[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-120856-en":3,"doc-seo-120856-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},120856,1649267921044,"Ava Thompson","https://us-avatar.wpscdn.com/avatar/1800007509477c92dfb?_k=1782875107921204101",8,"Research & Report","Cocaine Use Prediction with Tensor-based Machine Learning on Multimodal MRI Connectome Data","Machine learning methods are developed to predict cocaine use using multimodal MRI connectomic data. Functional MRI (fMRI) and diffusion MRI (dMRI) from 275 individuals are parcellated into 246 ROIs via the Brainnetome atlas, then preprocessed and converted into tensor form. A tensor-based unsupervised method applies the high-order Lloyd algorithm to compress the tensor into a 6-cluster representation, extracting reduced tensor features and combining them with demographics (age, gender, race, HIV status). A Catboost model trained with subsampling and nested cross-validation attains 0.857 accuracy for identifying cocaine users, with model comparisons and feature importance reported.","arXiv :2310 . 14146v1 [ stat .AP] 22 Oct 2023  \nCocaine Use Prediction with Tensor-based Machine Learning on Multimodal MRI Connectome Data  \nAnru R. Zhanga,d,1 , Ryan P. Bellb,1 , Chen Anc,1 , Runshi Tange,1 , Shana A. Hallb , Cliburn Chana , Kareem Al-Khalilb , Christina S. Meadeb,˚  \na Department of Biostatistics and Bioinformatics, Duke  \nUniveristy, Durham, 27710, NC, U.S.  \nb Department of Psychiatry and Behavioral Sciences, Duke  \nUniveristy, Durham, 27710, NC, U.S.  \nc Department of Mathematics, Duke Univeristy, Durham, 27708, NC, U.S. d Department of Computer Science, Duke University, Durham, 27710, NC, U. S.  \ne Department of Statistics, University of Wisconsin–Madison, Madison, 53706, WI, U.S.  \nAbstract  \nThis paper considers the use of machine learning algorithms for predicting cocaine use based on magnetic resonance imaging (MRI) connectomic data. The study utilized functional MRI (fMRI) and diffusion MRI (dMRI) data collected from 275 individuals, which was then parcellated into 246 regions of interest (ROIs) using the Brainnetome atlas. After data preprocessing, the datasets were transformed into tensor form. We developed a tensor-based unsupervised machine learning algorithm to reduce the size of the data tensor from 275 (individuals) ˆ2 (fMRI and dMRI) ˆ246 (ROIs) ˆ246 (ROIs) to 275 (individuals) ˆ2 (fMRI and dMRI) ˆ6 (clusters) ˆ6 (clusters) . This was achieved by applying the high-order Lloyd algorithm to group the ROI data into 6 clusters. Features were extracted from the reduced tensor and combined with demographic features (age, gender, race, and HIV status) . The resulting dataset was used to train a Catboost model using subsampling and nested cross-validation techniques, which achieved a prediction accuracy of 0 .857 for identifying cocaine users. The model was also compared with other models, and the feature importance of the model was presented.  \nOverall, this study highlights the potential for using tensor-based ma-  \n˚ Corresponding author. Email address:  \nEmail address: [christina.meade@duke.edu](christina.meade@duke.edu) (Christina S. Meade)  \n1 These authors have equally contributed to this paper.  \nPreprint submitted to arXiv October 24, 2023  \nchine learning algorithms to predict cocaine use based on MRI connectomic data and presents a promising approach for identifying individuals at risk of substance abuse.  \nKeywords: tensor methods, cocaine addiction, functional MRI, structural MRI  \n1. Introduction  \nWorldwide, cocaine was used by 20 million (0.4%) individuals between the ages of 15-64 in 2019 (United Nations Office on Drugs and Crime, 2021), and an estimated 2% of persons aged 12 or older in the United States used cocaine in the past year [1] . Chronic cocaine use is associated with both structural and functional deficits in the brain. Individuals who use cocaine show lower white matter integrity consistently within the corpus callosum and frontal regions [2 , 3], and also across diffuse association and projection fibers [4], as measured by diffusion tensor imaging (DTI), a method of examining white matter microstructure in magnetic resonance imaging (MRI) data. Resting-state functional MRI (fMRI) studies have been utilized to identify alterations in cerebral blood flow within and between functional networks in people who use cocaine. These alterations include deficits in functional connectivity within and between neural networks associated with reward processing and executive functioning [5 , 6 , 7] . Research shows altered reorganization of functional connectomes involving executive, reward, salience, and default mode networks [8 , 9 , 10] . Other studies integrating structural and functional connectomics show cocaine-related effects involving the interoceptive networks [11] .  \nConnectomics is a method of describing brain networks based on data collected via non-invasive MRI techniques, including fMRI and diffusion MRI (dMRI) [12] . MRI connectome data provides a powerful way to map","cbCaifXfz11b7CQN","https://ap.wps.com/l/cbCaifXfz11b7CQN","pdf",6025590,1,41,"English","en",105,"# Introduction\n## Background on cocaine-related brain deficits\n## MRI connectomics and multimodal imaging\n## Tensor-based machine learning motivation\n# Method Overview\n## Data preprocessing and ROI parcellation\n## Tensor transformation and blockwise reduction\n## Feature extraction and model training\n# Results\n## Prediction accuracy and validation strategy\n## Model comparison and feature importance","[{\"question\":\"How is cocaine use prediction modeled in this study?\",\"answer\":\"A Catboost model is trained using tensor-reduced MRI connectome features combined with demographic variables, then evaluated with subsampling and nested cross-validation.\"},{\"question\":\"What tensor-based step reduces the dimensionality of the MRI connectome data?\",\"answer\":\"The study uses a tensor-based unsupervised approach applying the high-order Lloyd algorithm to cluster ROI data and reduce the tensor representation into 6 clusters.\"},{\"question\":\"Which MRI modalities and demographic features are included?\",\"answer\":\"Functional MRI (fMRI) and diffusion MRI (dMRI) are used, along with demographic features including age, gender, race, and HIV status.\"}]","Cocaine Use Prediction with Tensor-based Machine Learning on Multimodal MRI Connectome Data | PDF",1785732370,103,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"cocaine-use-prediction-with-tensor-based-machine-learning-on-multimodal-mri-connectome-data","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/cocaine-use-prediction-with-tensor-based-machine-learning-on-multimodal-mri-connectome-data/120856/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-03",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"How is cocaine use prediction modeled in this study?","Question",{"text":75,"@type":76},"A Catboost model is trained using tensor-reduced MRI connectome features combined with demographic variables, then evaluated with subsampling and nested cross-validation.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What tensor-based step reduces the dimensionality of the MRI connectome data?",{"text":80,"@type":76},"The study uses a tensor-based unsupervised approach applying the high-order Lloyd algorithm to cluster ROI data and reduce the tensor representation into 6 clusters.",{"name":82,"@type":73,"acceptedAnswer":83},"Which MRI modalities and demographic features are included?",{"text":84,"@type":76},"Functional MRI (fMRI) and diffusion MRI (dMRI) are used, along with demographic features including age, gender, race, and HIV status.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]