Showing papers by "Yi-Ping Phoebe Chen published in 2015"

Image based computer aided diagnosis system for cancer detection

Abstract: Studies and compare the recent works in different types of cancer detection.Low level features comparison for detecting different cancer types.Compare image modalities and associated segmentation algorithms.Research extension discussion in intermediate feature analysis and cloud structure for cancer detection. Cancer is one of the major causes of non-accidental death in human. Early diagnosis of the disease allows clinician to administer suitable treatment, and can improve the patient's survival rate. Traditional diagnosis involves trained clinicians to visually examine the respective medical images for any signs of nodule development in the body. However due to the large scale of the medical image data, this manual diagnosis is often laborious and can be highly subjective due to inter-observer variability. Inspired by the advanced computing technology which is capable of performing complex image processing and machine learning, researches had been carried out in the past few decades to develop computer aided diagnosis (CAD) systems to assist clinicians detecting different forms of cancer. This paper reviews computer vision techniques adopted in medical image analysis, in particular, for cancer detection. The review focused on the detection of the most common form of cancer types, namely breast cancer, prostate cancer, lung cancer and skin cancer. A recent proposed cloud computing frame work has inspired the researchers to utilize the existing works on image based cancer research and develop a more versatile CAD system for detection.

A catalogue of novel bovine long noncoding RNA across 18 tissues.

Abstract: Long non-coding RNA (lncRNA) have been implicated in diverse biological roles including gene regulation and genomic imprinting Identifying lncRNA in bovine across many differing tissue would contribute to the current repertoire of bovine lncRNA, and help further improve our understanding of the evolutionary importance and constraints of these transcripts Additionally, it could aid in identifying sites in the genome outside of protein coding genes where mutations could contribute to variation in complex traits This is particularly important in bovine as genomic predictions are increasingly used in genetic improvement for milk and meat production Our aim was to identify and annotate novel long non coding RNA transcripts in the bovine genome captured from RNA Sequencing (RNA-Seq) data across 18 tissues, sampled in triplicate from a single cow To address the main challenge in identifying lncRNA, namely distinguishing lncRNA transcripts from unannotated genes and protein coding genes, a lncRNA identification pipeline with a number of filtering steps was developed A total of 9,778 transcripts passed the filtering pipeline The bovine lncRNA catalogue includes MALAT1 and HOTAIR, both of which have been well described in human and mouse genomes We attempted to validate the lncRNA in libraries from three additional cows 726 (8747%) liver and 1,668 (5527%) blood class 3 lncRNA were validated with stranded liver and blood libraries respectively Additionally, this study identified a large number of novel unknown transcripts in the bovine genome with high protein coding potential, illustrating a clear need for better annotations of protein coding genes

Data mining in lung cancer pathologic staging diagnosis

Abstract: We utilised data mining techniques in cancer staging diagnosis.We found the correlation between pathology report and clinical information.Many interesting rules have been generated and evaluated.The evaluation results demonstrated the availability of the proposed framework. Lung cancer is one of the leading cancers for both genders all over the world. It is the most common cause of cancer death and almost reaches 20% of the total. The incidence of lung cancer has significantly increased from the early 19th century. In this manuscript we have discussed various data mining techniques that have been utilised for cancer diagnosis. The lung cancer pathologic staging is set based on the pathology report to describe the size and/or the extent of the original tumour and whether the cancer has spread (metastasis). Being aware of the lung cancer pathologic staging is important because it can be used to estimate a patient's prognosis and also can help physicians plan a suitable treatment. A sample of tissue from the patient's lung is required in order to complete the pathology report for the lung cancer pathologic staging diagnosis. In this procedure, a surgery biopsy is necessary but it may put the patient's health in jeopardy. Therefore, this study focuses on taking the clinical information which can be obtained without surgery to replace the pathology report. The data mining techniques are used to find the correlation between the clinical information and the pathology report in order to support lung cancer pathologic staging diagnosis.

A computationally efficient algorithm for genomic prediction using a Bayesian model

Abstract: Genomic prediction of breeding values from dense single nucleotide polymorphisms (SNP) genotypes is used for livestock and crop breeding, and can also be used to predict disease risk in humans. For some traits, the most accurate genomic predictions are achieved with non-linear estimates of SNP effects from Bayesian methods that treat SNP effects as random effects from a heavy tailed prior distribution. These Bayesian methods are usually implemented via Markov chain Monte Carlo (MCMC) schemes to sample from the posterior distribution of SNP effects, which is computationally expensive. Our aim was to develop an efficient expectation–maximisation algorithm (emBayesR) that gives similar estimates of SNP effects and accuracies of genomic prediction than the MCMC implementation of BayesR (a Bayesian method for genomic prediction), but with greatly reduced computation time. emBayesR is an approximate EM algorithm that retains the BayesR model assumption with SNP effects sampled from a mixture of normal distributions with increasing variance. emBayesR differs from other proposed non-MCMC implementations of Bayesian methods for genomic prediction in that it estimates the effect of each SNP while allowing for the error associated with estimation of all other SNP effects. emBayesR was compared to BayesR using simulated data, and real dairy cattle data with 632 003 SNPs genotyped, to determine if the MCMC and the expectation-maximisation approaches give similar accuracies of genomic prediction. We were able to demonstrate that allowing for the error associated with estimation of other SNP effects when estimating the effect of each SNP in emBayesR improved the accuracy of genomic prediction over emBayesR without including this error correction, with both simulated and real data. When averaged over nine dairy traits, the accuracy of genomic prediction with emBayesR was only 0.5% lower than that from BayesR. However, emBayesR reduced computing time up to 8-fold compared to BayesR. The emBayesR algorithm described here achieved similar accuracies of genomic prediction to BayesR for a range of simulated and real 630 K dairy SNP data. emBayesR needs less computing time than BayesR, which will allow it to be applied to larger datasets.

Bioinformatics in protein kinases regulatory network and drug discovery

Abstract: Protein kinases have been implicated in a number of diseases, where kinases participate many aspects that control cell growth, movement and death. The deregulated kinase activities and the knowledge of these disorders are of great clinical interest of drug discovery. The most critical issue is the development of safe and efficient disease diagnosis and treatment for less cost and in less time. It is critical to develop innovative approaches that aim at the root cause of a disease, not just its symptoms. Bioinformatics including genetic, genomic, mathematics and computational technologies, has become the most promising option for effective drug discovery, and has showed its potential in early stage of drug-target identification and target validation. It is essential that these aspects are understood and integrated into new methods used in drug discovery for diseases arisen from deregulated kinase activity. This article reviews bioinformatics techniques for protein kinase data management and analysis, kinase pathways and drug targets and describes their potential application in pharma ceutical industry.

Pairwise RNA secondary structure alignment with conserved stem pattern

Abstract: Motivation The regulatory functions performed by non-coding RNAs are related to their 3D structures, which are, in turn, determined by their secondary structures. Pairwise secondary structure alignment gives insight into the functional similarity between a pair of RNA sequences. Numerous exact or heuristic approaches have been proposed for computational alignment. However, the alignment becomes intractable when arbitrary pseudoknots are allowed. Also, since non-coding RNAs are, in general, more conserved in structures than sequences, it is more effective to perform alignment based on the common structural motifs discovered. Results We devised a method to approximate the true conserved stem pattern for a secondary structure pair, and constructed the alignment from it. Experimental results suggest that our method identified similar RNA secondary structures better than the existing tools, especially for large structures. It also successfully indicated the conservation of some pseudoknot features with biological significance. More importantly, even for large structures with arbitrary pseudoknots, the alignment can usually be obtained efficiently. Availability and implementation Our algorithm has been implemented in a tool called PSMAlign. The source code of PSMAlign is freely available at http://homepage.cs.latrobe.edu.au/ypchen/psmalign/.

Efficient Conversion of RNA Pseudoknots to Knot-Free Structures Using a Graphical Model

Abstract: RNA secondary structures are vital in determining the 3-D structures of noncoding RNA molecules, which in turn affect their functions. Computational RNA secondary structure alignment and analysis are biologically significant, because they help identify numerous functionally important motifs. Unfortunately, many analysis methods suffer from computational intractability in the presence of pseudoknots. The conversion of knotted to knot-free secondary structures is an essential preprocessing step, and is regarded as pseudoknot removal. Although exact methods have been proposed for this task, their computational complexities are undetermined, and so their efficiencies in processing complex pseudoknots are currently unknown. We transformed the pseudoknot removal problem into a circle graph maximum weight independent set (MWIS) problem, in which each MWIS represents a unique optimal deknotted structure. An existing circle graph MWIS algorithm was extended to report either single or all solutions. Its time complexity depends on the number of MWISs, and is guaranteed to report one solution in polynomial time. Experimental results suggest that our extended algorithm is much more efficient than the state-of-the-art tool. We also devised a novel concept called the structural scoring function, and investigated its effectiveness in more accurate solution candidate selection for a certain criteria.

Evaluation of Recent Computational Approaches in Short-Term Traffic Forecasting

Abstract: Computational technologies under the domain of intelligent systems are expected to help the rapidly increasing traffic congestion problem in recent traffic management. Traffic management requires efficient and accurate forecasting models to assist real time traffic control systems. Researchers have proposed various computational approaches, especially in short-term traffic flow forecasting, in order to establish reliable traffic patterns models and generate timely prediction results. Forecasting models should have high accuracy and low computational time to be applied in intelligent traffic management. Therefore, this paper aims to evaluate recent computational modeling approaches utilized in short-term traffic flow forecasting. These approaches are evaluated by real-world data collected on the British freeway (M6) from 1st to 30th November in 2014. The results indicate that neural network model outperforms generalized additive model and autoregressive integrated moving average model on the accuracy of freeway traffic forecasting.

Computer-Based Rehabilitation for Developing Speech and Language in Hearing-Impaired Children: A Systematic Review

Abstract: The purpose of this systematic review was to examine whether online or computer-based technologies were effective in assisting the development of speech and language skills in children with hearing loss. Relevant studies of children with hearing loss were analysed with reference to (1) therapy outcomes, (2) factors affecting outcomes, and (3) publication and methodological quality. The study quality was assessed using the 11-point PEDro scale. The review identified ten studies of relevance to the question of interest. All studies had relatively low PEDro quality scores with only four studies scoring in the mid-range on the scale. For these four studies, computer-based training appeared favourable at the group level. However, the small number of studies found significantly limits the generalizations and indicates the usage of these technologies in this population as an area requiring further rigorous research.