Pub Date : 2024-01-01Epub Date: 2024-02-03DOI: 10.1007/978-3-031-47425-5_21
Zixuan Wen, Jingxuan Bao, Shu Yang, Shannon L Risacher, Andrew J Saykin, Paul M Thompson, Christos Davatzikos, Heng Huang, Yize Zhao, Li Shen
We introduce an informative metric, called morphometric correlation, as a measure of shared neuroanatomic similarity between two cognitive traits. Traditional estimates of trait correlations can be confounded by factors beyond brain morphology. To exclude these confounding factors, we adopt a Gaussian kernel to measure the morphological similarity between individuals and compare pure neuroanatomic correlations among cognitive traits. In our empirical study, we employ a multiscale strategy. Given a set of cognitive traits, we first perform morphometric correlation analysis for each pair of traits to reveal their shared neuroanatomic correlation at the whole brain (or global) level. After that, we extend our whole brain concept to regional morphometric correlation and estimate shared neuroanatomic similarity between two cognitive traits at the regional (or local) level. Our results demonstrate that morphometric correlation can provide insights into shared neuroanatomic architecture between cognitive traits. Furthermore, we also estimate the morphometricity of each cognitive trait at both global and local levels, which can be used to better understand how neuroanatomic changes influence individuals' cognitive status.
{"title":"Identifying Shared Neuroanatomic Architecture between Cognitive Traits through Multiscale Morphometric Correlation Analysis.","authors":"Zixuan Wen, Jingxuan Bao, Shu Yang, Shannon L Risacher, Andrew J Saykin, Paul M Thompson, Christos Davatzikos, Heng Huang, Yize Zhao, Li Shen","doi":"10.1007/978-3-031-47425-5_21","DOIUrl":"https://doi.org/10.1007/978-3-031-47425-5_21","url":null,"abstract":"<p><p>We introduce an informative metric, called morphometric correlation, as a measure of shared neuroanatomic similarity between two cognitive traits. Traditional estimates of trait correlations can be confounded by factors beyond brain morphology. To exclude these confounding factors, we adopt a Gaussian kernel to measure the morphological similarity between individuals and compare pure neuroanatomic correlations among cognitive traits. In our empirical study, we employ a multiscale strategy. Given a set of cognitive traits, we first perform morphometric correlation analysis for each pair of traits to reveal their shared neuroanatomic correlation at the whole brain (or global) level. After that, we extend our whole brain concept to regional morphometric correlation and estimate shared neuroanatomic similarity between two cognitive traits at the regional (or local) level. Our results demonstrate that morphometric correlation can provide insights into shared neuroanatomic architecture between cognitive traits. Furthermore, we also estimate the morphometricity of each cognitive trait at both global and local levels, which can be used to better understand how neuroanatomic changes influence individuals' cognitive status.</p>","PeriodicalId":517997,"journal":{"name":"Medical image computing and computer assisted intervention - MICCAI 2023 workshops : ISIC 2023, Care-AI 2023, MedAGI 2023, DeCaF 2023, held in conjunction with MICCAI 2023, Vancouver, BC, Canada, October 8-12, 2023, proceedings","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10993314/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140854691","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-01DOI: 10.1007/978-3-031-47425-5_18
Najla Al Turkestani, Lingrui Cai, Lucia Cevidanes, Jonas Bianchi, Winston Zhang, Marcela Gurgel, Maxime Gillot, Baptiste Baquero, Reza Soroushmehr
This paper proposes a machine learning model using privileged information (LUPI) and normalized mutual information feature selection method (NMIFS) to build a robust and accurate framework to diagnose patients with Temporomandibular Joint Osteoarthritis (TMJ OA). To build such a model, we employ clinical, quantitative imaging and additional biological markers as privileged information. We show that clinical features play a leading role in the TMJ OA diagnosis and quantitative imaging features, extracted from cone-beam computerized tomography (CBCT) scans, improve the model performance. As the proposed LUPI model employs biological data in the training phase (which boosted the model performance), this data is unnecessary for the testing stage, indicating the model can be widely used even when only clinical and imaging data are collected. The model was validated using 5-fold stratified cross-validation with hyperparameter tuning to avoid the bias of data splitting. Our method achieved an AUC, specificity and precision of 0.81, 0.79 and 0.77, respectively.
{"title":"Osteoarthritis Diagnosis Integrating Whole Joint Radiomics and Clinical Features for Robust Learning Models Using Biological Privileged Information.","authors":"Najla Al Turkestani, Lingrui Cai, Lucia Cevidanes, Jonas Bianchi, Winston Zhang, Marcela Gurgel, Maxime Gillot, Baptiste Baquero, Reza Soroushmehr","doi":"10.1007/978-3-031-47425-5_18","DOIUrl":"10.1007/978-3-031-47425-5_18","url":null,"abstract":"<p><p>This paper proposes a machine learning model using privileged information (LUPI) and normalized mutual information feature selection method (NMIFS) to build a robust and accurate framework to diagnose patients with Temporomandibular Joint Osteoarthritis (TMJ OA). To build such a model, we employ clinical, quantitative imaging and additional biological markers as privileged information. We show that clinical features play a leading role in the TMJ OA diagnosis and quantitative imaging features, extracted from cone-beam computerized tomography (CBCT) scans, improve the model performance. As the proposed LUPI model employs biological data in the training phase (which boosted the model performance), this data is unnecessary for the testing stage, indicating the model can be widely used even when only clinical and imaging data are collected. The model was validated using 5-fold stratified cross-validation with hyperparameter tuning to avoid the bias of data splitting. Our method achieved an AUC, specificity and precision of 0.81, 0.79 and 0.77, respectively.</p>","PeriodicalId":517997,"journal":{"name":"Medical image computing and computer assisted intervention - MICCAI 2023 workshops : ISIC 2023, Care-AI 2023, MedAGI 2023, DeCaF 2023, held in conjunction with MICCAI 2023, Vancouver, BC, Canada, October 8-12, 2023, proceedings","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10964798/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140295780","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号