eprintid: 10146010 rev_number: 7 eprint_status: archive userid: 699 dir: disk0/10/14/60/10 datestamp: 2022-03-30 09:03:57 lastmod: 2022-03-30 09:04:19 status_changed: 2022-03-30 09:03:57 type: article metadata_visibility: show sword_depositor: 699 creators_name: de Lange, Ann-Marie G creators_name: Anatürk, Melis creators_name: Rokicki, Jaroslav creators_name: Han, Laura KM creators_name: Franke, Katja creators_name: Alnaes, Dag creators_name: Ebmeier, Klaus P creators_name: Draganski, Bogdan creators_name: Kaufmann, Tobias creators_name: Westlye, Lars T creators_name: Hahn, Tim creators_name: Cole, James H title: Mind the gap: Performance metric evaluation in brain-age prediction ispublished: inpress divisions: C05 divisions: F48 divisions: B04 divisions: UCL keywords: brain-age prediction, machine learning, neuroimaging, statistics note: © 2022 The Authors. Human Brain Mapping published by Wiley Periodicals LLC. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. abstract: Estimating age based on neuroimaging-derived data has become a popular approach to developing markers for brain integrity and health. While a variety of machine-learning algorithms can provide accurate predictions of age based on brain characteristics, there is significant variation in model accuracy reported across studies. We predicted age in two population-based datasets, and assessed the effects of age range, sample size and age-bias correction on the model performance metrics Pearson's correlation coefficient (r), the coefficient of determination (R2 ), Root Mean Squared Error (RMSE) and Mean Absolute Error (MAE). The results showed that these metrics vary considerably depending on cohort age range; r and R2 values are lower when measured in samples with a narrower age range. RMSE and MAE are also lower in samples with a narrower age range due to smaller errors/brain age delta values when predictions are closer to the mean age of the group. Across subsets with different age ranges, performance metrics improve with increasing sample size. Performance metrics further vary depending on prediction variance as well as mean age difference between training and test sets, and age-bias corrected metrics indicate high accuracy-also for models showing poor initial performance. In conclusion, performance metrics used for evaluating age prediction models depend on cohort and study-specific data characteristics, and cannot be directly compared across different studies. Since age-bias corrected metrics generally indicate high accuracy, even for poorly performing models, inspection of uncorrected model results provides important information about underlying model attributes such as prediction variance. date: 2022-03-21 date_type: published publisher: Wiley official_url: https://doi.org/10.1002/hbm.25837 oa_status: green full_text_type: pub language: eng primo: open primo_central: open_green verified: verified_manual elements_id: 1946283 doi: 10.1002/hbm.25837 medium: Print-Electronic lyricists_name: Cole, James lyricists_id: JCOLE07 actors_name: Barczynska, Patrycja actors_id: PBARC91 actors_role: owner funding_acknowledgements: 5R24AG061421-03 [Collaboratory on Research Definitions for Reserve and Resilience in Cognitive Aging and Dementia]; FR 3709/1-2 [Deutsche Forschungsgemeinschaft]; HA7070/2-2 [Deutsche Forschungsgemeinschaft]; HA7070/3 [Deutsche Forschungsgemeinschaft]; HA7070/4 [Deutsche Forschungsgemeinschaft]; ERA PerMed project "IMPLEMENT" [ERA-net Cofound]; [Fondation Leenaards]; 802998 [H2020 European Research Council]; 1117747 [HDH Wills 1965 Charitable Trust]; 2015073 [Helse Sør-Øst RHF]; 2019107 [Helse Sør-Øst RHF]; AMSP 07 [Interdisciplinary Center for Clinical Research of the Jena University hospital]; MzH 3/020/20 [Interdisciplinary Center for Clinical Research of the Medical Faculty of Münster]; G1001354 [Medical Research Council]; MR/R024790/2 [Medical Research Council]; 223273 [Norges Forskningsråd]; 249795 [Norges Forskningsråd]; 273345 [Norges Forskningsråd]; 276082 [Norges Forskningsråd]; 32003B_135679 [Swiss National Science Foundation]; 32003B_159780 [Swiss National Science Foundation]; 324730_192755 [Swiss National Science Foundation]; CRSK-3_190185 [Swiss National Science Foundation]; PZ00P3_193658 [Swiss National Science Foundation] full_text_status: public publication: Human Brain Mapping event_location: United States issn: 1065-9471 citation: de Lange, Ann-Marie G; Anatürk, Melis; Rokicki, Jaroslav; Han, Laura KM; Franke, Katja; Alnaes, Dag; Ebmeier, Klaus P; ... Cole, James H; + view all <#> de Lange, Ann-Marie G; Anatürk, Melis; Rokicki, Jaroslav; Han, Laura KM; Franke, Katja; Alnaes, Dag; Ebmeier, Klaus P; Draganski, Bogdan; Kaufmann, Tobias; Westlye, Lars T; Hahn, Tim; Cole, James H; - view fewer <#> (2022) Mind the gap: Performance metric evaluation in brain-age prediction. Human Brain Mapping 10.1002/hbm.25837 <https://doi.org/10.1002/hbm.25837>. (In press). Green open access document_url: https://discovery.ucl.ac.uk/id/eprint/10146010/1/Cole_%20Mind%20the%20gap%20Performance%20metric%20evaluation%20in%20brain%20age%20prediction.pdf