Multi-contrast high quality high-resolution (HR) Magnetic Resonance (MR) images enrich available information for diagnosis and analysis. Deep convolutional neural network methods have shown promising ability for MR image super-resolution (SR) given low-resolution (LR) MR images. Methods taking HR images as references (Ref) have made progress to enhance the effect of MR images SR. However, existing multi-contrast MR image SR approaches are based on contrasting-expanding backbones, which lose high frequency information of Ref image during downsampling. They also failed to transfer textures of Ref image into target domain. In this paper, we propose Edge Mask Transformer UNet (EMFU) for accelerating MR images SR. We propose Edge Mask Transformer (EMF) to generate global details and texture representation of target domain. Dual domain fusion module in UNet aggregates semantic information of the representation and LR image of target domain. Specifically, we extract and encode edge masks to guide the attention in EMF by re-distributing the embedding tensors, so that the network allocates more attention to image edge area. We also design a dual domain fusion module with self-attention and cross-attention to deeply fuse semantic information of multiple protocols for MRI. Extensive experiments show the effectiveness of our proposed EMFU, which surpasses state-of-the-art methods on benchmarks quantitatively and visually. Codes will be released to the community.
{"title":"Multi-contrast High Quality MR Image Super-Resolution with Dual Domain Knowledge Fusion","authors":"Runhan Wang, Ruiwei Zhao, Weijia Fu, X. Zhang, Yuejie Zhang, Rui Feng","doi":"10.1109/BIBM55620.2022.9995219","DOIUrl":"https://doi.org/10.1109/BIBM55620.2022.9995219","url":null,"abstract":"Multi-contrast high quality high-resolution (HR) Magnetic Resonance (MR) images enrich available information for diagnosis and analysis. Deep convolutional neural network methods have shown promising ability for MR image super-resolution (SR) given low-resolution (LR) MR images. Methods taking HR images as references (Ref) have made progress to enhance the effect of MR images SR. However, existing multi-contrast MR image SR approaches are based on contrasting-expanding backbones, which lose high frequency information of Ref image during downsampling. They also failed to transfer textures of Ref image into target domain. In this paper, we propose Edge Mask Transformer UNet (EMFU) for accelerating MR images SR. We propose Edge Mask Transformer (EMF) to generate global details and texture representation of target domain. Dual domain fusion module in UNet aggregates semantic information of the representation and LR image of target domain. Specifically, we extract and encode edge masks to guide the attention in EMF by re-distributing the embedding tensors, so that the network allocates more attention to image edge area. We also design a dual domain fusion module with self-attention and cross-attention to deeply fuse semantic information of multiple protocols for MRI. Extensive experiments show the effectiveness of our proposed EMFU, which surpasses state-of-the-art methods on benchmarks quantitatively and visually. Codes will be released to the community.","PeriodicalId":210337,"journal":{"name":"2022 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-12-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121319431","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}
Pub Date : 2022-12-06DOI: 10.1109/BIBM55620.2022.9995383
Norah Saeed Awn, Yiming Li, Baoying Zhao, Min Zeng, Min Li
Recent studies have confirmed the significant effects of long non-coding RNAs (1ncRNAs) in understanding the mechanism of diseases. Because of the relatively small number of validated associations between 1ncRNAs and diseases, and previous computational methods have limited performance without capturing important features of sequences and ontology information, we developed LDAGSO, a novel deep learning framework to predict 1ncRNA and disease associations from 1ncRNA sequences and disease ontology. For 1ncRNA sequences, we converted them into graph structure based on k-mer technique and de Bruijn graph, and captured high-level features of the graph using graph convolutional networks. For diseases, we extracted ontology term paths from the disease ontology tree, and treated them as sentences to obtain their feature representation using Bidirectional Encoder Representations from Transformers (BERT) technique. Finally, these two kinds of features were fed into a fully connected layer to perform the task of association prediction between 1ncRNAs and diseases. According to the results, our approach provides state-of-the-art results when evaluated by leave-one-out cross-validation.
最近的研究证实了长链非编码rna (1ncRNAs)在理解疾病机制方面的重要作用。由于验证的1ncRNA与疾病之间关联的数量相对较少,并且先前的计算方法在没有捕获序列和本体信息的重要特征方面性能有限,因此我们开发了LDAGSO,一种新的深度学习框架,用于从1ncRNA序列和疾病本体预测1ncRNA和疾病关联。对于1ncRNA序列,我们基于k-mer技术和de Bruijn图将其转换为图结构,并使用图卷积网络捕获图的高级特征。对于疾病,我们从疾病本体树中提取本体术语路径,并将其作为句子处理,利用BERT (Bidirectional Encoder Representations from Transformers)技术获得其特征表示。最后,将这两种特征输入到一个完全连接的层中,以执行1ncrna与疾病之间的关联预测任务。根据结果,我们的方法提供了最先进的结果时,通过留一交叉验证评估。
{"title":"LDAGSO: Predicting 1ncRNA-Disease Associations from Graph Sequences and Disease Ontology via Deep Learning techniques","authors":"Norah Saeed Awn, Yiming Li, Baoying Zhao, Min Zeng, Min Li","doi":"10.1109/BIBM55620.2022.9995383","DOIUrl":"https://doi.org/10.1109/BIBM55620.2022.9995383","url":null,"abstract":"Recent studies have confirmed the significant effects of long non-coding RNAs (1ncRNAs) in understanding the mechanism of diseases. Because of the relatively small number of validated associations between 1ncRNAs and diseases, and previous computational methods have limited performance without capturing important features of sequences and ontology information, we developed LDAGSO, a novel deep learning framework to predict 1ncRNA and disease associations from 1ncRNA sequences and disease ontology. For 1ncRNA sequences, we converted them into graph structure based on k-mer technique and de Bruijn graph, and captured high-level features of the graph using graph convolutional networks. For diseases, we extracted ontology term paths from the disease ontology tree, and treated them as sentences to obtain their feature representation using Bidirectional Encoder Representations from Transformers (BERT) technique. Finally, these two kinds of features were fed into a fully connected layer to perform the task of association prediction between 1ncRNAs and diseases. According to the results, our approach provides state-of-the-art results when evaluated by leave-one-out cross-validation.","PeriodicalId":210337,"journal":{"name":"2022 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-12-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121419818","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}
Pub Date : 2022-12-06DOI: 10.1109/BIBM55620.2022.9994861
Salim Sazzed, P. Scheible, Jing He, W. Wriggers
The disordered nature of the actin network in Dictyostelium discoideum filopodia makes identifying filaments within noisy cryo-electron tomograms extremely challenging. In this work, we present a computationally efficient dynamic programming-based framework for tracing arbitrarily oriented actin filaments. Starting from locally determined seed points, it accumulates densities along paths of a particular length within 45° of the three Cartesian coordinate axes. This novel approach covers all possible orientations, so there is no need to assume a dominant direction as in earlier work. For each seed point, the path with the highest density value is selected, and it acts as a candidate filament segment (CFS) that is likely to form a part of a filament when it has a high path density value. The subsequent stages involve identifying groups of CFSs with high path densities by binning and merging them. The merging step considers the relative orientations and distances of CFSs to connect them. In addition, the CFSs are extended to fill the noise-induced gaps to some extent. In the current prototype software, we focused on the proof of the concept, using a noisy simulated tomogram with a known ground truth that closely mimics the appearance of an experimental map. We achieved an almost perfect precision score of 0.999, but this success came at the expense of a lower recall score 0.462 due to false negatives. We discuss the dependencies as well as the limitations of the current filament merging that need to be overcome to achieve a higher recall score in the future.
{"title":"Tracing Randomly Oriented Filaments in a Simulated Actin Network Tomogram","authors":"Salim Sazzed, P. Scheible, Jing He, W. Wriggers","doi":"10.1109/BIBM55620.2022.9994861","DOIUrl":"https://doi.org/10.1109/BIBM55620.2022.9994861","url":null,"abstract":"The disordered nature of the actin network in Dictyostelium discoideum filopodia makes identifying filaments within noisy cryo-electron tomograms extremely challenging. In this work, we present a computationally efficient dynamic programming-based framework for tracing arbitrarily oriented actin filaments. Starting from locally determined seed points, it accumulates densities along paths of a particular length within 45° of the three Cartesian coordinate axes. This novel approach covers all possible orientations, so there is no need to assume a dominant direction as in earlier work. For each seed point, the path with the highest density value is selected, and it acts as a candidate filament segment (CFS) that is likely to form a part of a filament when it has a high path density value. The subsequent stages involve identifying groups of CFSs with high path densities by binning and merging them. The merging step considers the relative orientations and distances of CFSs to connect them. In addition, the CFSs are extended to fill the noise-induced gaps to some extent. In the current prototype software, we focused on the proof of the concept, using a noisy simulated tomogram with a known ground truth that closely mimics the appearance of an experimental map. We achieved an almost perfect precision score of 0.999, but this success came at the expense of a lower recall score 0.462 due to false negatives. We discuss the dependencies as well as the limitations of the current filament merging that need to be overcome to achieve a higher recall score in the future.","PeriodicalId":210337,"journal":{"name":"2022 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-12-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122209366","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}
Pub Date : 2022-12-06DOI: 10.1109/BIBM55620.2022.9995198
Liang Zhao, Yu-Hsiang Shao, Chaoran Jia, Jiajun Ma
In order to better explore the evolution process of lung nodules in lung cancer patients, we collect lung CT data at multiple time points of lung cancer patients, track and mark the CT positions of the same lung nodules in lung cancer patients at different time points, and make time-series CT data sets of lung cancer patients. After that, 3D-UNet model is used to detect lung nodules on our data set. Experiment proves the effectiveness and availability of the data set, and also proved that the image data at multiple time points could improve the accuracy of the model’s identification of lung nodules.
{"title":"Time-series lung cancer CT dataset","authors":"Liang Zhao, Yu-Hsiang Shao, Chaoran Jia, Jiajun Ma","doi":"10.1109/BIBM55620.2022.9995198","DOIUrl":"https://doi.org/10.1109/BIBM55620.2022.9995198","url":null,"abstract":"In order to better explore the evolution process of lung nodules in lung cancer patients, we collect lung CT data at multiple time points of lung cancer patients, track and mark the CT positions of the same lung nodules in lung cancer patients at different time points, and make time-series CT data sets of lung cancer patients. After that, 3D-UNet model is used to detect lung nodules on our data set. Experiment proves the effectiveness and availability of the data set, and also proved that the image data at multiple time points could improve the accuracy of the model’s identification of lung nodules.","PeriodicalId":210337,"journal":{"name":"2022 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-12-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120949921","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}
Pub Date : 2022-12-06DOI: 10.1109/BIBM55620.2022.9995187
Rajaram Anantharaman, Anwika Bhandary, Raveesh Nandakumar, R. R. Kumar, Pranav Vajapeyam
Osteoporosis, a chronic disease, can be managed through medication and lifestyle changes if detected early. Therefore, there is need for a cost effective method of screening for osteoporosis. In this paper, we propose a deep learning based implementation for developing an automated computer aided diagnostic (CAD) system that harnesses additional information contained in dental panoramic radiographs to detect a person’s risk for developing osteoporosis. Our proposed method follows a two-step approach. First, we apply deep convolutional neural networks (CNNs) to segment key areas of a panoramic radiograph including the mandible, mental foramen, and the mandibular cortical bone. Second, we follow it up with image processing techniques using OpenCV to calculate the ratio of pixels to help arrive at a key ratio called the Panoramic Mandibular Index (PMI). This ratio is instrumental in determining the risk of bone loss in individuals. When compared to the dental clinicians, our model achieved an F1 score of 0.943 on the test set, whereas the performance of dental clinicians was regarded as the standard with a perfect score. Our paper focuses on automating the measurement of PMI to create a CAD system suitable for routine screening of osteoporosis.
{"title":"Utilizing Deep Learning to Opportunistically Screen for Osteoporosis from Dental Panoramic Radiographs","authors":"Rajaram Anantharaman, Anwika Bhandary, Raveesh Nandakumar, R. R. Kumar, Pranav Vajapeyam","doi":"10.1109/BIBM55620.2022.9995187","DOIUrl":"https://doi.org/10.1109/BIBM55620.2022.9995187","url":null,"abstract":"Osteoporosis, a chronic disease, can be managed through medication and lifestyle changes if detected early. Therefore, there is need for a cost effective method of screening for osteoporosis. In this paper, we propose a deep learning based implementation for developing an automated computer aided diagnostic (CAD) system that harnesses additional information contained in dental panoramic radiographs to detect a person’s risk for developing osteoporosis. Our proposed method follows a two-step approach. First, we apply deep convolutional neural networks (CNNs) to segment key areas of a panoramic radiograph including the mandible, mental foramen, and the mandibular cortical bone. Second, we follow it up with image processing techniques using OpenCV to calculate the ratio of pixels to help arrive at a key ratio called the Panoramic Mandibular Index (PMI). This ratio is instrumental in determining the risk of bone loss in individuals. When compared to the dental clinicians, our model achieved an F1 score of 0.943 on the test set, whereas the performance of dental clinicians was regarded as the standard with a perfect score. Our paper focuses on automating the measurement of PMI to create a CAD system suitable for routine screening of osteoporosis.","PeriodicalId":210337,"journal":{"name":"2022 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-12-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116069061","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}
Pub Date : 2022-12-06DOI: 10.1109/BIBM55620.2022.9995658
Weipeng Lv, Changkun Jiang, Jianqiang Li
The discovery of protein-protein interaction sites (PPIs) is vital for exploring the principle of PPI and understanding the nature of life activities. Developing computational approaches to predict PPIs can effectively compensate for the shortcomings of biological experiments, which are mostly time-consuming and vulnerable to noise. In recent years, deep learning has been used to develop PPIs prediction models. Most of them consider the contextual information of the target amino acid residues and use a local protein sequence to represent the targets. However, the traditional deep-learning techniques, e.g., deep neural networks (DNNs) and convolutional neural networks (CNNs), disregard the important spatial hierarchies contained in the features of protein sequences, leading to their failure to effectively distinguish the interaction sites from different residue regions. In this work, we design MSE-CapsPPISP, a new deep-learning model to address the PPIs prediction with spatial hierarchies. The key idea of MSE-CapsPPISP is to take into account the hierarchical relationships between the features of protein sequences. We characterize the hierarchical relationships by designing a tailored Capsule Network (CapsNet) model, which is a novel type of neural network with vector neurons. Moreover, to make the network representation more robust, MSE-CapsPPISP uses multi-scale CNNs to extract multi-scale features of protein sequences and Squeeze-and-Excitation blocks to recalibrate the features. The validation results show that our MSE-CapsPPISP outperforms the baseline CNNs-based architecture DeepPPISP and other competing schemes in the PPIs prediction task.
{"title":"MSE-CapsPPISP: Spatial Hierarchical Protein-Protein Interaction Sites Prediction Using Squeeze-and-Excitation Capsule Networks","authors":"Weipeng Lv, Changkun Jiang, Jianqiang Li","doi":"10.1109/BIBM55620.2022.9995658","DOIUrl":"https://doi.org/10.1109/BIBM55620.2022.9995658","url":null,"abstract":"The discovery of protein-protein interaction sites (PPIs) is vital for exploring the principle of PPI and understanding the nature of life activities. Developing computational approaches to predict PPIs can effectively compensate for the shortcomings of biological experiments, which are mostly time-consuming and vulnerable to noise. In recent years, deep learning has been used to develop PPIs prediction models. Most of them consider the contextual information of the target amino acid residues and use a local protein sequence to represent the targets. However, the traditional deep-learning techniques, e.g., deep neural networks (DNNs) and convolutional neural networks (CNNs), disregard the important spatial hierarchies contained in the features of protein sequences, leading to their failure to effectively distinguish the interaction sites from different residue regions. In this work, we design MSE-CapsPPISP, a new deep-learning model to address the PPIs prediction with spatial hierarchies. The key idea of MSE-CapsPPISP is to take into account the hierarchical relationships between the features of protein sequences. We characterize the hierarchical relationships by designing a tailored Capsule Network (CapsNet) model, which is a novel type of neural network with vector neurons. Moreover, to make the network representation more robust, MSE-CapsPPISP uses multi-scale CNNs to extract multi-scale features of protein sequences and Squeeze-and-Excitation blocks to recalibrate the features. The validation results show that our MSE-CapsPPISP outperforms the baseline CNNs-based architecture DeepPPISP and other competing schemes in the PPIs prediction task.","PeriodicalId":210337,"journal":{"name":"2022 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-12-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115621393","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}
Pub Date : 2022-12-06DOI: 10.1109/BIBM55620.2022.9995254
Dongfang Shen, Ming Wu, Song Zheng, Jianhui Chen, Yijiang Chen, Yinran Chen, Xióngbiao Luó
Unsupervised domain adaptation is to transfer knowledge from a well-annotated source domain and learn an accurate classifier for an unlabeled target domain, which is particularly useful in multimodal medical image processing. Currently available adaptation approaches strongly reduce the domain bias or inconsistency in the latent space, deteriorating inherent data structures. To appropriately leverage the reduction of the domain discrepancy and the maintenance of the intrinsic structure, this paper proposes a dual U-DenseTransformer generation domain adaptation framework to bridge the gap between source and target domains and achieve translation. Specifically, we create a DenseTransformer with multi-head attention embedded in U-shape network to establish a dual-generator strategy, which is further enhanced by a new hybrid loss function and an edge-aware mechanism that preserve inherent data structure consistent. We apply our proposed method to medical image segmentation, with the experimental results showing that it works more effective and stable than currently available approaches. Particularly, the dice similarity was improved from 79.3% to 82.8%, while the average symmetric surface distance was reduced from 2.5 to 1.9.
{"title":"Unsupervised Domain Adaptation with Dual U-DenseTransformer Generation","authors":"Dongfang Shen, Ming Wu, Song Zheng, Jianhui Chen, Yijiang Chen, Yinran Chen, Xióngbiao Luó","doi":"10.1109/BIBM55620.2022.9995254","DOIUrl":"https://doi.org/10.1109/BIBM55620.2022.9995254","url":null,"abstract":"Unsupervised domain adaptation is to transfer knowledge from a well-annotated source domain and learn an accurate classifier for an unlabeled target domain, which is particularly useful in multimodal medical image processing. Currently available adaptation approaches strongly reduce the domain bias or inconsistency in the latent space, deteriorating inherent data structures. To appropriately leverage the reduction of the domain discrepancy and the maintenance of the intrinsic structure, this paper proposes a dual U-DenseTransformer generation domain adaptation framework to bridge the gap between source and target domains and achieve translation. Specifically, we create a DenseTransformer with multi-head attention embedded in U-shape network to establish a dual-generator strategy, which is further enhanced by a new hybrid loss function and an edge-aware mechanism that preserve inherent data structure consistent. We apply our proposed method to medical image segmentation, with the experimental results showing that it works more effective and stable than currently available approaches. Particularly, the dice similarity was improved from 79.3% to 82.8%, while the average symmetric surface distance was reduced from 2.5 to 1.9.","PeriodicalId":210337,"journal":{"name":"2022 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-12-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121195349","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}
Pub Date : 2022-12-06DOI: 10.1109/BIBM55620.2022.9994864
Shuting Sun, Chang Yan, Juntong Lyu, Yueran Xin, Jieyuan Zheng, Zhaolong Yu, B. Hu
Neural circuit dysfunction underlies the biological mechanisms of major depressive disorder (MDD). However, little is known about how the brain’s dynamic connectomes differentiate between depressed patients and normal controls. As a result, we collected resting-state Electroencephalography from 16 MDD patients and 16 controls using 128-electrode geodesic sensor net. Static and dynamic network metrics were later applied to explore the abnormal topological structure of MDD patients and identify them from normal controls using traditional machine learning algorithms with feature selection methods. Results showed that the MDD tend to have a more randomized formation both in static and dynamic network. We also found that the combined static-dynamic feature set usually outperformed others with a highest accuracy of 79.25% under delta band. Lower frequency band (delta, theta) showed relatively better outcomes compared to higher frequency band (alpha, beta). It also indicate the role of functional segregation features as a potential biomarker for depression. In conclusion, neuropathological mechanism of depression may be more objectively quantified and evaluated from the perspective of combining static and dynamic network.
{"title":"EEG Based Depression Recognition by Employing Static and Dynamic Network Metrics","authors":"Shuting Sun, Chang Yan, Juntong Lyu, Yueran Xin, Jieyuan Zheng, Zhaolong Yu, B. Hu","doi":"10.1109/BIBM55620.2022.9994864","DOIUrl":"https://doi.org/10.1109/BIBM55620.2022.9994864","url":null,"abstract":"Neural circuit dysfunction underlies the biological mechanisms of major depressive disorder (MDD). However, little is known about how the brain’s dynamic connectomes differentiate between depressed patients and normal controls. As a result, we collected resting-state Electroencephalography from 16 MDD patients and 16 controls using 128-electrode geodesic sensor net. Static and dynamic network metrics were later applied to explore the abnormal topological structure of MDD patients and identify them from normal controls using traditional machine learning algorithms with feature selection methods. Results showed that the MDD tend to have a more randomized formation both in static and dynamic network. We also found that the combined static-dynamic feature set usually outperformed others with a highest accuracy of 79.25% under delta band. Lower frequency band (delta, theta) showed relatively better outcomes compared to higher frequency band (alpha, beta). It also indicate the role of functional segregation features as a potential biomarker for depression. In conclusion, neuropathological mechanism of depression may be more objectively quantified and evaluated from the perspective of combining static and dynamic network.","PeriodicalId":210337,"journal":{"name":"2022 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-12-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121207013","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}
Pub Date : 2022-12-06DOI: 10.1109/BIBM55620.2022.9995287
Jiawei Li, Fan Yang, Fang Wang, Yu Rong, P. Zhao, Shizhan Chen, Jianhua Yao, Jijun Tang, Fei Guo
Inferring gene regulatory networks based on single-cell transcriptomes is critical for systematically understanding cell-specific regulatory networks and discovering drug targets in tumor cells. Here we show that existing methods mainly perform co-expression analysis and apply the image-based model to deal with the non-euclidean scRNA-seq data, which may not reasonably handle the dropout problem and not fully take advantage of the validated gene regulatory topology. We propose a graph-based end-to-end deep learning model for GRN inference (GRNInfer) with the help of known regulatory relations through transductive learning. The robustness and superiority of the model are demonstrated by comparative experiments.
{"title":"Integrating Prior Knowledge with Graph Encoder for Gene Regulatory Inference from Single-cell RNA-Seq Data","authors":"Jiawei Li, Fan Yang, Fang Wang, Yu Rong, P. Zhao, Shizhan Chen, Jianhua Yao, Jijun Tang, Fei Guo","doi":"10.1109/BIBM55620.2022.9995287","DOIUrl":"https://doi.org/10.1109/BIBM55620.2022.9995287","url":null,"abstract":"Inferring gene regulatory networks based on single-cell transcriptomes is critical for systematically understanding cell-specific regulatory networks and discovering drug targets in tumor cells. Here we show that existing methods mainly perform co-expression analysis and apply the image-based model to deal with the non-euclidean scRNA-seq data, which may not reasonably handle the dropout problem and not fully take advantage of the validated gene regulatory topology. We propose a graph-based end-to-end deep learning model for GRN inference (GRNInfer) with the help of known regulatory relations through transductive learning. The robustness and superiority of the model are demonstrated by comparative experiments.","PeriodicalId":210337,"journal":{"name":"2022 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-12-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116542202","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}
Pub Date : 2022-12-06DOI: 10.1109/BIBM55620.2022.9994976
Zhi-Yuan Li, Ying-Lian Gao, Zhen Niu, Shasha Yuan, C. Zheng, Jin-Xing Liu
Ensemble learning is to train and combine multiple learners to complete the corresponding learning tasks. It can improve the stability of the overall model, and a good ensemble method can further improve the accuracy of the model. At the same time, as one of the outstanding representatives of machine learning, Extreme Learning Machine has attracted the continuous attention of experts and scholars. to get a better representation of the feature space, we extend the Gaussian kernel in the kernel risk-sensitive loss and propose a Kernel Risk-Sensitive Loss of q-Gaussian kernel and Hyper-graph Regularized Extreme Learning Machine method. Since the contingency in the ELM training process cannot be completely avoided, the stability of most ELM methods is affected to some extent. What’s more, we introduce the voting mechanism and a new ELM classification model named Kernel Risk-Sensitive Loss of q-Gaussian kernel and Hyper-graph Regularized Integrated Extreme Learning Machine based on Voting Mechanism is proposed. It improves the stability of the model through the idea of ensemble learning. We apply the new model on six real data sets, and through observation and analysis of experimental results, we find that the new model has certain competitiveness, especially in classification accuracy and stability.
{"title":"An integrated Extreme learning machine based on kernel risk-sensitive loss of q-Gaussian and voting mechanism for sample classification","authors":"Zhi-Yuan Li, Ying-Lian Gao, Zhen Niu, Shasha Yuan, C. Zheng, Jin-Xing Liu","doi":"10.1109/BIBM55620.2022.9994976","DOIUrl":"https://doi.org/10.1109/BIBM55620.2022.9994976","url":null,"abstract":"Ensemble learning is to train and combine multiple learners to complete the corresponding learning tasks. It can improve the stability of the overall model, and a good ensemble method can further improve the accuracy of the model. At the same time, as one of the outstanding representatives of machine learning, Extreme Learning Machine has attracted the continuous attention of experts and scholars. to get a better representation of the feature space, we extend the Gaussian kernel in the kernel risk-sensitive loss and propose a Kernel Risk-Sensitive Loss of q-Gaussian kernel and Hyper-graph Regularized Extreme Learning Machine method. Since the contingency in the ELM training process cannot be completely avoided, the stability of most ELM methods is affected to some extent. What’s more, we introduce the voting mechanism and a new ELM classification model named Kernel Risk-Sensitive Loss of q-Gaussian kernel and Hyper-graph Regularized Integrated Extreme Learning Machine based on Voting Mechanism is proposed. It improves the stability of the model through the idea of ensemble learning. We apply the new model on six real data sets, and through observation and analysis of experimental results, we find that the new model has certain competitiveness, especially in classification accuracy and stability.","PeriodicalId":210337,"journal":{"name":"2022 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-12-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116742978","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}