Pub Date : 2022-12-06DOI: 10.1109/BIBM55620.2022.9995146
Ying Lv, Xiaodong Yue, Zhikang Xu, Yufei Chen, Zihao Li
Pre-training deep learning models on ImageNet and transferring the models to medical image applications facilitate to improve the medical image analysis and reduce the need for labeled medical image data. However, some images from ImageNet may be fundamentally different from medical images in feature representation and lead to the negative transfer effects. To deal with this issue, we propose a novel strategy based on evidence theory to select reliable instances from ImageNet for medical image domain adaptation. Specifically, we formulate an evidential mass function to measure the ignorance and reliability of the images from ImageNet with respect to the classification tasks of medical images. Through selecting reliable instances with low ignorance degree from ImageNet, we can enhance the transfer performances of deep neural networks in medical image domain adaptation. Moreover, the proposed data selection strategy is independent of specific learning algorithm and can be viewed as a common preprocessing technique. Numerical experiments on tomography images, X-Ray images, and ultrasound images are given to comprehensively demonstrate the effectiveness of the selection strategy.
{"title":"Selecting Reliable Instances from ImageNet for Medical Image Domain Adaptation","authors":"Ying Lv, Xiaodong Yue, Zhikang Xu, Yufei Chen, Zihao Li","doi":"10.1109/BIBM55620.2022.9995146","DOIUrl":"https://doi.org/10.1109/BIBM55620.2022.9995146","url":null,"abstract":"Pre-training deep learning models on ImageNet and transferring the models to medical image applications facilitate to improve the medical image analysis and reduce the need for labeled medical image data. However, some images from ImageNet may be fundamentally different from medical images in feature representation and lead to the negative transfer effects. To deal with this issue, we propose a novel strategy based on evidence theory to select reliable instances from ImageNet for medical image domain adaptation. Specifically, we formulate an evidential mass function to measure the ignorance and reliability of the images from ImageNet with respect to the classification tasks of medical images. Through selecting reliable instances with low ignorance degree from ImageNet, we can enhance the transfer performances of deep neural networks in medical image domain adaptation. Moreover, the proposed data selection strategy is independent of specific learning algorithm and can be viewed as a common preprocessing technique. Numerical experiments on tomography images, X-Ray images, and ultrasound images are given to comprehensively demonstrate the effectiveness of the selection strategy.","PeriodicalId":210337,"journal":{"name":"2022 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116826772","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/BIBM55620.2022.9995372
Mei Li, Sihan Xu, Xiangrui Cai, Zhong Zhang, Hua Ji
Effective drug-target binding affinity (DTA) prediction is essential for drug discovery and development. The development of machine learning techniques considerably advances it. However, the cold-start problems in DTA prediction are still under-explored, which significantly degrades prediction performances on novel drugs and novel targets. In this paper, we propose a contrastive meta-learning (CML) framework to address these issues. We define drug-anchored tasks and target-anchored tasks, which enables the employment of meta-learning to accumulate common knowledge from various tasks so as to adapt to new tasks faster and better. Besides, we utilize a task inequality loss to measure task disparities and enhance model sensitivities to new tasks. We also propose a contrastive learning block (CLB) to explore correlations among drug-target pairs across tasks, which facilitates DTA prediction performance improvements. We compare CML with various baselines on two benchmarks and comparison results show that CML outperforms or achieves competitive results to its competitors.
{"title":"Contrastive Meta-Learning for Drug-Target Binding Affinity Prediction","authors":"Mei Li, Sihan Xu, Xiangrui Cai, Zhong Zhang, Hua Ji","doi":"10.1109/BIBM55620.2022.9995372","DOIUrl":"https://doi.org/10.1109/BIBM55620.2022.9995372","url":null,"abstract":"Effective drug-target binding affinity (DTA) prediction is essential for drug discovery and development. The development of machine learning techniques considerably advances it. However, the cold-start problems in DTA prediction are still under-explored, which significantly degrades prediction performances on novel drugs and novel targets. In this paper, we propose a contrastive meta-learning (CML) framework to address these issues. We define drug-anchored tasks and target-anchored tasks, which enables the employment of meta-learning to accumulate common knowledge from various tasks so as to adapt to new tasks faster and better. Besides, we utilize a task inequality loss to measure task disparities and enhance model sensitivities to new tasks. We also propose a contrastive learning block (CLB) to explore correlations among drug-target pairs across tasks, which facilitates DTA prediction performance improvements. We compare CML with various baselines on two benchmarks and comparison results show that CML outperforms or achieves competitive results to its competitors.","PeriodicalId":210337,"journal":{"name":"2022 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129574226","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/BIBM55620.2022.9995038
Teng Teng, Xuebo Li
Objective: To explore and analyze the experience and potential rules of Chinese medical master Li Zhenhua in the treatment of spleen and stomach diseases [1]. Methods: The medical records of 75 patients with spleen and stomach diseases treated by Mr. Li Zhenhua have been collected, and the data of initial diagnosis were entered into the database. After data standardization, IBM SPSS Statistics 26 and IBM SPSS Modeler 18 statistical software were used to process the data and explore the internal diagnosis and treatment rules. Results: Among the 75 medical records of spleen and stomach diseases, the most common syndromes were spleen and stomach deficiency cold syndrome, liver depression and spleen deficiency syndrome, spleen and kidney Yang deficiency syndrome, commonly used herbs are licorice, atractylodes macrocephala, poria, tangerine peel, etc. The commonly used treatment rules were invigorated the spleen, harmonized the stomach and soothed the liver. The prescriptions used more often were Xiangsha Wenzhong Decoction, Xiangsha Liujunzi Decoction, Buzhong Yiqi Decoction, Dingxiang Shidi Decoction and Chaihu Shugan Decoction. The commonly used additional herbs were Lindera aggregata, xiangfu, cassia twig and so on. Among all the used herbs, the herb combinations with the highest support were atractylodes macrocephala → poria and poria → atractylodes macrocephala, and the herb combinations with the highest confidence were poria → atractylodes macrocephala, poria and licorice → atractylodes macrocephala. Among the treatment rules, the combination of treatment rules with the highest support were harmonized the stomach → invigorated the spleen and invigorated the spleen → harmonized the stomach, the combination of treatment rules with the highest confidence was tonified Qi → invigorated the spleen. Among all the additional herbs, the herb combination with the highest support and the highest confidence was malt → hawthorn. Conclusion: The main treatment rules of Mr. Li Zhenhua in treating spleen and stomach diseases were invigorated the spleen, soothed the liver and harmonized the stomach. Xiangsha Wenzhong Decoction is commonly used in the treatment of spleen and stomach diseases, and the syndrome of deficiency and cold is common, therefore, Mr. Li Zhenhua often adds warm herbs combined with tonic herbs to achieve the purpose of treatment.
{"title":"Data Mining and Analysis of The Medical Records of Chinese Medical Master Li Zhenhua in The Treatment of Spleen and Stomach Diseases","authors":"Teng Teng, Xuebo Li","doi":"10.1109/BIBM55620.2022.9995038","DOIUrl":"https://doi.org/10.1109/BIBM55620.2022.9995038","url":null,"abstract":"Objective: To explore and analyze the experience and potential rules of Chinese medical master Li Zhenhua in the treatment of spleen and stomach diseases [1]. Methods: The medical records of 75 patients with spleen and stomach diseases treated by Mr. Li Zhenhua have been collected, and the data of initial diagnosis were entered into the database. After data standardization, IBM SPSS Statistics 26 and IBM SPSS Modeler 18 statistical software were used to process the data and explore the internal diagnosis and treatment rules. Results: Among the 75 medical records of spleen and stomach diseases, the most common syndromes were spleen and stomach deficiency cold syndrome, liver depression and spleen deficiency syndrome, spleen and kidney Yang deficiency syndrome, commonly used herbs are licorice, atractylodes macrocephala, poria, tangerine peel, etc. The commonly used treatment rules were invigorated the spleen, harmonized the stomach and soothed the liver. The prescriptions used more often were Xiangsha Wenzhong Decoction, Xiangsha Liujunzi Decoction, Buzhong Yiqi Decoction, Dingxiang Shidi Decoction and Chaihu Shugan Decoction. The commonly used additional herbs were Lindera aggregata, xiangfu, cassia twig and so on. Among all the used herbs, the herb combinations with the highest support were atractylodes macrocephala → poria and poria → atractylodes macrocephala, and the herb combinations with the highest confidence were poria → atractylodes macrocephala, poria and licorice → atractylodes macrocephala. Among the treatment rules, the combination of treatment rules with the highest support were harmonized the stomach → invigorated the spleen and invigorated the spleen → harmonized the stomach, the combination of treatment rules with the highest confidence was tonified Qi → invigorated the spleen. Among all the additional herbs, the herb combination with the highest support and the highest confidence was malt → hawthorn. Conclusion: The main treatment rules of Mr. Li Zhenhua in treating spleen and stomach diseases were invigorated the spleen, soothed the liver and harmonized the stomach. Xiangsha Wenzhong Decoction is commonly used in the treatment of spleen and stomach diseases, and the syndrome of deficiency and cold is common, therefore, Mr. Li Zhenhua often adds warm herbs combined with tonic herbs to achieve the purpose of treatment.","PeriodicalId":210337,"journal":{"name":"2022 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)","volume":"108 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129587515","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/BIBM55620.2022.9994981
Haoran Zheng, Qiu Xiao, Jiancheng Zhong
Extensive research evidence shows that variation and dysregulation of microRNAs(miRNAs) are important causes of disease, and therefore the study of miRNA-disease associations has important theoretical and applied implications in the field of human disease research and treatment. Based on the time and cost of validating miRNA-disease associations in traditional medicine clinical experiments, using multiple biological datasets to predict potential miRNA-disease associations (MDAs) has become a hot topic in the field of biological research in recent years. This paper develops a novel model of MDA-RGCN based on a multi-channel graph convolutional network and graph attention for MDAs prediction. Based on graph theory, this study treats MDAs prediction as a node classification task. To learn the topology and various interactions between feature graph nodes of various strengths, we employ two independent graph attention networks, which increases training efficiency and accuracy. In order to learn information that is shared by both graphs, we employ a GCN with a shared weight matrix simultaneously. Comprehensive experiments reveal that the prediction performance of MDA-RGCN excels other more sophisticated models for MDAs prediction. Furthermore, we further confirmed the predictive ability of MDA-RGCN to identify potential disease-related miRNAs by selecting two human diseases for case study.
{"title":"Predicting miRNA-disease associations via multi-channel graph convolutional networks","authors":"Haoran Zheng, Qiu Xiao, Jiancheng Zhong","doi":"10.1109/BIBM55620.2022.9994981","DOIUrl":"https://doi.org/10.1109/BIBM55620.2022.9994981","url":null,"abstract":"Extensive research evidence shows that variation and dysregulation of microRNAs(miRNAs) are important causes of disease, and therefore the study of miRNA-disease associations has important theoretical and applied implications in the field of human disease research and treatment. Based on the time and cost of validating miRNA-disease associations in traditional medicine clinical experiments, using multiple biological datasets to predict potential miRNA-disease associations (MDAs) has become a hot topic in the field of biological research in recent years. This paper develops a novel model of MDA-RGCN based on a multi-channel graph convolutional network and graph attention for MDAs prediction. Based on graph theory, this study treats MDAs prediction as a node classification task. To learn the topology and various interactions between feature graph nodes of various strengths, we employ two independent graph attention networks, which increases training efficiency and accuracy. In order to learn information that is shared by both graphs, we employ a GCN with a shared weight matrix simultaneously. Comprehensive experiments reveal that the prediction performance of MDA-RGCN excels other more sophisticated models for MDAs prediction. Furthermore, we further confirmed the predictive ability of MDA-RGCN to identify potential disease-related miRNAs by selecting two human diseases for case study.","PeriodicalId":210337,"journal":{"name":"2022 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)","volume":"114 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124562163","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/BIBM55620.2022.9995553
Ling-ling Zhu, Kai Zheng, Guihua Duan, Jianxin Wang
Receptor-binding is the first step of viral infection. Discovering potential virus-receptor interactions may give insight into potential strategies for treating viral infectious diseases. Most of computational methods for the virus-receptor interaction prediction are mainly based on sequence information. They neither makes effective use of structure information nor effectively handles with missing values of multiple similarities. In addition, the Link Prediction via linear optimization (LP) only uses contribution of neighbors of a node and ignores contribution of neighbors of another node on the network link. In this article, we present a virus-receptor interaction prediction method (MVLP) based on Multi-View learning and LP via contributions of all neighbors of two nodes on the network link. First, missing values of the receptor secondary structure similarity, the receptor conserved domain secondary structure similarity, the viral protein secondary structure similarity, the viral protein sequence similarity and the viral genome sequence similarity are updated by the gaussian radial basis function (GRB). To improve these similarities, we fuse updated and initial values of each similarity with multi-view learning, respectively. Next, three virus values and receptor similarities are integrated into the comprehensive virus and receptor similarity by the averaging method, respectively. Finally, LP based on contribution of neighbors of two nodes is presented for the virus-receptor interaction prediction. To evaluate the ability of MVLP, we compare MVLP with four related methods in 10 fold Cross-Validation (10CV). Computational results indicate that an average Area Under Curve (AUC) values of MVLP on viralReceptor sup and viralReceptor are 0.9427 and 0.9444, respectively, which are superior to other related methods. Furthermore, a case study also demonstrates the ability of MVLP in practice.
受体结合是病毒感染的第一步。发现潜在的病毒-受体相互作用可能会为治疗病毒性传染病的潜在策略提供见解。大多数病毒-受体相互作用预测的计算方法主要基于序列信息。它们既没有有效地利用结构信息，也没有有效地处理多重相似度缺失值。此外，线性优化的链路预测(Link Prediction via linear optimization, LP)只利用一个节点的邻居的贡献，而忽略了网络链路上另一个节点的邻居的贡献。本文提出了一种基于多视图学习和LP的病毒-受体相互作用预测方法(MVLP)，该方法利用网络链路上两个节点的所有邻居的贡献。首先，利用高斯径向基函数(GRB)对缺失的受体二级结构相似度、受体保守域二级结构相似度、病毒蛋白二级结构相似度、病毒蛋白序列相似度和病毒基因组序列相似度进行更新。为了提高这些相似度，我们将每个相似度的更新值和初始值分别融合到多视图学习中。接下来，将三个病毒值和受体相似度分别用平均法整合到综合病毒和受体相似度中。最后，提出了基于两个节点邻居贡献的LP预测病毒与受体相互作用。为了评估MVLP的能力，我们在10倍交叉验证(10CV)中将MVLP与四种相关方法进行了比较。计算结果表明，MVLP在病毒受体sup和病毒受体上的平均AUC值分别为0.9427和0.9444，优于其他相关方法。最后，通过实例验证了该方法在实际应用中的能力。
{"title":"Prediction of virus-receptor interactions based on multi-view learning and link prediction","authors":"Ling-ling Zhu, Kai Zheng, Guihua Duan, Jianxin Wang","doi":"10.1109/BIBM55620.2022.9995553","DOIUrl":"https://doi.org/10.1109/BIBM55620.2022.9995553","url":null,"abstract":"Receptor-binding is the first step of viral infection. Discovering potential virus-receptor interactions may give insight into potential strategies for treating viral infectious diseases. Most of computational methods for the virus-receptor interaction prediction are mainly based on sequence information. They neither makes effective use of structure information nor effectively handles with missing values of multiple similarities. In addition, the Link Prediction via linear optimization (LP) only uses contribution of neighbors of a node and ignores contribution of neighbors of another node on the network link. In this article, we present a virus-receptor interaction prediction method (MVLP) based on Multi-View learning and LP via contributions of all neighbors of two nodes on the network link. First, missing values of the receptor secondary structure similarity, the receptor conserved domain secondary structure similarity, the viral protein secondary structure similarity, the viral protein sequence similarity and the viral genome sequence similarity are updated by the gaussian radial basis function (GRB). To improve these similarities, we fuse updated and initial values of each similarity with multi-view learning, respectively. Next, three virus values and receptor similarities are integrated into the comprehensive virus and receptor similarity by the averaging method, respectively. Finally, LP based on contribution of neighbors of two nodes is presented for the virus-receptor interaction prediction. To evaluate the ability of MVLP, we compare MVLP with four related methods in 10 fold Cross-Validation (10CV). Computational results indicate that an average Area Under Curve (AUC) values of MVLP on viralReceptor sup and viralReceptor are 0.9427 and 0.9444, respectively, which are superior to other related methods. Furthermore, a case study also demonstrates the ability of MVLP in practice.","PeriodicalId":210337,"journal":{"name":"2022 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129902971","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/BIBM55620.2022.9994941
Jiyao Liu, Hao Wu, Li Zhang
Electroencephalogram (EEG) emotion recognition has become a research focus in the field of human-computer interaction (HCI). However, the process of EEG signal collection requires lots of expertise, which makes the amount of labeled EEG data very limited. It constrains the performance of supervised methods which require large amounts of annotated data in some sense. Self-supervised learning paradigm, which aims to train models that do not require any labeled samples can make full use of a large amount of unlabeled EEG samples. But a drawback is that they fall short of learning class discriminative sample representations since no labeled information is utilized during training. To solve the above problem, we propose a semi-supervised model, named consistency regularization enhanced graph attention network (CR-GAT) for EEG emotion recognition. The CR-GAT mainly consists of three modules, namely the feature extraction and fusion (FEF) module, the feature graph building and augment (GBA) module as well as the consistency regularization (CR) module. Specifically, t he F EFm odule is to extract task-specific EEG features and highlight the most valuable features from the EEG signals. The GBA module is to build a sample-related graph representation of the EEG feature set. The CR module, which draws support samples from labeled samples and anchor samples from the entire sample set, intends to minimize the difference between the predicted class distributions from different graphs constructed by multi-views of the sample set to push samples that belong to the same class to be grouped together. We conduct our experiment on three real-world datasets, the experimental results show the method surpasses most of competitive models.
{"title":"CR-GAT: Consistency Regularization Enhanced Graph Attention Network for Semi-supervised EEG Emotion Recognition","authors":"Jiyao Liu, Hao Wu, Li Zhang","doi":"10.1109/BIBM55620.2022.9994941","DOIUrl":"https://doi.org/10.1109/BIBM55620.2022.9994941","url":null,"abstract":"Electroencephalogram (EEG) emotion recognition has become a research focus in the field of human-computer interaction (HCI). However, the process of EEG signal collection requires lots of expertise, which makes the amount of labeled EEG data very limited. It constrains the performance of supervised methods which require large amounts of annotated data in some sense. Self-supervised learning paradigm, which aims to train models that do not require any labeled samples can make full use of a large amount of unlabeled EEG samples. But a drawback is that they fall short of learning class discriminative sample representations since no labeled information is utilized during training. To solve the above problem, we propose a semi-supervised model, named consistency regularization enhanced graph attention network (CR-GAT) for EEG emotion recognition. The CR-GAT mainly consists of three modules, namely the feature extraction and fusion (FEF) module, the feature graph building and augment (GBA) module as well as the consistency regularization (CR) module. Specifically, t he F EFm odule is to extract task-specific EEG features and highlight the most valuable features from the EEG signals. The GBA module is to build a sample-related graph representation of the EEG feature set. The CR module, which draws support samples from labeled samples and anchor samples from the entire sample set, intends to minimize the difference between the predicted class distributions from different graphs constructed by multi-views of the sample set to push samples that belong to the same class to be grouped together. We conduct our experiment on three real-world datasets, the experimental results show the method surpasses most of competitive models.","PeriodicalId":210337,"journal":{"name":"2022 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130681036","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/BIBM55620.2022.9995677
Tao Peng, Yiyun Wu, Jing Zhao, Bo Zhang, Jin Wang, Jing Cai
Accurate segmentation of the prostate is important to image-guided prostate biopsy and brachytherapy treatment planning. However, the incompleteness of prostate boundary increases the challenges in the automatic ultrasound prostate segmentation task. In this work, an automatic coarse-to-fine framework for prostate segmentation was developed and tested. Our framework has four metrics: first, it combines the ability of deep learning model to automatically locate the prostate and integrates the characteristics of principal curve that can automatically fit the data center for refinement. Second, to well balance the accuracy and efficiency of our method, we proposed an intelligent determination of the data radius algorithm-based modified polygon tracking method. Third, we modified the traditional quantum evolution network by adding the numerous-operator scheme and global optimum search scheme for ensuring population diversity and achieving the optimal model parameters. Fourth, we found a suitable mathematical function expressed by the parameters of the machine learning model to smooth the contour of the prostate. Results on the multiple datasets demonstrate that our method has good segmentation performance.
{"title":"Explainability-guided Mathematical Model-Based Segmentation of Transrectal Ultrasound Images for Prostate Brachytherapy","authors":"Tao Peng, Yiyun Wu, Jing Zhao, Bo Zhang, Jin Wang, Jing Cai","doi":"10.1109/BIBM55620.2022.9995677","DOIUrl":"https://doi.org/10.1109/BIBM55620.2022.9995677","url":null,"abstract":"Accurate segmentation of the prostate is important to image-guided prostate biopsy and brachytherapy treatment planning. However, the incompleteness of prostate boundary increases the challenges in the automatic ultrasound prostate segmentation task. In this work, an automatic coarse-to-fine framework for prostate segmentation was developed and tested. Our framework has four metrics: first, it combines the ability of deep learning model to automatically locate the prostate and integrates the characteristics of principal curve that can automatically fit the data center for refinement. Second, to well balance the accuracy and efficiency of our method, we proposed an intelligent determination of the data radius algorithm-based modified polygon tracking method. Third, we modified the traditional quantum evolution network by adding the numerous-operator scheme and global optimum search scheme for ensuring population diversity and achieving the optimal model parameters. Fourth, we found a suitable mathematical function expressed by the parameters of the machine learning model to smooth the contour of the prostate. Results on the multiple datasets demonstrate that our method has good segmentation performance.","PeriodicalId":210337,"journal":{"name":"2022 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123957870","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/BIBM55620.2022.9995444
Kaiwen Yang, Aiga Suzuki, Jiaxing Ye, H. Nosato, Ayumi Izumori, H. Sakanashi
Segmentation and classification a re h ighly correlated tasks in tumor detection from breast ultrasound images. Recent studies have successfully applied multi-task learning to breast ultrasound image analysis to explore the correlation between tasks. However, there exists potential inconsistency between individual tasks that critically affect the overall performance of breast ultrasound image analysis. Therefore, this study designs a consistency branch for harmonizing the segmentation and classification t ask 0 ptimization. T he c onsistency b ranch characterizes the outputs of individual task-specific models to maintain consistency during training, thereby generating highly consistent results. Specifically, the consistency branch outputs a consistency probability while determining the inconsistency types predicted by both tasks. Subsequently, the segmentation and classification loss weights are reconciled using consistency probabilities based on the inconsistent prediction behavior for each sample, thus constraining the two tasks to produce consistent predictions close to the ground truth. The evaluation using private and public breast ultrasound image datasets indicates that the proposed method can effectively remedy the inconsistent predictions between tasks for improved computerized breast ultrasound image analysis.
{"title":"Multi-task Learning with Consistent Prediction for Efficient Breast Ultrasound Tumor Detection","authors":"Kaiwen Yang, Aiga Suzuki, Jiaxing Ye, H. Nosato, Ayumi Izumori, H. Sakanashi","doi":"10.1109/BIBM55620.2022.9995444","DOIUrl":"https://doi.org/10.1109/BIBM55620.2022.9995444","url":null,"abstract":"Segmentation and classification a re h ighly correlated tasks in tumor detection from breast ultrasound images. Recent studies have successfully applied multi-task learning to breast ultrasound image analysis to explore the correlation between tasks. However, there exists potential inconsistency between individual tasks that critically affect the overall performance of breast ultrasound image analysis. Therefore, this study designs a consistency branch for harmonizing the segmentation and classification t ask 0 ptimization. T he c onsistency b ranch characterizes the outputs of individual task-specific models to maintain consistency during training, thereby generating highly consistent results. Specifically, the consistency branch outputs a consistency probability while determining the inconsistency types predicted by both tasks. Subsequently, the segmentation and classification loss weights are reconciled using consistency probabilities based on the inconsistent prediction behavior for each sample, thus constraining the two tasks to produce consistent predictions close to the ground truth. The evaluation using private and public breast ultrasound image datasets indicates that the proposed method can effectively remedy the inconsistent predictions between tasks for improved computerized breast ultrasound image analysis.","PeriodicalId":210337,"journal":{"name":"2022 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)","volume":"100 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120964173","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/BIBM55620.2022.9995418
Kai Chen, Yu Liu, Xuqi Wang, Shanwen Zhang, Chuanghui Zhang
This paper aims to automatically diagnose thoracic diseases on chest X-ray (CXR) images using convolutional neural networks (CNN). Most existing approaches typically employ a global learning strategy and use CNN with small convolutional kernels for thoracic disease classification. However, irrelevant noisy regions may affect the global learning strategy; small convolutional kernels can only capture fewer discriminant features. To address the above problems, we construct a multi-feature fusion neural network (MFCNet), which can fully use the global and weighted local features. Specifically, the global features are first generated by the global branch. Weighted local features are generated by multiplying the global feature and the heart-lung region mask identified by the Lung-heart Region Generator (LHRG). At last, the fusion branch integrates the global and weighted local features to complement the lost discriminative feature of the global branch and the local branch, thus enabling a better feature presentation for thoracic disease classification. Extensive experiments on the NIH ChestX-ray 14 dataset demonstrate that the MFCNet model achieves superior performance (average AUC=0.844) compared to state-of-the-art methods. Source code is released in https://github.com/Warrior996/MFCNet.
本文旨在利用卷积神经网络(CNN)对胸部x射线(CXR)图像进行胸部疾病的自动诊断。大多数现有方法通常采用全局学习策略，并使用具有小卷积核的CNN进行胸部疾病分类。然而，不相关的噪声区域可能会影响全局学习策略;较小的卷积核只能捕获较少的判别特征。为了解决上述问题，我们构建了一个多特征融合神经网络(MFCNet)，该网络可以充分利用全局特征和加权的局部特征。具体来说，全局特征首先由全局分支生成。将全局特征与肺脏区域生成器(LHRG)识别的心肺区域掩模相乘，生成加权局部特征。最后，融合分支融合了全局特征和加权局部特征，弥补了全局分支和局部分支缺失的判别特征，使胸椎疾病分类的特征表现更好。在NIH chestx - x - 14数据集上进行的大量实验表明，与最先进的方法相比，MFCNet模型具有优越的性能(平均AUC=0.844)。源代码发布在https://github.com/Warrior996/MFCNet。
{"title":"MFCNet: Multi-Feature Fusion Neural Network for Thoracic Disease Classification","authors":"Kai Chen, Yu Liu, Xuqi Wang, Shanwen Zhang, Chuanghui Zhang","doi":"10.1109/BIBM55620.2022.9995418","DOIUrl":"https://doi.org/10.1109/BIBM55620.2022.9995418","url":null,"abstract":"This paper aims to automatically diagnose thoracic diseases on chest X-ray (CXR) images using convolutional neural networks (CNN). Most existing approaches typically employ a global learning strategy and use CNN with small convolutional kernels for thoracic disease classification. However, irrelevant noisy regions may affect the global learning strategy; small convolutional kernels can only capture fewer discriminant features. To address the above problems, we construct a multi-feature fusion neural network (MFCNet), which can fully use the global and weighted local features. Specifically, the global features are first generated by the global branch. Weighted local features are generated by multiplying the global feature and the heart-lung region mask identified by the Lung-heart Region Generator (LHRG). At last, the fusion branch integrates the global and weighted local features to complement the lost discriminative feature of the global branch and the local branch, thus enabling a better feature presentation for thoracic disease classification. Extensive experiments on the NIH ChestX-ray 14 dataset demonstrate that the MFCNet model achieves superior performance (average AUC=0.844) compared to state-of-the-art methods. Source code is released in https://github.com/Warrior996/MFCNet.","PeriodicalId":210337,"journal":{"name":"2022 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116373556","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/BIBM55620.2022.9995708
Dongning Ma, Rahul Thapa, Xun Jiao
In this paper, we propose MoleHD, an efficient learning model based on brain-inspired hyperdimensional computing (HDC) for molecular property prediction. We develop HDC encoders to project SMILES representation of a molecule into high-dimensional vectors that are used for HDC training and inference. We perform an extensive evaluation using 29 classification tasks from 3 widely-used molecule datasets (Clintox, BBBP, SIDER) under three splits methods (random, scaffold, and stratified). By a comprehensive comparison with 8 existing learning models, we show that MoleHD achieves highest ROC-AUC score on random and scaffold splits on average across 3 datasets and achieve second-highest on stratified split. More importantly, MoleHD achieves such performance with significantly reduced computing cost: no back-propagation needed, only around 10 minutes training time using CPU.MoleHD is open-sourced and available at https://github.com/VU-DETAIL/MoleHD.
{"title":"MoleHD: Efficient Drug Discovery using Brain Inspired Hyperdimensional Computing","authors":"Dongning Ma, Rahul Thapa, Xun Jiao","doi":"10.1109/BIBM55620.2022.9995708","DOIUrl":"https://doi.org/10.1109/BIBM55620.2022.9995708","url":null,"abstract":"In this paper, we propose MoleHD, an efficient learning model based on brain-inspired hyperdimensional computing (HDC) for molecular property prediction. We develop HDC encoders to project SMILES representation of a molecule into high-dimensional vectors that are used for HDC training and inference. We perform an extensive evaluation using 29 classification tasks from 3 widely-used molecule datasets (Clintox, BBBP, SIDER) under three splits methods (random, scaffold, and stratified). By a comprehensive comparison with 8 existing learning models, we show that MoleHD achieves highest ROC-AUC score on random and scaffold splits on average across 3 datasets and achieve second-highest on stratified split. More importantly, MoleHD achieves such performance with significantly reduced computing cost: no back-propagation needed, only around 10 minutes training time using CPU.MoleHD is open-sourced and available at https://github.com/VU-DETAIL/MoleHD.","PeriodicalId":210337,"journal":{"name":"2022 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)","volume":"57 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127771006","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：