Dataset Information

Lung Radiomics Features Selection for COPD Stage Classification Based on Auto-Metric Graph Neural Network.

ABSTRACT: Chronic obstructive pulmonary disease (COPD) is a preventable, treatable, progressive chronic disease characterized by persistent airflow limitation. Patients with COPD deserve special consideration regarding treatment in this fragile population for preclinical health management. Therefore, this paper proposes a novel lung radiomics combination vector generated by a generalized linear model (GLM) and Lasso algorithm for COPD stage classification based on an auto-metric graph neural network (AMGNN) with a meta-learning strategy. Firstly, the parenchyma images were segmented from chest high-resolution computed tomography (HRCT) images by ResU-Net. Second, lung radiomics features are extracted from the parenchyma images by PyRadiomics. Third, a novel lung radiomics combination vector (3 + 106) is constructed by the GLM and Lasso algorithm for determining the radiomics risk factors (K = 3) and radiomics node features (d = 106). Last, the COPD stage is classified based on the AMGNN. The results show that compared with the convolutional neural networks and machine learning models, the AMGNN based on constructed novel lung radiomics combination vector performs best, achieving an accuracy of 0.943, precision of 0.946, recall of 0.943, F1-score of 0.943, and ACU of 0.984. Furthermore, it is found that our method is effective for COPD stage classification.

SUBMITTER: Yang Y

PROVIDER: S-EPMC9600898 | biostudies-literature | 2022 Sep

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Lung Radiomics Features Selection for COPD Stage Classification Based on Auto-Metric Graph Neural Network.

Yang Yingjian Y Wang Shicong S Zeng Nanrong N Duan Wenxin W Chen Ziran Z Liu Yang Y Li Wei W Guo Yingwei Y Chen Huai H Li Xian X Chen Rongchang R Kang Yan Y

Diagnostics (Basel, Switzerland) 20220920 10

Chronic obstructive pulmonary disease (COPD) is a preventable, treatable, progressive chronic disease characterized by persistent airflow limitation. Patients with COPD deserve special consideration regarding treatment in this fragile population for preclinical health management. Therefore, this paper proposes a novel lung radiomics combination vector generated by a generalized linear model (GLM) and Lasso algorithm for COPD stage classification based on an auto-metric graph neural network (AMGN ...[more]

PMID: 36291964

Similar Datasets

Project description:BackgroundsThe respiratory microbiota and radiomics correlate with the disease severity and prognosis of chronic obstructive pulmonary disease (COPD). We aim to characterize the respiratory microbiota and radiomics features of COPD patients and explore the relationship between them.MethodsSputa from stable COPD patients were collected for bacterial 16 S rRNA gene sequencing and fungal Internal Transcribed Spacer (ITS) sequencing. Chest computed tomography (CT) and 3D-CT analysis were conducted for radiomics information, including the percentages of low attenuation area below - 950 Hounsfield Units (LAA%), wall thickness (WT), and intraluminal area (Ai). WT and Ai were adjusted by body surface area (BSA) to WT/[Formula: see text] and Ai/BSA, respectively. Some key pulmonary function indicators were collected, which included forced expiratory volume in one second (FEV1), forced vital capacity (FVC), diffusion lung carbon monoxide (DLco). Differences and correlations of microbiomics with radiomics and clinical indicators between different patient subgroups were assessed.ResultsTwo bacterial clusters dominated by Streptococcus and Rothia were identified. Chao and Shannon indices were higher in the Streptococcus cluster than that in the Rothia cluster. Principal Co-ordinates Analysis (PCoA) indicated significant differences between their community structures. Higher relative abundance of Actinobacteria was detected in the Rothia cluster. Some genera were more common in the Streptococcus cluster, mainly including Leptotrichia, Oribacterium, Peptostreptococcus. Peptostreptococcus was positively correlated with DLco per unit of alveolar volume as a percentage of predicted value (DLco/VA%pred). The patients with past-year exacerbations were more in the Streptococcus cluster. Fungal analysis revealed two clusters dominated by Aspergillus and Candida. Chao and Shannon indices of the Aspergillus cluster were higher than that in the Candida cluster. PCoA showed distinct community compositions between the two clusters. Greater abundance of Cladosporium and Penicillium was found in the Aspergillus cluster. The patients of the Candida cluster had upper FEV1 and FEV1/FVC levels. In radiomics, the patients of the Rothia cluster had higher LAA% and WT/[Formula: see text] than those of the Streptococcus cluster. Haemophilus, Neisseria and Cutaneotrichosporon positively correlated with Ai/BSA, but Cladosporium negatively correlated with Ai/BSA.ConclusionsAmong respiratory microbiota in stable COPD patients, Streptococcus dominance was associated with an increased risk of exacerbation, and Rothia dominance was relevant to worse emphysema and airway lesions. Peptostreptococcus, Haemophilus, Neisseria and Cutaneotrichosporon probably affected COPD progression and potentially could be disease prediction biomarkers.

Project description:BackgroundCancer has been a leading cause of death in the United States with significant health care costs. Accurate prediction of cancers at an early stage and understanding the genomic mechanisms that drive cancer development are vital to the improvement of treatment outcomes and survival rates, thus resulting in significant social and economic impacts. Attempts have been made to classify cancer types with machine learning techniques during the past two decades and deep learning approaches more recently.ResultsIn this paper, we established four models with graph convolutional neural network (GCNN) that use unstructured gene expressions as inputs to classify different tumor and non-tumor samples into their designated 33 cancer types or as normal. Four GCNN models based on a co-expression graph, co-expression+singleton graph, protein-protein interaction (PPI) graph, and PPI+singleton graph have been designed and implemented. They were trained and tested on combined 10,340 cancer samples and 731 normal tissue samples from The Cancer Genome Atlas (TCGA) dataset. The established GCNN models achieved excellent prediction accuracies (89.9-94.7%) among 34 classes (33 cancer types and a normal group). In silico gene-perturbation experiments were performed on four models based on co-expression graph, co-expression+singleton, PPI graph, and PPI+singleton graphs. The co-expression GCNN model was further interpreted to identify a total of 428 markers genes that drive the classification of 33 cancer types and normal. The concordance of differential expressions of these markers between the represented cancer type and others are confirmed. Successful classification of cancer types and a normal group regardless of normal tissues' origin suggested that the identified markers are cancer-specific rather than tissue-specific.ConclusionNovel GCNN models have been established to predict cancer types or normal tissue based on gene expression profiles. We demonstrated the results from the TCGA dataset that these models can produce accurate classification (above 94%), using cancer-specific markers genes. The models and the source codes are publicly available and can be readily adapted to the diagnosis of cancer and other diseases by the data-driven modeling research community.

Project description:The Brain Tumor presents a highly critical situation concerning the brain, characterized by the uncontrolled growth of an abnormal cell cluster. Early brain tumor detection is essential for accurate diagnosis and effective treatment planning. In this paper, a novel Convolutional Neural Network (CNN) based Graph Neural Network (GNN) model is proposed using the publicly available Brain Tumor dataset from Kaggle to predict whether a person has brain tumor or not and if yes then which type (Meningioma, Pituitary or Glioma). The objective of this research and the proposed models is to provide a solution to the non-consideration of non-Euclidean distances in image data and the inability of conventional models to learn on pixel similarity based upon the pixel proximity. To solve this problem, we have proposed a Graph based Convolutional Neural Network (GCNN) model and it is found that the proposed model solves the problem of considering non-Euclidean distances in images. We aimed at improving brain tumor detection and classification using a novel technique which combines GNN and a 26 layered CNN that takes in a Graph input pre-convolved using Graph Convolution operation. The objective of Graph Convolution is to modify the node features (data linked to each node) by combining information from nearby nodes. A standard pre-computed Adjacency matrix is used, and the input graphs were updated as the averaged sum of local neighbor nodes, which carry the regional information about the tumor. These modified graphs are given as the input matrices to a standard 26 layered CNN with Batch Normalization and Dropout layers intact. Five different networks namely Net-0, Net-1, Net-2, Net-3 and Net-4 are proposed, and it is found that Net-2 outperformed the other networks namely Net-0, Net-1, Net-3 and Net-4. The highest accuracy achieved was 95.01% by Net-2. With its current effectiveness, the model we propose represents a critical alternative for the statistical detection of brain tumors in patients who are suspected of having one.

Project description:BackgroundMolecular interaction networks summarize complex biological processes as graphs, whose structure is informative of biological function at multiple scales. Simultaneously, omics technologies measure the variation or activity of genes, proteins, or metabolites across individuals or experimental conditions. Integrating the complementary viewpoints of biological networks and omics data is an important task in bioinformatics, but existing methods treat networks as discrete structures, which are intrinsically difficult to integrate with continuous node features or activity measures. Graph neural networks map graph nodes into a low-dimensional vector space representation, and can be trained to preserve both the local graph structure and the similarity between node features.ResultsWe studied the representation of transcriptional, protein-protein and genetic interaction networks in E. coli and mouse using graph neural networks. We found that such representations explain a large proportion of variation in gene expression data, and that using gene expression data as node features improves the reconstruction of the graph from the embedding. We further proposed a new end-to-end Graph Feature Auto-Encoder framework for the prediction of node features utilizing the structure of the gene networks, which is trained on the feature prediction task, and showed that it performs better at predicting unobserved node features than regular MultiLayer Perceptrons. When applied to the problem of imputing missing data in single-cell RNAseq data, the Graph Feature Auto-Encoder utilizing our new graph convolution layer called FeatGraphConv outperformed a state-of-the-art imputation method that does not use protein interaction information, showing the benefit of integrating biological networks and omics data with our proposed approach.ConclusionOur proposed Graph Feature Auto-Encoder framework is a powerful approach for integrating and exploiting the close relation between molecular interaction networks and functional genomics data.

Project description:BackgroundThe recent development of high-throughput sequencing has created a large collection of multi-omics data, which enables researchers to better investigate cancer molecular profiles and cancer taxonomy based on molecular subtypes. Integrating multi-omics data has been proven to be effective for building more precise classification models. Most current multi-omics integrative models use either an early fusion in the form of concatenation or late fusion with a separate feature extractor for each omic, which are mainly based on deep neural networks. Due to the nature of biological systems, graphs are a better structural representation of bio-medical data. Although few graph neural network (GNN) based multi-omics integrative methods have been proposed, they suffer from three common disadvantages. One is most of them use only one type of connection, either inter-omics or intra-omic connection; second, they only consider one kind of GNN layer, either graph convolution network (GCN) or graph attention network (GAT); and third, most of these methods have not been tested on a more complex classification task, such as cancer molecular subtypes.ResultsIn this study, we propose a novel end-to-end multi-omics GNN framework for accurate and robust cancer subtype classification. The proposed model utilizes multi-omics data in the form of heterogeneous multi-layer graphs, which combine both inter-omics and intra-omic connections from established biological knowledge. The proposed model incorporates learned graph features and global genome features for accurate classification. We tested the proposed model on the Cancer Genome Atlas (TCGA) Pan-cancer dataset and TCGA breast invasive carcinoma (BRCA) dataset for molecular subtype and cancer subtype classification, respectively. The proposed model shows superior performance compared to four current state-of-the-art baseline models in terms of accuracy, F1 score, precision, and recall. The comparative analysis of GAT-based models and GCN-based models reveals that GAT-based models are preferred for smaller graphs with less information and GCN-based models are preferred for larger graphs with extra information.

Dataset Information

Lung Radiomics Features Selection for COPD Stage Classification Based on Auto-Metric Graph Neural Network.

Publications

Lung Radiomics Features Selection for COPD Stage Classification Based on Auto-Metric Graph Neural Network.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets