51. Longitudinal measurement and hierarchical classification framework for the prediction of Alzheimer's disease
- Author
-
Mary L. Hynes, Liberty Teodoro, John C. Brockington, Connie Brand, Paul Malloy, Russell H. Swerdlow, Ronald C. Petersen, Judith L. Heidebrink, Pierre N. Tariot, Curtis Caldwell, Clifford R. Jack, David G. Clark, Neill R. Graff-Radford, Charles D. Smith, Geoffrey Tremont, Ranjan Duara, Stephen Pasternack, T. Y. Lee, Owen Carmichael, Nadira Trncic, Irina Rachisky, Daniel D'Agostino, James J. Lah, Steven G. Potkin, Howard Bergman, Dana M. Pogorelec, Lon S. Schneider, Anna Burke, Sherye A. Sirrel, Henry W. Querfurth, Michael Lin, David Bachman, Edward Coleman, Michele Assaly, Allyson C. Rosen, Jeffrey M. Burns, Balebail Ashok Raj, Jared R. Brosch, Joanne L. Lord, William Brooks, Brigid Reynolds, Karen S. Anderson, Sandra Jacobson, Nunzio Pomara, Patricia Lynn Johnson, George Bartzokis, Parianne Fatica, Benita Mudge, Dana Nguyen, Carl H. Sadowsky, Michael Borrie, Qianjin Feng, Chiadi U. Onyike, Partha Sinha, Gloria Chaing, Howard Chertkow, Leyla deToledo-Morrell, Bojana Stefanovic, Richard E. Carson, Wufan Chen, Ronald J. Killiany, Mimi Dang, Thomas O. Obisesan, Christopher H. van Dyck, Maria Carroll, Gaby Thai, Arthur W. Toga, Chuang Kuo Wu, Erik D. Roberson, Effie M. Mitsis, Smita Kittur, Keith A. Johnson, Dana Mathews, Sara Dolen, Raj C. Shah, M.-Marsel Mesulam, Howard J. Rosen, Karen L. Bell, Ging-Yuek Robin Hsiung, Teresa Villena, Kris Johnson, Saba Wolday, Douglas W. Scharre, Kyle B. Womack, Maria Kataki, Barton Lane, Angela Oliver, Greg Jicha, Reisa A. Sperling, Wei Yang, David S. Geldmacher, Lawrence S. Honig, Sanjay Asthana, Janet S. Cellar, William J. Jagust, Dzintra Celmins, Susan Rountree, Christina A. Michel, Allan I. Levey, Tracy Kendall, Lisa D. Ravdin, Jared R. Tinklenberg, Brittany Cerbone, Alice D. Brown, Marilyn S. Albert, Andrew J. Saykin, Raymundo Hernando, Sandra Weintraub, John Q. Trojanowki, Raina Carter, Betty Lind, Kristin Fargher, Sterling C. Johnson, P. Murali Doraiswamy, Jeffrey R. Petrella, Neil W. Kowall, Sara S. Mason, Heather Johnson, Mary L. Creech, Stacy Schneider, Donna Munic, Liana G. Apostolova, Peter A. Hardy, Munir Chowdhury, Bruce L. Miller, Ruth A. Mulnard, Curtis Tatsuoka, Po H. Lu, Daniel C. Marson, Pauline Maillard, John C. Morris, Marwan N. Sabbagh, Jeffrey Kaye, Hillel Grossman, Gary R. Conrad, Karen Blank, Meiyan Huang, Stephanie Kielb, Andrew Kertesz, Jerome A. Yesavage, Leslie Shaw, Martin R. Farlow, Maria T. Greig, Jacobo Mintzer, Susan De Santi, David S. Knopman, Marc Seltzer, Scott Herring, Joy L. Taylor, Vernice Bates, Rob Bartha, Cynthia Hunt, Henry Rusinek, Randall Griffith, Cynthia M. Carlsson, Charles Bernick, Bonnie S. Goldstein, Rachelle S. Doody, Leslie Gordineer, Catherine Mc-Adams-Ortiz, Kim Martin, Howard Feldman, David C. Perry, Horacio Capote, Lidia Glodzik, Stephen Correia, James B. Brewer, Elizabeth Finger, Jeff D. Williamson, Franklin Watkins, Borna Bonakdarpour, Colleen S. Albers, M. Saleem Ismail, Alan J. Lerner, Daniel Varon, Christine M. Belden, Sonia Pawluczyk, Paul S. Aisen, Pradeep Garg, Kelly M. Makino, Laurel A. Beckett, Peggy Roberts, Nancy Johnson, Anahita Adeli, Terence Z. Wong, Michelle Rainka, Elizabeth Oates, Amanda Smith, Kenneth M. Spicer, Laura A. Flashman, Kristine Lipowski, Charles DeCarli, Stephanie Reeder, Mauricio Beccera, Dick Trost, Alexander Norbash, Lisa C. Silbert, Michael W. Weiner, Gad A. Marshall, Ann Marie Hake, Pradeep Varma, Francine Parfitt, Chris Hosein, Adam S. Fleisher, Marissa Natelson Love, Joanne S. Allard, Earl A. Zimmerman, Kathleen Tingus, Brian R. Ott, Joseph F. Quinn, Anton P. Porsteinsson, Paula Ogrocki, Raymond Scott Turner, Salvador Borges-Neto, James E. Galvin, Yaakov Stern, Andrew E. Budson, Martha G. MacAvoy, Daniel H.S. Silverman, Robert B. Santulli, Adrian Preda, Godfrey D. Pearlson, Mark A. Mintun, Stephen Salloway, Mary Quiceno, Kaycee M. Sink, John M Olichney, Antero Sarrael, Beau M. Ances, Javier Villanueva-Meyer, Mony J. de Leon, Sandra E. Black, Bryan M. Spann, Diana R. Kerwin, Ellen Woo, Helen Vanderswag, Erin E. Franklin, Robert C. Green, and Norman R. Relkin
- Subjects
Male ,Computer science ,Neuroimaging ,Disease ,computer.software_genre ,Sensitivity and Specificity ,Article ,030218 nuclear medicine & medical imaging ,03 medical and health sciences ,0302 clinical medicine ,Alzheimer Disease ,Predictive Value of Tests ,mental disorders ,medicine ,Dementia ,Humans ,Cognitive Dysfunction ,Longitudinal Studies ,Cognitive impairment ,Aged ,Aged, 80 and over ,Multidisciplinary ,medicine.diagnostic_test ,business.industry ,Brain ,Pattern recognition ,Magnetic resonance imaging ,Middle Aged ,medicine.disease ,Magnetic Resonance Imaging ,Early Diagnosis ,Disease Progression ,Female ,Data mining ,Artificial intelligence ,business ,computer ,030217 neurology & neurosurgery ,Alzheimer's Disease Neuroimaging Initiative ,Follow-Up Studies - Abstract
Accurate prediction of Alzheimer’s disease (AD) is important for the early diagnosis and treatment of this condition. Mild cognitive impairment (MCI) is an early stage of AD. Therefore, patients with MCI who are at high risk of fully developing AD should be identified to accurately predict AD. However, the relationship between brain images and AD is difficult to construct because of the complex characteristics of neuroimaging data. To address this problem, we present a longitudinal measurement of MCI brain images and a hierarchical classification method for AD prediction. Longitudinal images obtained from individuals with MCI were investigated to acquire important information on the longitudinal changes, which can be used to classify MCI subjects as either MCI conversion (MCIc) or MCI non-conversion (MCInc) individuals. Moreover, a hierarchical framework was introduced to the classifier to manage high feature dimensionality issues and incorporate spatial information for improving the prediction accuracy. The proposed method was evaluated using 131 patients with MCI (70 MCIc and 61 MCInc) based on MRI scans taken at different time points. Results showed that the proposed method achieved 79.4% accuracy for the classification of MCIc versus MCInc, thereby demonstrating very promising performance for AD prediction.
- Published
- 2017