The widespread of 3D-printing technology has resulted in the appearance of many open-source prosthetic hand models, especially for partial hand amputations. However, most of these designs are not editable and while some are parametric to some degree, customization for every user is limited to scaling the size of a base design. As consequence, most prostheses fail to closely match the user specific anthropometry and have poor aesthetics, which could result in abandonment of the device. Furthermore, achieving a high degree of customization could be a time-consuming task and requires previous knowledge of CAD design. This work presents a prosthetic hand easy to customize by changing parametric dimensions of the finger phalanges and palm on an Excel sheet. Additionally, the design tackles common issues from previous 3D-printed body-powered prosthetic hands by incorporating new features such as the use of linkages instead of cables as finger flexors and a new cable-adjusting system which requires no additional tools and makes the tensioning of finger tendons easier and quicker.
{"title":"[Regular Paper] A Parametric 3D-Printed Body-Powered Hand Prosthesis Based on the Four-Bar Linkage Mechanism","authors":"Marlene Bustamante, Rodrigo Vega-Centeno, Midori Sanchez, Renato Mio","doi":"10.1109/BIBE.2018.00022","DOIUrl":"https://doi.org/10.1109/BIBE.2018.00022","url":null,"abstract":"The widespread of 3D-printing technology has resulted in the appearance of many open-source prosthetic hand models, especially for partial hand amputations. However, most of these designs are not editable and while some are parametric to some degree, customization for every user is limited to scaling the size of a base design. As consequence, most prostheses fail to closely match the user specific anthropometry and have poor aesthetics, which could result in abandonment of the device. Furthermore, achieving a high degree of customization could be a time-consuming task and requires previous knowledge of CAD design. This work presents a prosthetic hand easy to customize by changing parametric dimensions of the finger phalanges and palm on an Excel sheet. Additionally, the design tackles common issues from previous 3D-printed body-powered prosthetic hands by incorporating new features such as the use of linkages instead of cables as finger flexors and a new cable-adjusting system which requires no additional tools and makes the tensioning of finger tendons easier and quicker.","PeriodicalId":127507,"journal":{"name":"2018 IEEE 18th International Conference on Bioinformatics and Bioengineering (BIBE)","volume":"87 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126336523","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}
Yuta Furudate, Nanami Onuki, Kaori Chiba, Yuji Ishida, S. Mikami
Paralysis of fingers, which is caused by Hemiplegia, is difficult to recover. Patients often forced to leave hospital with paralysis remaining at hand. By this, a continuous rehabilitation at home is needed. However, it is difficult to carry out finger rehabilitation without help of therapists. To this end, we have been proposing an automated finger rehabilitation device which realizes home rehabilitation. A patient is asked by device to lift a finger, and the device measures whether undesirable movements are found on the other fingers by pressure sensors. To monitor an involuntary movement, it is necessary to evaluate the degree of the patient's condition of recovery. For this, we proposed a quantification method in our previous study. The method is based on the hypothesis that a patient is regarded as making recovery if his/her movement gets close to that of a healthy person. However, we consider only four fingers (index, middle, ring, little) are used to evaluate the degree of recovery because the thumb is different from the other finger in an anatomical structure. In this paper, we show a new recovery evaluation method that involves the sensor signals of all 5 fingers. We explain two possible evaluation methods: one is the model less simple integration method, and another is an integration by Generalized Linear Model (GLM). Comparing these methods, we conclude that the integration method by GLM provides a good scalar measurement of recovery, which was validated by the experiments conducted with patients who were previously evaluated by clinical scale.
{"title":"[Regular Paper] Automated Evaluation of Hand Motor Function Recovery by Using Finger Pressure Sensing Device for Home Rehabilitation","authors":"Yuta Furudate, Nanami Onuki, Kaori Chiba, Yuji Ishida, S. Mikami","doi":"10.1109/BIBE.2018.00047","DOIUrl":"https://doi.org/10.1109/BIBE.2018.00047","url":null,"abstract":"Paralysis of fingers, which is caused by Hemiplegia, is difficult to recover. Patients often forced to leave hospital with paralysis remaining at hand. By this, a continuous rehabilitation at home is needed. However, it is difficult to carry out finger rehabilitation without help of therapists. To this end, we have been proposing an automated finger rehabilitation device which realizes home rehabilitation. A patient is asked by device to lift a finger, and the device measures whether undesirable movements are found on the other fingers by pressure sensors. To monitor an involuntary movement, it is necessary to evaluate the degree of the patient's condition of recovery. For this, we proposed a quantification method in our previous study. The method is based on the hypothesis that a patient is regarded as making recovery if his/her movement gets close to that of a healthy person. However, we consider only four fingers (index, middle, ring, little) are used to evaluate the degree of recovery because the thumb is different from the other finger in an anatomical structure. In this paper, we show a new recovery evaluation method that involves the sensor signals of all 5 fingers. We explain two possible evaluation methods: one is the model less simple integration method, and another is an integration by Generalized Linear Model (GLM). Comparing these methods, we conclude that the integration method by GLM provides a good scalar measurement of recovery, which was validated by the experiments conducted with patients who were previously evaluated by clinical scale.","PeriodicalId":127507,"journal":{"name":"2018 IEEE 18th International Conference on Bioinformatics and Bioengineering (BIBE)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125234218","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 the new era of genetic epidemiology, there have been growing interest in studying genetic variants and their associations to complex diseases. Advances in modern computational approaches have led to the search for useful interacting genetic variants that are associated to the manifestation of a disease. However, these conventional strategies face number of challenges in predicting interesting interactions when data acquisition and dimensionality increases. Deep learning promises empirical success in number of applications including bioinformatics to drive insights of biological complexities. A deep neural network was previously proposed to identify true causative two-locus SNP interactions. The method was evaluated on various simulated and real datasets. In this study, the performance of the previously proposed deep learning method is maximized by improving network learning and avoiding overfitting. The method is further extended for performing sensitivity analysis. The performance of the method is evaluated on chronical dialysis patient's data for identifying higher-order interactions. It was observed that the highly ranked two-locus and three-locus SNP interactions in mitochondrial D-loop has the highest risk for the manifestation of disease.
{"title":"[Regular Paper] An Intensive Search for Higher-Order Gene-Gene Interactions by Improving Deep Learning Model","authors":"Suneetha Uppu, A. Krishna","doi":"10.1109/BIBE.2018.00027","DOIUrl":"https://doi.org/10.1109/BIBE.2018.00027","url":null,"abstract":"In the new era of genetic epidemiology, there have been growing interest in studying genetic variants and their associations to complex diseases. Advances in modern computational approaches have led to the search for useful interacting genetic variants that are associated to the manifestation of a disease. However, these conventional strategies face number of challenges in predicting interesting interactions when data acquisition and dimensionality increases. Deep learning promises empirical success in number of applications including bioinformatics to drive insights of biological complexities. A deep neural network was previously proposed to identify true causative two-locus SNP interactions. The method was evaluated on various simulated and real datasets. In this study, the performance of the previously proposed deep learning method is maximized by improving network learning and avoiding overfitting. The method is further extended for performing sensitivity analysis. The performance of the method is evaluated on chronical dialysis patient's data for identifying higher-order interactions. It was observed that the highly ranked two-locus and three-locus SNP interactions in mitochondrial D-loop has the highest risk for the manifestation of disease.","PeriodicalId":127507,"journal":{"name":"2018 IEEE 18th International Conference on Bioinformatics and Bioengineering (BIBE)","volume":"113 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133895143","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}
Shaikh Farhad Hossain, S. Wijaya, Ming Huang, I. Batubara, S. Kanaya, M. A. Farhad
Various medicinal plants are available in Bangladesh and these plants are used as traditional medicines for healing and health maintenance. Unani is one of the traditional medicine systems popular among Bangladeshi people because of its high success rate. Disease phenotype is changing constantly. It is Challenging for researchers to get the right medicinal ingredients, for the right disease, within a reasonable time. So we need to analyze the right plants for the right disease based on the existing formulas and to find out the relationship between plant and disease. The predicted plant-disease relations will help the health researcher or pharmacist for finding new drugs for new diseases. In our datasets, we have 409 plants, which are used as ingredients of 609 Unani formulas. Based on 609 formulas, we enlisted and sorted the relationship between diseases and plants. We assigned 609 Unani formulas to 18 National Center for Biotechnology Information (NCBI) disease classes. We then constructed the network of Unani formulas based on their ingredient similarity and applied DPclusO algorithm to find clusters. Clusters are associated with dominant disease and dominant plants by voting thus we established relations between plants and diseases. We predicted associations between 12 diseases and 151 plants. We validated our prediction based on the global set of Unani formulas and obtained 85.57% accuracy
{"title":"Prediction of Plant-Disease Relations Based on Unani Formulas by Network Analysis","authors":"Shaikh Farhad Hossain, S. Wijaya, Ming Huang, I. Batubara, S. Kanaya, M. A. Farhad","doi":"10.1109/BIBE.2018.00075","DOIUrl":"https://doi.org/10.1109/BIBE.2018.00075","url":null,"abstract":"Various medicinal plants are available in Bangladesh and these plants are used as traditional medicines for healing and health maintenance. Unani is one of the traditional medicine systems popular among Bangladeshi people because of its high success rate. Disease phenotype is changing constantly. It is Challenging for researchers to get the right medicinal ingredients, for the right disease, within a reasonable time. So we need to analyze the right plants for the right disease based on the existing formulas and to find out the relationship between plant and disease. The predicted plant-disease relations will help the health researcher or pharmacist for finding new drugs for new diseases. In our datasets, we have 409 plants, which are used as ingredients of 609 Unani formulas. Based on 609 formulas, we enlisted and sorted the relationship between diseases and plants. We assigned 609 Unani formulas to 18 National Center for Biotechnology Information (NCBI) disease classes. We then constructed the network of Unani formulas based on their ingredient similarity and applied DPclusO algorithm to find clusters. Clusters are associated with dominant disease and dominant plants by voting thus we established relations between plants and diseases. We predicted associations between 12 diseases and 151 plants. We validated our prediction based on the global set of Unani formulas and obtained 85.57% accuracy","PeriodicalId":127507,"journal":{"name":"2018 IEEE 18th International Conference on Bioinformatics and Bioengineering (BIBE)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134028650","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}
This paper shows how a standard proportional-integral-plus controller, based on a non-minimal state space (NMSS) design, can be extended to reduce the effects of measurement noise and so yield smoother control inputs for automated drug delivery control applications. Use of a NMSS model for state variable feedback control design, in which all the states are based on the directly measured input and output variables, removes the need for state estimation. Nonetheless, a stochastic NMSS form, with a Kalman filter, can optionally be introduced to reduce the effect of measurement noise and so yield smoother control inputs. In this case, the Kalman filter attenuates measurement noise but does not address state disturbances. In this article, we propose a modification to the stochastic NMSS control system to enable the elimination of such state disturbances. This involves further extending the non–minimal state vector to include additional terms based on the innovations. We compare the deterministic, stochastic and extended stochastic NMSS controllers via a simulation of the control of anaesthesia using propofol.
{"title":"[Regular Paper] Stochastic Non-minimal State Space Control with Application to Automated Drug Delivery","authors":"Emma D. Wilson, Q. Clairon, C. Taylor","doi":"10.1109/BIBE.2018.00014","DOIUrl":"https://doi.org/10.1109/BIBE.2018.00014","url":null,"abstract":"This paper shows how a standard proportional-integral-plus controller, based on a non-minimal state space (NMSS) design, can be extended to reduce the effects of measurement noise and so yield smoother control inputs for automated drug delivery control applications. Use of a NMSS model for state variable feedback control design, in which all the states are based on the directly measured input and output variables, removes the need for state estimation. Nonetheless, a stochastic NMSS form, with a Kalman filter, can optionally be introduced to reduce the effect of measurement noise and so yield smoother control inputs. In this case, the Kalman filter attenuates measurement noise but does not address state disturbances. In this article, we propose a modification to the stochastic NMSS control system to enable the elimination of such state disturbances. This involves further extending the non–minimal state vector to include additional terms based on the innovations. We compare the deterministic, stochastic and extended stochastic NMSS controllers via a simulation of the control of anaesthesia using propofol.","PeriodicalId":127507,"journal":{"name":"2018 IEEE 18th International Conference on Bioinformatics and Bioengineering (BIBE)","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134354456","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}
Yuan-Hsiang Chang, H. Yokota, K. Abe, Ming-Dar Tsai
This paper proposes an accurate 3D segmentation method for visualization and quantitative analysis of differentiation activities of mouse embryonic stem (ES) cells using time-lapse confocal fluorescence microscopy images. One of critical tasks in ES cell segmentation arises due to that ES cell nuclei are often close to each other. Several segmentation methods by convexities and concavities on cell or nucleus contours to detect possible touching cells or nuclei were proposed. Comparing to image processing methods, these methods are more accurate in some conditions, however, still cannot detect touching nuclei without concavities on nucleus contours. Our method uses the nucleus size and convex, concave, strait and extrusion features on nucleus contour to touch a boundary between touching cell nuclei in 2D slices and interslices. Experimental results show our method can well detect touching boundaries of 2D and 3D nucleus for confocal microscopy images of mouse ES cells in an early stage of differentiating into neural progenitor cells. Based on the accurate ES cell segmentation, cell activities (velocities and shape changes) during differentiation can be accurately visualized and quantitatively analyzed.
{"title":"[Regular Paper] Three-Dimensional Segmentation of Mouse Embryonic Stem Cell Nuclei for Quantitative Analysis of Differentiation Activity Using Time-Lapse Fluorescence Microscopy Images","authors":"Yuan-Hsiang Chang, H. Yokota, K. Abe, Ming-Dar Tsai","doi":"10.1109/BIBE.2018.00065","DOIUrl":"https://doi.org/10.1109/BIBE.2018.00065","url":null,"abstract":"This paper proposes an accurate 3D segmentation method for visualization and quantitative analysis of differentiation activities of mouse embryonic stem (ES) cells using time-lapse confocal fluorescence microscopy images. One of critical tasks in ES cell segmentation arises due to that ES cell nuclei are often close to each other. Several segmentation methods by convexities and concavities on cell or nucleus contours to detect possible touching cells or nuclei were proposed. Comparing to image processing methods, these methods are more accurate in some conditions, however, still cannot detect touching nuclei without concavities on nucleus contours. Our method uses the nucleus size and convex, concave, strait and extrusion features on nucleus contour to touch a boundary between touching cell nuclei in 2D slices and interslices. Experimental results show our method can well detect touching boundaries of 2D and 3D nucleus for confocal microscopy images of mouse ES cells in an early stage of differentiating into neural progenitor cells. Based on the accurate ES cell segmentation, cell activities (velocities and shape changes) during differentiation can be accurately visualized and quantitatively analyzed.","PeriodicalId":127507,"journal":{"name":"2018 IEEE 18th International Conference on Bioinformatics and Bioengineering (BIBE)","volume":"60 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134558995","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}
Seunghyun Park, You Jin Kim, Jeong-Whun Kim, Jin Joo Park, Borim Ryu, Jung-Woo Ha
Precise prediction of severe diseases resulting in mortality is one of the main issues in medical fields. Even if pathological and radiological measurements provide competitive precision, they usually require large costs of time and expense to obtain and analyze the data for prediction. Recently, end-to-end approaches based on deep neural networks have been proposed, however, they still suffer from the low classification performance and difficulties of interpretation. In this study, we propose a novel disease prediction method, EHAN (EHR History-based prediction using Attention Network), based on the recurrent neural network (RNN) and attention mechanism. The proposed method incorporates (1) a bidirectional gated recurrent units (GRU) for automated sequential modeling, (2) attention mechanism for improving long-term dependence modeling, (3) RNN-based gradient-weighted class activation mapping (Grad-CAM) to visualize the class specific attention-weights. We conducted the experiments to predict the occurrence of risky disease containing cardiovascular and cerebrovascular diseases from more than 40,000 hypertension patients' electronic health records (EHR). The results showed that the proposed method outperformed the state-of-the-art model with respect to the various performance metrics. Furthermore, we confirmed that the proposed visualizing methods can be used to assist data-driven discovery.
对导致死亡的严重疾病的精确预测是医学领域的主要问题之一。即使病理学和放射学测量提供了相当的精度，它们通常需要大量的时间和费用来获取和分析预测的数据。近年来，人们提出了基于深度神经网络的端到端方法，但它们仍然存在分类性能低和解释困难的问题。在本研究中，我们提出了一种基于递归神经网络(RNN)和注意机制的疾病预测新方法EHAN (EHR History-based prediction using Attention Network)。该方法结合了(1)用于自动顺序建模的双向门控循环单元(GRU)，(2)用于改进长期依赖建模的注意机制，(3)基于rnn的梯度加权类激活映射(Grad-CAM)来可视化类特定的注意权重。我们对4万多名高血压患者的电子健康记录(EHR)进行了包含心脑血管疾病的危险疾病发生预测实验。结果表明，该方法在各种性能指标方面优于最先进的模型。此外，我们证实了所提出的可视化方法可以用于协助数据驱动的发现。
{"title":"[Regular Paper] Interpretable Prediction of Vascular Diseases from Electronic Health Records via Deep Attention Networks","authors":"Seunghyun Park, You Jin Kim, Jeong-Whun Kim, Jin Joo Park, Borim Ryu, Jung-Woo Ha","doi":"10.1109/BIBE.2018.00028","DOIUrl":"https://doi.org/10.1109/BIBE.2018.00028","url":null,"abstract":"Precise prediction of severe diseases resulting in mortality is one of the main issues in medical fields. Even if pathological and radiological measurements provide competitive precision, they usually require large costs of time and expense to obtain and analyze the data for prediction. Recently, end-to-end approaches based on deep neural networks have been proposed, however, they still suffer from the low classification performance and difficulties of interpretation. In this study, we propose a novel disease prediction method, EHAN (EHR History-based prediction using Attention Network), based on the recurrent neural network (RNN) and attention mechanism. The proposed method incorporates (1) a bidirectional gated recurrent units (GRU) for automated sequential modeling, (2) attention mechanism for improving long-term dependence modeling, (3) RNN-based gradient-weighted class activation mapping (Grad-CAM) to visualize the class specific attention-weights. We conducted the experiments to predict the occurrence of risky disease containing cardiovascular and cerebrovascular diseases from more than 40,000 hypertension patients' electronic health records (EHR). The results showed that the proposed method outperformed the state-of-the-art model with respect to the various performance metrics. Furthermore, we confirmed that the proposed visualizing methods can be used to assist data-driven discovery.","PeriodicalId":127507,"journal":{"name":"2018 IEEE 18th International Conference on Bioinformatics and Bioengineering (BIBE)","volume":"107 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132642585","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}
Breast cancer diagnosis requires a pathologist to analyze the histology slides under various magnifications. An automated diagnosis method to aid pathologists that is magnification independent will significantly save time, reduce cost and mitigate subjectivity and errors in current histopathological diagnosis procedures. This paper presents a deep learning network, called MVPNet and a customized data augmentation technique, called NuView, for magnification independent diagnosis. MVPNet is tailored to tackle the most common issues (diversity, relatively small size of datasets and manifestation of diagnostic biomarkers at various magnification levels) with breast cancer histology data to perform the classification. The network simultaneously analyzes local and global features of a given tissue image. It does so by viewing the tissue at varying levels of relative nuclei sizes. MVPNet has significantly less parameters than standard transfer learning deep models with comparable performance and it combines and processes local and global features simulatenously for effective diagnosis. Additionally, NuView extracts tumor nuclei location and points the attention of MVPNet to the informative region specifically. The method gives an average magnification independent classification accuracy of 92.2% as compared to 83% reported in literature on the BreaKHis database.
{"title":"[Regular Paper] MVPNets: Multi-viewing Path Deep Learning Neural Networks for Magnification Invariant Diagnosis in Breast Cancer","authors":"P. Jonnalagedda, D. Schmolze, B. Bhanu","doi":"10.1109/BIBE.2018.00044","DOIUrl":"https://doi.org/10.1109/BIBE.2018.00044","url":null,"abstract":"Breast cancer diagnosis requires a pathologist to analyze the histology slides under various magnifications. An automated diagnosis method to aid pathologists that is magnification independent will significantly save time, reduce cost and mitigate subjectivity and errors in current histopathological diagnosis procedures. This paper presents a deep learning network, called MVPNet and a customized data augmentation technique, called NuView, for magnification independent diagnosis. MVPNet is tailored to tackle the most common issues (diversity, relatively small size of datasets and manifestation of diagnostic biomarkers at various magnification levels) with breast cancer histology data to perform the classification. The network simultaneously analyzes local and global features of a given tissue image. It does so by viewing the tissue at varying levels of relative nuclei sizes. MVPNet has significantly less parameters than standard transfer learning deep models with comparable performance and it combines and processes local and global features simulatenously for effective diagnosis. Additionally, NuView extracts tumor nuclei location and points the attention of MVPNet to the informative region specifically. The method gives an average magnification independent classification accuracy of 92.2% as compared to 83% reported in literature on the BreaKHis database.","PeriodicalId":127507,"journal":{"name":"2018 IEEE 18th International Conference on Bioinformatics and Bioengineering (BIBE)","volume":"32 2","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120924979","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}
Bioinformatics workloads are characterized by huge data sets and complex algorithms, requiring enormous data processing and making high performance heterogeneous computation platforms such as FPGAs and GPUs highly relevant. We compare three accelerated implementations of the widely used BWA-MEM genomic mapping tool as a case study on design-time optimization for heterogeneous architectures: BWA-MEM-CUDA, BWA-MEM-OpenCL, and BWA-MEMVHDL, each using an optimized Smith-Waterman algorithm implementation. Optimization of design-time is important because of the significant development effort of such implementations: BWA-MEM-CUDA and BWA-MEM-OpenCL require 5-7x more lines of code to express the Smith-Waterman algorithm, while BWA-MEM-VHDL requires more than 40x as many lines of code. Similar differences hold for required implementation time, ranging from one month for BWA-MEMOpenCL to six months for BWA-MEM-VHDL. The advantages and disadvantages of each implementation are described using both quantitative and qualitative metrics, and recommendations are given for future algorithm implementations.
{"title":"Comparative Analysis of System-Level Acceleration Techniques in Bioinformatics: A Case Study of Accelerating the Smith-Waterman Algorithm for BWA-MEM","authors":"Ernst Houtgast, V. Sima, K. Bertels, Z. Al-Ars","doi":"10.1109/BIBE.2018.00053","DOIUrl":"https://doi.org/10.1109/BIBE.2018.00053","url":null,"abstract":"Bioinformatics workloads are characterized by huge data sets and complex algorithms, requiring enormous data processing and making high performance heterogeneous computation platforms such as FPGAs and GPUs highly relevant. We compare three accelerated implementations of the widely used BWA-MEM genomic mapping tool as a case study on design-time optimization for heterogeneous architectures: BWA-MEM-CUDA, BWA-MEM-OpenCL, and BWA-MEMVHDL, each using an optimized Smith-Waterman algorithm implementation. Optimization of design-time is important because of the significant development effort of such implementations: BWA-MEM-CUDA and BWA-MEM-OpenCL require 5-7x more lines of code to express the Smith-Waterman algorithm, while BWA-MEM-VHDL requires more than 40x as many lines of code. Similar differences hold for required implementation time, ranging from one month for BWA-MEMOpenCL to six months for BWA-MEM-VHDL. The advantages and disadvantages of each implementation are described using both quantitative and qualitative metrics, and recommendations are given for future algorithm implementations.","PeriodicalId":127507,"journal":{"name":"2018 IEEE 18th International Conference on Bioinformatics and Bioengineering (BIBE)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116488403","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 model of gene-gene interaction contributing to the biological insight of disease pathology have received significant attention from both medical and computing communities. Through the modeled interactome map, the biological significant of the mutated genes can be revealed and treatments targeting these genes can be taken to prevent further proliferation of the mutated genes. In this paper we propose a novel computational way to interrogate interaction between genes. We utilize centroid computation in the hybrid genetic algorithm and neural network to model interaction between leukemia-related genes. Results indicated the effectiveness of centroid value in detecting significant interactions of gene. Hub genes were also identified.
{"title":"Pathway Analysis of Marker Genes for Leukemia Cancer using Enhanced Genetic Algorithm-Neural Network (enGANN)","authors":"Hau Cherng Wong, C. Lee, Dong-Ling Tong","doi":"10.1109/BIBE.2018.00029","DOIUrl":"https://doi.org/10.1109/BIBE.2018.00029","url":null,"abstract":"The model of gene-gene interaction contributing to the biological insight of disease pathology have received significant attention from both medical and computing communities. Through the modeled interactome map, the biological significant of the mutated genes can be revealed and treatments targeting these genes can be taken to prevent further proliferation of the mutated genes. In this paper we propose a novel computational way to interrogate interaction between genes. We utilize centroid computation in the hybrid genetic algorithm and neural network to model interaction between leukemia-related genes. Results indicated the effectiveness of centroid value in detecting significant interactions of gene. Hub genes were also identified.","PeriodicalId":127507,"journal":{"name":"2018 IEEE 18th International Conference on Bioinformatics and Bioengineering (BIBE)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128731504","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：