{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,13]],"date-time":"2026-09-13T23:03:37Z","timestamp":1789340617629,"version":"build-2803163510"},"reference-count":25,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2026,6,6]],"date-time":"2026-06-06T00:00:00Z","timestamp":1780704000000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,9,14]],"date-time":"2026-09-14T00:00:00Z","timestamp":1789344000000},"content-version":"vor","delay-in-days":100,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["npj Digit. Med."],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>\n                    Whether individual-level cognitive trajectories in Parkinson's disease are predictable remains unresolved. Here, we provide convergent evidence that measurement fidelity, rather than model complexity, governs the prediction ceiling. We evaluated ten machine learning paradigm families across 26 configurations in two independent cohorts\u2014the Parkinson\u2019s Progression Markers Initiative (PPMI,\n                    <jats:italic>N<\/jats:italic>\n                    \u2009=\u20091018) and the National Alzheimer\u2019s Coordinating Center (NACC,\n                    <jats:italic>N<\/jats:italic>\n                    \u2009=\u2009523). Motor subtype classification succeeded as a positive control (AUROC\u2009=\u20090.869), while cognitive trajectory prediction from the Montreal Cognitive Assessment remained uninformative (AUROC\u2009=\u20090.581) across all methods, including 42 genetic features. Synthetic experiments confirmed that models recover trajectories at high signal-to-noise ratios but collapse uniformly at levels present in clinical screening data. Replacing coarse screening with detailed neuropsychological tests improved prediction in both cohorts (PPMI: Symbol Digit Modalities Test AUROC\u2009=\u20090.725 vs. MoCA 0.596; NACC: Logical Memory AUROC\u2009=\u20090.684 vs. MMSE 0.648). Within-patient trajectory\n                    <jats:italic>R<\/jats:italic>\n                    \u00b2 = 0.20 for MoCA domains confirmed that approximately 80% of score variance represents non-signal variance. These findings redirect the clinical AI agenda from algorithm development toward measurement innovation: higher-frequency, higher-fidelity cognitive instruments are necessary before individual-level prediction becomes feasible.\n                  <\/jats:p>","DOI":"10.1038\/s41746-026-02681-8","type":"journal-article","created":{"date-parts":[[2026,6,6]],"date-time":"2026-06-06T06:34:07Z","timestamp":1780727647000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":1,"title":["Measurement fidelity, not model complexity, governs individual cognitive trajectory prediction in Parkinson's disease"],"prefix":"10.1038","volume":"9","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-2332-924X","authenticated-orcid":false,"given":"Wei","family":"Lin","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Sanjeet S.","family":"Grewal","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2026,6,6]]},"reference":[{"key":"2681_CR1","doi-asserted-by":"publisher","first-page":"47","DOI":"10.1038\/s41572-021-00280-3","volume":"7","author":"D Aarsland","year":"2021","unstructured":"Aarsland, D. et al. Parkinson disease-associated cognitive impairment. Nat. Rev. Dis. Prim. 7, 47 (2021).","journal-title":"Nat. Rev. Dis. Prim."},{"key":"2681_CR2","doi-asserted-by":"publisher","first-page":"863","DOI":"10.1001\/jamaneurol.2015.0703","volume":"72","author":"SM Fereshtehnejad","year":"2015","unstructured":"Fereshtehnejad, S. M. et al. New clinical subtypes of Parkinson disease and their longitudinal progression. JAMA Neurol. 72, 863\u2013873 (2015).","journal-title":"JAMA Neurol."},{"key":"2681_CR3","doi-asserted-by":"publisher","first-page":"319","DOI":"10.1038\/s41586-024-08328-6","volume":"637","author":"N Hollmann","year":"2025","unstructured":"Hollmann, N. et al. Accurate predictions on small data with a tabular foundation model. Nature 637, 319\u2013326 (2025).","journal-title":"Nature"},{"key":"2681_CR4","doi-asserted-by":"crossref","unstructured":"Tang, H., Liu, J., Zhao, M. & Gong, X. Progressive layered extraction (PLE): a novel multi-task learning (MTL) model for personalized recommendations. In Proc. ACM Conference on Recommender Systems (ACM, 2020).","DOI":"10.1145\/3383313.3412236"},{"key":"2681_CR5","doi-asserted-by":"crossref","unstructured":"Ma, J. et al. Modeling task relationships in multi-task learning with multi-gate mixture-of-experts. In Proc. ACM SIGKDD International Conference on Knowledge Discovery & Data Mining (ACM, 2018).","DOI":"10.1145\/3219819.3220007"},{"key":"2681_CR6","unstructured":"Rubanova, Y. Chen, R. T. Q. & Duvenaud, D. Latent ordinary differential equations for irregularly-sampled time series. In Proc. Advances in Neural Information Processing Systems (ACM, 2019)."},{"key":"2681_CR7","unstructured":"Yu, Y., Chen, J., Gao, T. & Yu, M. DAG-GNN: DAG structure learning with graph neural networks. In Proc. International Conference on Machine Learning (PMLR, 2019)."},{"key":"2681_CR8","unstructured":"Zheng, X., Aragam, B., Ravikumar, P. & Xing, E. P. DAGs with NO TEARS: continuous optimization for structure learning. In Proc. 32nd International Conference on Neural Information Processing Systems (ACM, 2018)."},{"key":"2681_CR9","doi-asserted-by":"publisher","first-page":"908","DOI":"10.1016\/S1474-4422(17)30328-9","volume":"16","author":"JC Latourelle","year":"2017","unstructured":"Latourelle, J. C. et al. Large-scale identification of clinical and genetic predictors of motor progression in patients with newly diagnosed Parkinson\u2019s disease: a longitudinal cohort study and validation. Lancet Neurol. 16, 908\u2013916 (2017).","journal-title":"Lancet Neurol."},{"key":"2681_CR10","doi-asserted-by":"publisher","first-page":"e17819","DOI":"10.2196\/17819","volume":"8","author":"F Faghri","year":"2020","unstructured":"Faghri, F. et al. Predicting onset of dementia using clinical notes and machine learning: case-control study. JMIR Med. Inf. 8, e17819 (2020).","journal-title":"JMIR Med. Inf."},{"key":"2681_CR11","first-page":"64","volume":"103","author":"KP Nguyen","year":"2022","unstructured":"Nguyen, K. P. et al. Predicting Parkinson\u2019s disease trajectory using clinical and neuroimaging baseline measures. Park. Relat. Disord. 103, 64\u201369 (2022).","journal-title":"Park. Relat. Disord."},{"key":"2681_CR12","doi-asserted-by":"crossref","unstructured":"Grinsztajn, L., Oyallon, E. & Varoquaux, G. Why do tree-based models still outperform deep learning on typical tabular data? In Proc. 36th International Conference on Neural Information Processing Systems (ACM, 2022).","DOI":"10.52202\/068431-0037"},{"key":"2681_CR13","doi-asserted-by":"publisher","first-page":"2501","DOI":"10.1002\/mds.23362","volume":"25","author":"KL Chou","year":"2010","unstructured":"Chou, K. L. et al. A recommended scale for cognitive screening in clinical trials of Parkinson\u2019s disease. Mov. Disord. 25, 2501\u20132507 (2010).","journal-title":"Mov. Disord."},{"key":"2681_CR14","doi-asserted-by":"publisher","first-page":"1717","DOI":"10.1212\/WNL.0b013e3181fc29c9","volume":"75","author":"JC Dalrymple-Alford","year":"2010","unstructured":"Dalrymple-Alford, J. C. et al. The MoCA: well-suited screen for cognitive impairment in Parkinson disease. Neurology 75, 1717\u20131725 (2010).","journal-title":"Neurology"},{"key":"2681_CR15","doi-asserted-by":"publisher","first-page":"695","DOI":"10.1111\/j.1532-5415.2005.53221.x","volume":"53","author":"ZS Nasreddine","year":"2005","unstructured":"Nasreddine, Z. S. et al. The Montreal Cognitive Assessment, MoCA: a brief screening tool for mild cognitive impairment. J. Am. Geriatr. Soc. 53, 695\u2013699 (2005).","journal-title":"J. Am. Geriatr. Soc."},{"key":"2681_CR16","doi-asserted-by":"publisher","first-page":"392","DOI":"10.1093\/brain\/aws318","volume":"136","author":"SE Winder-Rhodes","year":"2013","unstructured":"Winder-Rhodes, S. E. et al. Glucocerebrosidase mutations influence the natural history of Parkinson\u2019s disease in a community-based incident cohort. Brain 136, 392\u2013399 (2013).","journal-title":"Brain"},{"key":"2681_CR17","doi-asserted-by":"publisher","first-page":"1287","DOI":"10.1002\/mds.27376","volume":"33","author":"F Lipsmeier","year":"2018","unstructured":"Lipsmeier, F. et al. Evaluation of smartphone-based testing to generate exploratory outcome measures in a phase 1 Parkinson\u2019s disease clinical trial. Mov. Disord. 33, 1287\u20131297 (2018).","journal-title":"Mov. Disord."},{"key":"2681_CR18","doi-asserted-by":"publisher","first-page":"46","DOI":"10.1186\/s12984-016-0154-5","volume":"13","author":"S Del Din","year":"2016","unstructured":"Del Din, S. et al. Free-living gait characteristics in ageing and Parkinson\u2019s disease: impact of environment and ambulatory bout length. J. Neuroeng. Rehabil. 13, 46 (2016).","journal-title":"J. Neuroeng. Rehabil."},{"key":"2681_CR19","doi-asserted-by":"publisher","first-page":"64","DOI":"10.1038\/s41531-024-00721-2","volume":"10","author":"JL Adams","year":"2024","unstructured":"Adams, J. L. et al. Using a smartwatch and smartphone to assess early Parkinson\u2019s disease in the WATCH-PD study over 12 months. npj Parkinsons Dis. 10, 64 (2024).","journal-title":"npj Parkinsons Dis."},{"key":"2681_CR20","doi-asserted-by":"publisher","first-page":"3613","DOI":"10.3390\/s24113613","volume":"24","author":"D Trabassi","year":"2024","unstructured":"Trabassi, D. et al. Optimizing rare disease gait classification through data balancing and generative AI: insights from hereditary cerebellar ataxia. Sensors 24, 3613 (2024).","journal-title":"Sensors"},{"key":"2681_CR21","doi-asserted-by":"publisher","first-page":"n2281","DOI":"10.1136\/bmj.n2281","volume":"375","author":"CL Andaur Navarro","year":"2021","unstructured":"Andaur Navarro, C. L. et al. Risk of bias in studies on prediction models developed using supervised machine learning techniques: systematic review. BMJ 375, n2281 (2021).","journal-title":"BMJ"},{"key":"2681_CR22","doi-asserted-by":"publisher","first-page":"e048008","DOI":"10.1136\/bmjopen-2020-048008","volume":"11","author":"GS Collins","year":"2021","unstructured":"Collins, G. S. et al. Protocol for development of a reporting guideline (TRIPOD-AI) and risk of bias tool (PROBAST-AI) for diagnostic and prognostic prediction model studies based on artificial intelligence. BMJ Open 11, e048008 (2021).","journal-title":"BMJ Open"},{"key":"2681_CR23","doi-asserted-by":"publisher","first-page":"171","DOI":"10.1242\/dmm.015123","volume":"7","author":"N Matosin","year":"2014","unstructured":"Matosin, N., Frank, E., Engel, M., Lum, J. S. & Newell, K. A. Negativity towards negative results: a discussion of the disconnect between scientific worth and scientific culture. Dis. Model Mech. 7, 171\u2013173 (2014).","journal-title":"Dis. Model Mech."},{"key":"2681_CR24","doi-asserted-by":"publisher","first-page":"24","DOI":"10.1093\/jrsssb\/qkac004","volume":"85","author":"E Candes","year":"2023","unstructured":"Candes, E., Lei, L. & Ren, Z. Conformalized survival analysis. J. R. Stat. Soc. Ser. B 85, 24\u201345 (2023).","journal-title":"J. R. Stat. Soc. Ser. B"},{"key":"2681_CR25","doi-asserted-by":"crossref","unstructured":"Spirtes, P., Glymour, C. N. & Scheines, R. Causation, Prediction, and Search. 2nd edn (MIT Press, 2000).","DOI":"10.7551\/mitpress\/1754.001.0001"}],"container-title":["npj Digital Medicine"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.nature.com\/articles\/s41746-026-02681-8","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41746-026-02681-8.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41746-026-02681-8.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,13]],"date-time":"2026-09-13T22:22:01Z","timestamp":1789338121000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.nature.com\/articles\/s41746-026-02681-8"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,6,6]]},"references-count":25,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2026,12]]}},"alternative-id":["2681"],"URL":"https:\/\/doi.org\/10.1038\/s41746-026-02681-8","relation":{},"ISSN":["2398-6352"],"issn-type":[{"value":"2398-6352","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,6,6]]},"assertion":[{"value":"10 March 2026","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"17 April 2026","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"6 June 2026","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"The authors declare no competing interests.","order":1,"name":"Ethics","label":"Competing interests","group":{"name":"EthicsHeading","label":"Ethics"}},{"value":"Computational code used in this study was developed with the assistance of Claude (Anthropic), a large language model. All analyses were designed, executed, verified, and interpreted by the first author. The AI tool was used solely for code generation assistance and did not contribute to study design, data interpretation, or scientific conclusions.","order":2,"name":"Ethics","label":"AI Disclosure","group":{"name":"EthicsHeading","label":"Ethics"}}],"article-number":"698"}}