our results on two datasets, which consist of a total of 113 malignant and 356 benign lesions. To assess the performance, we conduct fivefold cross validation using the following split: 70% for training data, 10% for validation data, and 20% testing data. The results showed that our proposed method performed better on benign lesions, with a top "mean Dice" score of 0.7626 with FCN-16s, when compared with the malignant lesions with a top mean Dice score of 0.5484 with FCN-8s. When considering the number of images with Dice score >0.5, 89.6% of the benign lesions were successfully segmented and correctly recognised, whereas 60.6% of the malignant lesions were successfully segmented and correctly recognized. We conclude the paper by addressing the future challenges of the work.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.