Background: Identifying prediagnostic neurodegenerative disease is a critical issue in neurodegenerative disease research, and Alzheimer’s disease (AD) in particular, to identify populations suitable for preventive and early disease-modifying trials. Evidence from genetic and other studies suggests the neurodegeneration of Alzheimer’s disease measured by brain atrophy starts many years before diagnosis, but it is unclear whether these changes can be used to reliably detect prediagnostic sporadic disease. Methods: We trained a Bayesian machine learning neural network model to generate a neuroimaging phenotype and AD score representing the probability of AD using structural MRI data in the Alzheimer’s Disease Neuroimaging Initiative (ADNI) Cohort (cut-off 0.5, AUC 0.92, PPV 0.90, NPV 0.93). We go on to validate the model in an independent real-world dataset of the National Alzheimer’s Coordinating Centre (AUC 0.74, PPV 0.65, NPV 0.80) and demonstrate the correlation of the AD-score with cognitive scores in those with an AD-score above 0.5. We then apply the model to a healthy population in the UK Biobank study to identify a cohort at risk for Alzheimer’s disease. Results: We show that the cohort with a neuroimaging Alzheimer’s phenotype has a cognitive profile in keeping with Alzheimer’s disease, with strong evidence for poorer fluid intelligence, and some evidence of poorer numeric memory, reaction time, working memory, and prospective memory. We found some evidence in the AD-score positive cohort for modifiable risk factors of hypertension and smoking. Conclusions: This approach demonstrates the feasibility of using AI methods to identify a potentially prediagnostic population at high risk for developing sporadic Alzheimer’s disease.
Identifying healthy individuals with Alzheimer’s disease neuroimaging phenotypes in the UK Biobank / Azevedo, T.; Bethlehem, R. A. I.; Whiteside, D. J.; Swaddiwudhipong, N.; Rowe, J. B.; Lio, P.; Rittman, T.; Li, G.; Petrie, E. C.; Peskind, E. R.; Villena, T.; Martinez, W.; Sadowsky, C.; Lee, A.; Ravdin, L.; Lin, M.; Chiang, G.; Relkin, N.; Santulli, R. B.; Hynes, M. L.; Seltzer, M.; Flashman, L. A.; Anderson, K.; Blank, K.; Pearlson, G. D.; Kittur, S.; Correia, S.; Malloy, P.; Salloway, S.; Potkin, S.; Newhouse, P.; Garcia, A.; Shi, J.; Masdeu, J.; Wiliams, A.; Mintzer, J.; O'Connelll, A.; Mintz, A.; Munic, D.; Bernick, C.; Ritter, A.; Drake, J. D.; Daiello, L. A.; Tremont, G.; Drake, J.; Ott, B. R.; Sink, K. M.; Yang, M.; Cleveland, J.; Craft, S.; Williamson, J. D.; Schultz, S. K.; Nam, K. W.; Shim, H.; Koleva, H.; Smith, K. E.; Boles Ponto, L. L.; Miller, D. D.; Celmins, D.; Zimmerman, E. A.; Hart, D.; Kelley, B.; Tarawneh, R.; Kataki, M.; Scharre, D. W.; Burke, A.; Perrin, A.; Carlsson, C. M.; Asthana, S.; Johnson, S.; Bartha, R.; Lee, T. -Y.; Borrie, M.; Rainka, M.; Capote, H.; Bates, V.; Carmichael, O.; Decarli, C.; Olichney, J.; Maillard, P.; Fletcher, E.; Fatica, P.; Tatsuoka, C.; Ogrocki, P.; Lerner, A.; Nadarajah, S.; Nwulia, E.; Khan, J. I.; Wolday, S.; Ntekim, O. E.; Obisesan, T. O.; Budson, A. E.; Kowall, N.; Mez, J.; Stern, R.; Killiany, R.; Sabbagh, M.; Clark Edward Zamrini, K. A.; Spann, B. M.; Atri, A.; Belden, C. M.; Tinklenberg, J.; Rosen, A.; Lane, B.; White, J. D.; Coleman, J.; Chao, S.; Taylor, J. L.; Yesavage, J.; Marshall, G. A.; Johnson, K. A.; Sperling, R. A.; Poe, J.; Mccann, K.; Reynolds, B.; Johnson, K.; Turner, R. S.; Perry, D.; Rosen, H.; Miller, B. L.; Fernandez, D.; Ong, E.; Stark, S.; Catalinotto, D.; Scherer, K.; Au, Y. H.; Bickford, D.; Nelson, C.; Mackin, S.; Rosen, H. J.; Sarrael, A.; Hernando, R.; Pomara, N.; Bonakdarpour, B.; Weintraub, S.; Lipowski, K.; Johnson, N.; Wu, C. -K.; Kerwin, D.; M. -Marsel Mesulam J., R.; Rogalski, E.; Muse, B.; Grant, I.; Rogers, J.; Drost, D.; Kertesz, A.; Rachinsky, I.; Pasternak, S.; Finger, E.; Assaly, M.; Feldman, H.; Sossi, V.; Mudge, B.; Kim, E.; Hsiung, G. -Y. R.; Heyn, C. C.; Stefanovic, B.; Black, S.; Hosein, C.; Vaitekunis, S.; Chertkow, H.; Varma, P.; Carson, R. E.; Macavoy, M. G.; Good, S. P.; Mecca, A. P.; van Dyck, C. H.; Herring, S.; Brosch, J. R.; Matthews, B. R.; Hake, A. M.; Farlow, M. R.; Poki-Walker, K.; Parfitt, F.; Graff-Radford, N. R.; Teng, E.; Woo, E.; Bartzokis, G.; Lu, P. H.; Tingus, K.; Apostolova, L.; Kremen, S.; Silverman, D. H. S.; Brooks, W. M.; Swerdlow, R. H.; Burns, J. M.; Cellar, J. S.; Hajjar, I.; Lah, J. J.; Levey, A. I.; Quiceno, M.; Mathews, D.; Womack, K.; Nguyen, T.; Kelley, B.; Witbracht, M.; Sosa, E.; Yanez, B.; Pierce, A.; Thai, G.; Reist, C.; Brand, C.; Ismail, M. S.; Makino, K. M.; Goldstein, B. S.; Keltz, M.; Martin, K. S.; Kowalski, N.; Martin, K.; Porsteinsson, A. P.; Simpson, D. M.; Oakley, M.; Butters, M.; Zmuda, M.; Lopez, O. L.; Raslau, F. D.; El Khouli, R.; Jicha, G. A.; Smith, C. D.; Arnold, S. E.; Clark, C. M.; Vaishnavi, S.; Wolk, D. A.; Karlawish, J. H.; Borges-Neto, S.; Wong, T. Z.; James, O.; Petrella, J. R.; Doraiswamy, P. M.; Singleton-Garvin, J.; Sheikh, M. O.; Chen, S.; Ulysse, A.; Castro, K. M.; Rao, J.; Faustin, A.; Shulman, M.; Wisniewski, T.; Sadowski, M.; Fargher, K.; Raj, B. A.; Smith, A.; Kielb, S.; Greig, M. T.; Brichko, R.; Rudow, S.; Farrington, L.; Onyike, C.; Varon, D.; Albert, M.; Rodriguez, R.; Greig-Custo, M. T.; Duara, R.; Samuels, P.; Arfanakis, K.; Fleischman, D.; Blanchard, K. S.; Sood, A.; Lamar, M.; Shah, R. C.; Mitsis, E.; Greenberg, J.; Goldstein, M. A.; Grossman, H.; Marson, D.; Brockington, J.; Clark, D.; Griffith, R.; Love, M. N.; Geldmacher, D.; Schneider, S.; Mintun, M. A.; Creech, M. L.; Oliver, A.; Stobbs-Cucchi, G.; Carroll, M.; Winkfield, D.; Morris, J. C.; Ances, B.; Mintz, A.; Honig, L. S.; Stern, Y.; Dang, M.; Rountree, S.; Chowdhury, M.; Shibley, V.; Doody, R. S.; Kass, J. S.; Lamb, A.; Pacini, N.; Pavlik, V.; Villanueva-Meyer, J.; Johnson, K.; Knopman, D.; Albers, C. S.; Mason, S. S.; Petersen, R.; Zbizek-Nulph, L.; Lord, J. L.; Nulph, Z.; Heidebrink, J. L.; Ziolkowski, J.; Fleisher, A.; Vanderswag, H.; Brewer, J.; Spann, B. M.; Dagerman, K.; Teodoro, L.; Becerra, M.; Pawluczyk, S.; Schneider, L. S.; Quinn, J.; Dolen, S.; Carter, R.; Kaye, J. A.; Crissey, R.; Lind, B.; Silbert, L. C.. - In: COMMUNICATIONS MEDICINE. - ISSN 2730-664X. - 3:1(2023). [10.1038/s43856-023-00313-w]
Identifying healthy individuals with Alzheimer’s disease neuroimaging phenotypes in the UK Biobank
Lio P.;
2023
Abstract
Background: Identifying prediagnostic neurodegenerative disease is a critical issue in neurodegenerative disease research, and Alzheimer’s disease (AD) in particular, to identify populations suitable for preventive and early disease-modifying trials. Evidence from genetic and other studies suggests the neurodegeneration of Alzheimer’s disease measured by brain atrophy starts many years before diagnosis, but it is unclear whether these changes can be used to reliably detect prediagnostic sporadic disease. Methods: We trained a Bayesian machine learning neural network model to generate a neuroimaging phenotype and AD score representing the probability of AD using structural MRI data in the Alzheimer’s Disease Neuroimaging Initiative (ADNI) Cohort (cut-off 0.5, AUC 0.92, PPV 0.90, NPV 0.93). We go on to validate the model in an independent real-world dataset of the National Alzheimer’s Coordinating Centre (AUC 0.74, PPV 0.65, NPV 0.80) and demonstrate the correlation of the AD-score with cognitive scores in those with an AD-score above 0.5. We then apply the model to a healthy population in the UK Biobank study to identify a cohort at risk for Alzheimer’s disease. Results: We show that the cohort with a neuroimaging Alzheimer’s phenotype has a cognitive profile in keeping with Alzheimer’s disease, with strong evidence for poorer fluid intelligence, and some evidence of poorer numeric memory, reaction time, working memory, and prospective memory. We found some evidence in the AD-score positive cohort for modifiable risk factors of hypertension and smoking. Conclusions: This approach demonstrates the feasibility of using AI methods to identify a potentially prediagnostic population at high risk for developing sporadic Alzheimer’s disease.File | Dimensione | Formato | |
---|---|---|---|
Azevedo_Identifying_2023.pdf
accesso aperto
Note: https://www.nature.com/articles/s43856-023-00313-w.pdf
Tipologia:
Versione editoriale (versione pubblicata con il layout dell'editore)
Licenza:
Creative commons
Dimensione
1.76 MB
Formato
Adobe PDF
|
1.76 MB | Adobe PDF |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.