Human metapneumovirus has persistently been the leading causative agent of acute respiratory infections in young children and the elderly worldwide. The respiratory tract illness caused by HMPV yields fatal levels of morbidity and mortality rate in young children under five and the immunocompromised. To study the genetic structure of HMPV, this paper conducts a genomic analysis of the nine genes (N, P, M, F, M2-1, M2-2, SH, G, and L) of human metapneumovirus subtype A1, A2, B1, and B2. Through multiple sequence alignments, decision tree, Apriori algorithm, and phylogenetic reconstruction, this paper investigates the genome-wise discrepancy and the protein-wise discrepancy between different HMPV strains. The results of the experiment indicate that the four HMPV subtypes show high similarity while displaying distinct attributes. The role of glycoprotein (G) and small hydrophobic protein (SH) are found to display the most variance among the four subtypes. The Apriori algorithm shows that amino acid serine and lysine are the most frequent among the four subtypes. Under Apriori algorithm 19 window, it has been found that the four subtypes display some degree of similarity in terms of their frequencies of the amino acid lysine(K). On the other hand, two clades of HMPV seem to split in terms of their frequencies of the amino acid serine(S). Hence, the role of glycoprotein and small hydrophobic protein and the contribution of amino acids serine and lysine to the nine polypeptides are suggested as a future research.
{"title":"Genomic Comparison of Four Metapneumovirus Strains Using Decision Tree, Apriori Algorithm, ClustalW, and Phylogenetic Reconstruction","authors":"Sang-Ran Lim, Taeseon Yoon","doi":"10.1145/3309129.3309130","DOIUrl":"https://doi.org/10.1145/3309129.3309130","url":null,"abstract":"Human metapneumovirus has persistently been the leading causative agent of acute respiratory infections in young children and the elderly worldwide. The respiratory tract illness caused by HMPV yields fatal levels of morbidity and mortality rate in young children under five and the immunocompromised. To study the genetic structure of HMPV, this paper conducts a genomic analysis of the nine genes (N, P, M, F, M2-1, M2-2, SH, G, and L) of human metapneumovirus subtype A1, A2, B1, and B2. Through multiple sequence alignments, decision tree, Apriori algorithm, and phylogenetic reconstruction, this paper investigates the genome-wise discrepancy and the protein-wise discrepancy between different HMPV strains. The results of the experiment indicate that the four HMPV subtypes show high similarity while displaying distinct attributes. The role of glycoprotein (G) and small hydrophobic protein (SH) are found to display the most variance among the four subtypes. The Apriori algorithm shows that amino acid serine and lysine are the most frequent among the four subtypes. Under Apriori algorithm 19 window, it has been found that the four subtypes display some degree of similarity in terms of their frequencies of the amino acid lysine(K). On the other hand, two clades of HMPV seem to split in terms of their frequencies of the amino acid serine(S). Hence, the role of glycoprotein and small hydrophobic protein and the contribution of amino acids serine and lysine to the nine polypeptides are suggested as a future research.","PeriodicalId":326530,"journal":{"name":"Proceedings of the 5th International Conference on Bioinformatics Research and Applications","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125958475","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Lifelong learning topic models identify the hidden concepts discussed in the collection of documents. The concepts are represented as topics having groups of ordered words based on their relevance to the topic. Lifelong learning models have an automatic learning mechanism which allows continuous learning without external support. In the process, the model gets more knowledgeable with experience as it learns from the past in the form of rules. It is carries rules to the future and utilize them when a similar scenario arises. The existing lifelong learning topic models heavily rely on statistical measures to learn rules that leads to two limitations. The rules are evaluated for fixed number of words while ignoring the natural arrangement of words within the documents. Moreover, the rules have arbitrary orientation that causes repeated patterns of transferring the impact of a rule into a topic during the early iterations of the inference technique. In this research work, we introduce complex networks analysis for learning rules which addresses both of the limitations discussed. The rules are obtained through hierarchical clustering of the complex network that have different number of words within a rule and have directed orientation. The proposed approach improves the utilization of rules for improved quality of topics at higher performance with unidirectional rules on the standard lifelong learning dataset.
{"title":"Graph Clustering Based Size Varying Rules for Lifelong Topic Modeling","authors":"Muhammad Taimoor Khan, S. Khalid, Furqan Aziz","doi":"10.1145/3309129.3309146","DOIUrl":"https://doi.org/10.1145/3309129.3309146","url":null,"abstract":"Lifelong learning topic models identify the hidden concepts discussed in the collection of documents. The concepts are represented as topics having groups of ordered words based on their relevance to the topic. Lifelong learning models have an automatic learning mechanism which allows continuous learning without external support. In the process, the model gets more knowledgeable with experience as it learns from the past in the form of rules. It is carries rules to the future and utilize them when a similar scenario arises. The existing lifelong learning topic models heavily rely on statistical measures to learn rules that leads to two limitations. The rules are evaluated for fixed number of words while ignoring the natural arrangement of words within the documents. Moreover, the rules have arbitrary orientation that causes repeated patterns of transferring the impact of a rule into a topic during the early iterations of the inference technique. In this research work, we introduce complex networks analysis for learning rules which addresses both of the limitations discussed. The rules are obtained through hierarchical clustering of the complex network that have different number of words within a rule and have directed orientation. The proposed approach improves the utilization of rules for improved quality of topics at higher performance with unidirectional rules on the standard lifelong learning dataset.","PeriodicalId":326530,"journal":{"name":"Proceedings of the 5th International Conference on Bioinformatics Research and Applications","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127010324","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Mohamed Haroon Abdul Jabbar, S. Shanmugam, P. Khiew
During standard hemodialysis (HD), there is tendency for a rise in body temperature, which can possibly cause life-threatening complications. The analysis of thermal energy exchange in a dialyzer can be significant to provide constant body temperature, which can necessitate the development of an effective temperature controller. In this paper, the main aim is to evaluate the heat transfer that takes place in a dialyzer model during HD and a Polyflux 210H dialyzer membrane model was developed using COMSOL Multiphysics® software. The clearance rate of toxins was computed and validated for various blood flow rates. Then the heat transfer physics was added to investigate the effect of heat transfer taking place in the dialyzer. The clearance rates show significant improvement (<5% error) compared to previous published work (>11.7% error) and a strong agreement with the manufacturer's data. The model exhibited a trend in temperature profile across the dialyzer membrane and the blood temperature has decreased up to 1.15°C using cool dialysate settings. The dialyzer acts as a heat exchanger during HD. Our study reveals the temperature changes taking place in the dialyzer, which necessitates a system to control and regulate the dialysate temperature to compensate for this heat loss.
{"title":"Investigation on Heat and Mass transfer in a Dialyzer Membrane Model for the Development of Dialysate Temperature Controller","authors":"Mohamed Haroon Abdul Jabbar, S. Shanmugam, P. Khiew","doi":"10.1145/3309129.3309135","DOIUrl":"https://doi.org/10.1145/3309129.3309135","url":null,"abstract":"During standard hemodialysis (HD), there is tendency for a rise in body temperature, which can possibly cause life-threatening complications. The analysis of thermal energy exchange in a dialyzer can be significant to provide constant body temperature, which can necessitate the development of an effective temperature controller. In this paper, the main aim is to evaluate the heat transfer that takes place in a dialyzer model during HD and a Polyflux 210H dialyzer membrane model was developed using COMSOL Multiphysics® software. The clearance rate of toxins was computed and validated for various blood flow rates. Then the heat transfer physics was added to investigate the effect of heat transfer taking place in the dialyzer. The clearance rates show significant improvement (<5% error) compared to previous published work (>11.7% error) and a strong agreement with the manufacturer's data. The model exhibited a trend in temperature profile across the dialyzer membrane and the blood temperature has decreased up to 1.15°C using cool dialysate settings. The dialyzer acts as a heat exchanger during HD. Our study reveals the temperature changes taking place in the dialyzer, which necessitates a system to control and regulate the dialysate temperature to compensate for this heat loss.","PeriodicalId":326530,"journal":{"name":"Proceedings of the 5th International Conference on Bioinformatics Research and Applications","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130780414","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
A. Santos, R. M. Sousa, M. Bianchi, Leandro Lima da Silva, E. Cordioli
Deep convolutional neural networks (CNNs) have proven its potential for many tasks related to object identification and classification. This study aims to show the performance of several convolutional neural networks architectures applied to the diagnosis and screening of skin lesions in patients using different training techniques: Random weights initialization, feature extraction and extending model. A dataset of 1000 clinical images proven by biopsy or consensus among specialists were the examples applied at the various architectures which were end-to-end trained from images directly, using only pixels and disease labels as inputs. The predictions provided from the models intended to claim whether the lesion could be treated by doctors with images only on a teledermatology approach or if it is necessary to prescribe a biopsy or referral to a face-to-face consultation. The model can also tell the urgency of the case and the group of diseases which that lesion belongs to. Performances of deep neural networks in all proposed tasks demonstrated that artificial intelligence has the potential to perform the screening of skin lesions with a level of competence comparable to dermatologists. It is projected 6.3 billion signatures of smartphone by the year 2021 [38]. Therefore, deep neural networks incorporated in mobile devices can amplify the reach of dermatologists outside their offices providing universal low-cost access to dermatological diagnostics.
{"title":"Screening Feasibility and Comparison of Deep Artificial Neural Networks Algorithms for Classification of Skin Lesions","authors":"A. Santos, R. M. Sousa, M. Bianchi, Leandro Lima da Silva, E. Cordioli","doi":"10.1145/3309129.3309137","DOIUrl":"https://doi.org/10.1145/3309129.3309137","url":null,"abstract":"Deep convolutional neural networks (CNNs) have proven its potential for many tasks related to object identification and classification. This study aims to show the performance of several convolutional neural networks architectures applied to the diagnosis and screening of skin lesions in patients using different training techniques: Random weights initialization, feature extraction and extending model. A dataset of 1000 clinical images proven by biopsy or consensus among specialists were the examples applied at the various architectures which were end-to-end trained from images directly, using only pixels and disease labels as inputs. The predictions provided from the models intended to claim whether the lesion could be treated by doctors with images only on a teledermatology approach or if it is necessary to prescribe a biopsy or referral to a face-to-face consultation. The model can also tell the urgency of the case and the group of diseases which that lesion belongs to. Performances of deep neural networks in all proposed tasks demonstrated that artificial intelligence has the potential to perform the screening of skin lesions with a level of competence comparable to dermatologists. It is projected 6.3 billion signatures of smartphone by the year 2021 [38]. Therefore, deep neural networks incorporated in mobile devices can amplify the reach of dermatologists outside their offices providing universal low-cost access to dermatological diagnostics.","PeriodicalId":326530,"journal":{"name":"Proceedings of the 5th International Conference on Bioinformatics Research and Applications","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121857854","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Qiang Cao, S. Liao, Xiangqian Meng, Han Ye, Zhenbin Yan, Puxi Wang
Identifying viable embryos for implantation is one of the most relevant aspects in assisted reproductive technology. However, embryo selection highly depends on visual examination by embryologists via microscopy, and their evaluations are often subjective. The rapid growth of image processing technology has resulted in increased interest in the use of machine learning methods for embryo selection in in vitro fertilization (IVF) programs. The present study uses deep learning method for the morphological classification of embryos based on medical images. The proposed system is trained and tested on a real data set of 1,310 images from 344 embryos and evaluated by comparison with other traditional machine learning methods to solve similar classification problems. The results indicate that our new deep learning model significantly outperforms other methods. Our work contributes immensely to the fields of assisted reproductive technology, medical image processing, and decision support system design.
{"title":"Identification of Viable Embryos Using Deep Learning for Medical Image","authors":"Qiang Cao, S. Liao, Xiangqian Meng, Han Ye, Zhenbin Yan, Puxi Wang","doi":"10.1145/3309129.3309143","DOIUrl":"https://doi.org/10.1145/3309129.3309143","url":null,"abstract":"Identifying viable embryos for implantation is one of the most relevant aspects in assisted reproductive technology. However, embryo selection highly depends on visual examination by embryologists via microscopy, and their evaluations are often subjective. The rapid growth of image processing technology has resulted in increased interest in the use of machine learning methods for embryo selection in in vitro fertilization (IVF) programs. The present study uses deep learning method for the morphological classification of embryos based on medical images. The proposed system is trained and tested on a real data set of 1,310 images from 344 embryos and evaluated by comparison with other traditional machine learning methods to solve similar classification problems. The results indicate that our new deep learning model significantly outperforms other methods. Our work contributes immensely to the fields of assisted reproductive technology, medical image processing, and decision support system design.","PeriodicalId":326530,"journal":{"name":"Proceedings of the 5th International Conference on Bioinformatics Research and Applications","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123889052","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Proceedings of the 5th International Conference on Bioinformatics Research and Applications","authors":"","doi":"10.1145/3309129","DOIUrl":"https://doi.org/10.1145/3309129","url":null,"abstract":"","PeriodicalId":326530,"journal":{"name":"Proceedings of the 5th International Conference on Bioinformatics Research and Applications","volume":"707 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115961687","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The skin protects our body from heat and light of the sun and other threats. One of the illnesses that threaten the skin is the skin cancer. Skin cancer may start with an irregular shaped mole with size greater than a pencil eraser. This study focuses on the non-invasive approach in detecting and classifying skin cancer. Geometrical features of the moles suspected for skin cancer are extracted following the asymmetry, border, and diameter parameters of the ABCD-Rule of Dermoscopy. In particular, greatest and shortest diameter, irregularity index and equivalent diameter are the parameters loaded in the dataset for classification. Classification of mole images is done through k-Nearest Neighbors (k-NN) algorithm. The overall result showed 86.67% accuracy in determining the classification.
{"title":"Skin Cancer Detection and Classification for Moles Using K-Nearest Neighbor Algorithm","authors":"N. Linsangan, J. Adtoon","doi":"10.1145/3309129.3309141","DOIUrl":"https://doi.org/10.1145/3309129.3309141","url":null,"abstract":"The skin protects our body from heat and light of the sun and other threats. One of the illnesses that threaten the skin is the skin cancer. Skin cancer may start with an irregular shaped mole with size greater than a pencil eraser. This study focuses on the non-invasive approach in detecting and classifying skin cancer. Geometrical features of the moles suspected for skin cancer are extracted following the asymmetry, border, and diameter parameters of the ABCD-Rule of Dermoscopy. In particular, greatest and shortest diameter, irregularity index and equivalent diameter are the parameters loaded in the dataset for classification. Classification of mole images is done through k-Nearest Neighbors (k-NN) algorithm. The overall result showed 86.67% accuracy in determining the classification.","PeriodicalId":326530,"journal":{"name":"Proceedings of the 5th International Conference on Bioinformatics Research and Applications","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132283967","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In chocolate production, post-harvest procedure is one of the most critical factors. Fermentation is a vital procedure to consider since exact generation of acid contemplate to aroma and quality of the final product. This innovative study aims to classify the quality of the cacao beans after the post-harvest procedures. Classified sample beans from partner cacao trader were analyzed and became data sets of the device. Photographs are taken to the subjects and undergo image processing procedure then through k-Nearest Neighbors Algorithm (k-NN). Beans are classified to be well-fermented under fermentation and over-fermentation process. Function test and statistical analysis using confusion matrix revealed 97.22 percent accuracy in analyzing well-fermented beans, 92.59 percent accuracy in under fermented, 75 percent in over-fermented and 80 percent in analyzing unknowns. These results generated 92.50 percent overall accuracy of the device.
{"title":"Fermentation Level Classification of Cross Cut Cacao Beans Using k-NN Algorithm","authors":"Randy E. Angelia, N. Linsangan","doi":"10.1145/3309129.3309142","DOIUrl":"https://doi.org/10.1145/3309129.3309142","url":null,"abstract":"In chocolate production, post-harvest procedure is one of the most critical factors. Fermentation is a vital procedure to consider since exact generation of acid contemplate to aroma and quality of the final product. This innovative study aims to classify the quality of the cacao beans after the post-harvest procedures. Classified sample beans from partner cacao trader were analyzed and became data sets of the device. Photographs are taken to the subjects and undergo image processing procedure then through k-Nearest Neighbors Algorithm (k-NN). Beans are classified to be well-fermented under fermentation and over-fermentation process. Function test and statistical analysis using confusion matrix revealed 97.22 percent accuracy in analyzing well-fermented beans, 92.59 percent accuracy in under fermented, 75 percent in over-fermented and 80 percent in analyzing unknowns. These results generated 92.50 percent overall accuracy of the device.","PeriodicalId":326530,"journal":{"name":"Proceedings of the 5th International Conference on Bioinformatics Research and Applications","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127262262","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：