Multiple imputation integrated to machine learning: predicting post-stroke recovery of ambulation after intensive inpatient rehabilitation.
Ambulation
Machine learning
Multiple imputation
Prediction
Rehabilitation
Stroke
Journal
Scientific reports
ISSN: 2045-2322
Titre abrégé: Sci Rep
Pays: England
ID NLM: 101563288
Informations de publication
Date de publication:
24 Oct 2024
24 Oct 2024
Historique:
received:
15
05
2024
accepted:
26
09
2024
medline:
25
10
2024
pubmed:
25
10
2024
entrez:
25
10
2024
Statut:
epublish
Résumé
Good data quality is vital for personalising plans in rehabilitation. Machine learning (ML) improves prognostics but integrating it with Multiple Imputation (MImp) for dealing missingness is an unexplored field. This work aims to provide post-stroke ambulation prognosis, integrating MImp with ML, and identify the prognostic influential factors. Stroke survivors in intensive rehabilitation were enrolled. Data on demographics, events, clinical, physiotherapy, and psycho-social assessment were collected. An independent ambulation at discharge, using the Functional Ambulation Category scale, was the outcome. After handling missingness using MImp, ML models were optimised, cross-validated, and tested. Interpretability techniques analysed predictor contributions. Pre-MImp, the dataset included 54.1% women, 79.2% ischaemic patients, median age 80.0 (interquartile range: 15.0). Post-MImp, 368 non-ambulatory patients on 10 imputed datasets were used for training, 80 for testing. The random forest (the validation best-performing algorithm) obtained 75.5% aggregated balanced accuracy on the test set. The main predictors included modified Barthel index, Fugl-Meyer assessment/motricity index, short physical performance battery, age, Charlson comorbidity index/cumulative illness rating scale, and trunk control test. This is among the first studies applying ML, together with MImp, to predict ambulation recovery in post-stroke rehabilitation. This pipeline reliably exploits the potential of incomplete datasets for healthcare prognosis, identifying relevant predictors.
Identifiants
pubmed: 39448629
doi: 10.1038/s41598-024-74537-8
pii: 10.1038/s41598-024-74537-8
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
25188Subventions
Organisme : Ministero dell'Università e della Ricerca
ID : Ricerca Corrente
Organisme : Ministero dell'Università e della Ricerca
ID : PNC0000007
Informations de copyright
© 2024. The Author(s).
Références
Rinott, R. et al. Prognostic data-driven clinical decision support - formulation and implications. Stud. Health Technol. Inf.169, 140–144 (2011).
Baptista, M., Goebel, K. & Henriques, E. Relation between Prognostics Predictor Evaluation Metrics and local interpretability SHAP values. Artif. Intell. 306, 103667 (2022).
doi: 10.1016/j.artint.2022.103667
Ostropolets, A., Zhang, L. & Hripcsak, G. A scoping review of clinical decision support tools that generate new knowledge to support decision making in real time. J. Am. Med. Inf. Assoc. JAMIA. 27, 1968–1976 (2020).
doi: 10.1093/jamia/ocaa200
Rincé, G. et al. Impact of an individual personalised rehabilitation program on mobility performance in older-old people. Aging Clin. Exp. Res. 33, 2821–2830 (2021).
pubmed: 33625687
doi: 10.1007/s40520-021-01812-3
Dür, M., Wenzel, C., Simon, P. & Tucek, G. Patients’ and professionals’ perspectives on the consideration of patients’ convenient therapy periods as part of personalised rehabilitation: a focus group study with patients and therapists from inpatient neurological rehabilitation. BMC Health Serv. Res. 22, 372 (2022).
pubmed: 35313879
pmcid: 8939130
doi: 10.1186/s12913-022-07755-3
Kokkotis, C. et al. Machine learning techniques for the prediction of functional outcomes in the Rehabilitation of Post-stroke patients: a scoping review. BioMed. 3, 1–20 (2023).
doi: 10.3390/biomed3010001
Campagnini, S. et al. Machine learning methods for functional recovery prediction and prognosis in post-stroke rehabilitation: a systematic review. J. Neuroeng. Rehabil. 19, 54 (2022).
pubmed: 35659246
pmcid: 9166382
doi: 10.1186/s12984-022-01032-4
Habehh, H., Gohel, S. Machine Learning in Healthcare. Curr Genomics. 22(4), 291–300. https://doi.org/10.2174/1389202922666210705124359 (2021).
Madley-Dowd, P., Hughes, R., Tilling, K. & Heron, J. The proportion of missing data should not be used to guide decisions on multiple imputation. J. Clin. Epidemiol. 110, 63–73 (2019).
pubmed: 30878639
pmcid: 6547017
doi: 10.1016/j.jclinepi.2019.02.016
Zhang, N. Journal of Hospital Medicine Leadership Team. Methodolgical Progress Note: Handling Missing Data in Clinical Research. J Hosp Med. 14(4), 237–239. https://doi.org/10.12788/jhm.3330 (2020).
Kang, H. The prevention and handling of the missing data. Korean J. Anesthesiol. 64, 402–406 (2013).
pubmed: 23741561
pmcid: 3668100
doi: 10.4097/kjae.2013.64.5.402
Shao, X. et al. Development and validation of risk prediction models for stroke and mortality among patients with type 2 diabetes in northern China. J. Endocrinol. Invest. 46, 271–283 (2023).
pubmed: 35972686
doi: 10.1007/s40618-022-01898-0
Rubin, D. B. Inference and Missing Data. Biometrika. 63, 581–592 (1976).
doi: 10.1093/biomet/63.3.581
Navarro, C. L. A. et al. Risk of bias in studies on prediction models developed using supervised machine learning techniques: systematic review. BMJ. 375, n2281 (2021).
doi: 10.1136/bmj.n2281
Rios, R. et al. Handling missing values in machine learning to predict patient-specific risk of adverse cardiac events: insights from REFINE SPECT registry. Comput. Biol. Med. 145, 105449 (2022).
pubmed: 35381453
pmcid: 9117456
doi: 10.1016/j.compbiomed.2022.105449
Hoogland, J. et al. Handling missing predictor values when validating and applying a prediction model to new patients. Stat. Med. 39, 3591–3607 (2020).
pubmed: 32687233
pmcid: 7586995
doi: 10.1002/sim.8682
Chang, M. C. et al. The Use of Artificial Intelligence to predict the prognosis of patients undergoing Central Nervous System Rehabilitation: a narrative review. Healthcare. 11, 2687 (2023).
pubmed: 37830724
pmcid: 10572243
doi: 10.3390/healthcare11192687
Hakiki, B. et al. Predictors of function, activity, and participation of Stroke patients undergoing Intensive Rehabilitation: a Multicenter prospective observational study protocol. Front. Neurol. 12, 632672 (2021).
pubmed: 33897593
pmcid: 8060493
doi: 10.3389/fneur.2021.632672
Chiavilli, M. et al. Design and implementation of a Stroke Rehabilitation Registry for the systematic assessment of processes and outcomes and the development of data-driven prediction models: the STRATEGY study protocol. Front. Neurol. 13, 919353 (2022).
pubmed: 36299268
pmcid: 9588928
doi: 10.3389/fneur.2022.919353
Bijleveld-Uitman, M., van de Port, I. & Kwakkel, G. Is gait speed or walking distance a better predictor for community walking after stroke? J. Rehabil Med. 45, 535–540 (2013).
pubmed: 23584080
doi: 10.2340/16501977-1147
Muren, M. A., Hütler, M. & Hooper, J. Functional capacity and health-related quality of life in individuals post stroke. Top. Stroke Rehabil. 15, 51–58 (2008).
pubmed: 18250074
doi: 10.1310/tsr1501-51
Salinas, J. et al. An International Standard Set of patient-centered outcome measures after stroke. Stroke. 47, 180–186 (2016).
pubmed: 26604251
doi: 10.1161/STROKEAHA.115.010898
Winstein, C.J., Stein, J., Arena, R., Bates, B., Cherney, L.R., Cramer, S.C., Deruyter, F., Eng, J.J., Fisher, B., Harvey, R.L., Lang, C.E., MacKay-Lyons, M., Ottenbacher, K.J., Pugh, S., Reeves, M.J., Richards, L.G., Stiers, W., Zorowitz, RD. American Heart Association Stroke Council, Council on Cardiovascular and Stroke Nursing, Council on Clinical Cardiology, and Council on Quality of Care and Outcomes Research. Guidelines for Adult Stroke Rehabilitation and Recovery: A Guideline for Healthcare Professionals From the American Heart Association/American Stroke Association. Stroke. 2016 Jun; 47(6), e98-e169. https://doi.org/10.1161/STR.0000000000000098 . [Erratum in: Stroke. 2017 Feb; 48(2):e78. https://doi.org/10.1161/STR.0000000000000120 ] [Erratum in: Stroke. 2017 Dec; 48(12):e369. https://doi.org/10.1161/STR.0000000000000156 ]
Inzitari, D. & Carlucci, G. Italian stroke guidelines (SPREAD): evidence and clinical practice. Neurol. Sci. 27, s225–s227 (2006).
pubmed: 16752053
doi: 10.1007/s10072-006-0622-y
Cecchi, F., Diverio, M., Arienti, C., Corbella, E., Marrazzo, F., Speranza, G., Del Zotto, E., Poggianti, G., Gigliotti, F., Polcaro, P., Zingoni, M., Antonioli, D., Avila, L., Barilli, M., Romano, E., Landucci Pellegrini, L., Gambini, M., Verdesca, S., Bertolucci, F., Mosca, I., Gemignani, P., Paperini, A., Castagnoli, C., Hochleitner, I., Luisi, M.L., Lucidi, G., Hakiki, B., Gabrielli, M.A., Fruzzetti, M., Bruzzi, A., Bacci Bonotti, E., Pancani, S., Galeri, S., Macchi, C., Aprile, I. Development and implementation of a stroke rehabilitation integrated care pathway in an Italian no profit institution: an observational study. Eur J Phys Rehabil Med. 56(6), 713–724. https://doi.org/10.23736/S1973-9087.20.06195-X (2020).
Chao, Y.S., Wu, C.J., Wu, H.C., McGolrick, D., Chen, WC. Interpretable Trials: Is Interpretability a Reason Why Clinical Trials Fail? Front Med (Lausanne). 8, 541405. https://doi.org/10.3389/fmed.2021.541405 (2021).
Geyh, S. et al. Identifying the concepts contained in outcome measures of clinical trials on stroke using the International classification of Functioning, Disability and Health as a reference. J. Rehabil Med. https://doi.org/10.1080/16501960410015399 (2004).
Shah, S., Vanclay, F. & Cooper, B. Improving the sensitivity of the Barthel Index for stroke rehabilitation. J. Clin. Epidemiol. 42, 703–709 (1989).
pubmed: 2760661
doi: 10.1016/0895-4356(89)90065-6
Nasreddine, Z. S. et al. The Montreal Cognitive Assessment, MoCA: a brief Screening Tool for mild cognitive impairment. J. Am. Geriatr. Soc. 53, 695–699 (2005).
pubmed: 15817019
doi: 10.1111/j.1532-5415.2005.53221.x
Folstein, M. F., Folstein, S. E. & McHugh, P. R. Mini-mental state’. A practical method for grading the cognitive state of patients for the clinician. J. Psychiatr Res. 12, 189–198 (1975).
pubmed: 1202204
doi: 10.1016/0022-3956(75)90026-6
Collen, F. M., Wade, D. T. & Bradshaw, C. M. Mobility after stroke: reliability of measures of impairment and disability. Int. Disabil. Stud. 12, 6–9 (1990).
pubmed: 2211468
doi: 10.3109/03790799009166594
Demeurisse, G., Demol, O. & Robaye, E. Motor evaluation in vascular hemiplegia. Eur. Neurol. 19, 382–389 (1980).
pubmed: 7439211
doi: 10.1159/000115178
Cecchi, F. et al. Transcultural translation and validation of Fugl-Meyer assessment to Italian. Disabil. Rehabil. 43, 3717–3722 (2021).
pubmed: 32356509
doi: 10.1080/09638288.2020.1746844
Charlson, M. E., Carrozzino, D., Guidi, J. & Patierno, C. Charlson Comorbidity Index: a critical review of Clinimetric Properties. Psychother. Psychosom. 91, 8–35 (2022).
pubmed: 34991091
doi: 10.1159/000521288
Parmelee, P. A., Thuras, P. D., Katz, I. R. & Lawton, M. P. Validation of the cumulative illness rating scale in a geriatric residential population. J. Am. Geriatr. Soc. 43, 130–137 (1995).
pubmed: 7836636
doi: 10.1111/j.1532-5415.1995.tb06377.x
Aiello, E. N., Pasotti, F., Appollonio, I. & Bolognini, N. Equating Mini-mental State Examination (MMSE) and Montreal Cognitive Assessment (MoCA) scores: conversion norms from a healthy Italian population sample. Aging Clin. Exp. Res. 34, 1721–1724 (2022).
pubmed: 35182351
doi: 10.1007/s40520-022-02089-w
Carpinelli Mazzi, M. et al. Mini-mental state examination: new normative values on subjects in Southern Italy. Aging Clin. Exp. Res. 32, 699–702 (2020).
pubmed: 31230268
doi: 10.1007/s40520-019-01250-2
Little, R. J. A. & Rubin, D. B. Statistical Analysis with Missing Data (Wiley, 2019).
van Buuren, S. & Groothuis-Oudshoorn, K. Mice: multivariate imputation by chained equations in R. J. Stat. Softw. 45, 1–67 (2011).
doi: 10.18637/jss.v045.i03
Pedregosa, F. et al. Scikit-learn: machine learning in Python. J. Mach. Learn. Res. 12, 2825–2830 (2011).
Shapley, L. S. A Value for N-Person Games. https://www.rand.org/pubs/papers/P295.html (1952).
Marino, M., Lucas, J., Latour, E. & Heintzman, J. D. Missing data in primary care research: importance, implications and approaches. Fam Pract. 38, 199–202 (2021).
pmcid: 8243609
doi: 10.1093/fampra/cmaa134
Di, J. et al. Considerations to address missing data when deriving clinical trial endpoints from digital health technologies. Contemp. Clin. Trials. 113, 106661 (2022).
pubmed: 34954098
doi: 10.1016/j.cct.2021.106661
Cecchi, F. et al. Redefining a minimal assessment protocol for stroke rehabilitation: the new ‘Protocollo Di Minima per l’ICtus’ (PMIC2020). Eur. J. Phys. Rehabil Med. 57, 669–676 (2021).
pubmed: 34042407
doi: 10.23736/S1973-9087.21.06638-7
Moss, L., Corsar, D., Shaw, M., Piper, I. & Hawthorne, C. Demystifying the Black Box: the importance of interpretability of Predictive models in Neurocritical Care. Neurocrit Care. 37, 185–191 (2022).
pubmed: 35523917
pmcid: 9343258
doi: 10.1007/s12028-022-01504-4
Holden, M. K., Gill, K. M., Magliozzi, M. R., Nathan, J. & Piehl-Baker, L. Clinical gait assessment in the neurologically impaired. Reliability and meaningfulness. Phys. Ther. 64, 35–40 (1984).
pubmed: 6691052
doi: 10.1093/ptj/64.1.35
Sodero, A. et al. Predicting the functional outcome of intensive inpatient rehabilitation after stroke: results from the RIPS Study. Eur. J. Phys. Rehabil Med. 60, 1–12 (2024).
pubmed: 37934187
doi: 10.23736/S1973-9087.23.07852-8
Moore, S. A., Boyne, P., Fulk, G., Verheyden, G. & Fini, N. A. Walk the talk: current evidence for walking Recovery after Stroke, Future pathways and a Mission for Research and Clinical Practice. Stroke. 53, 3494–3505 (2022).
pubmed: 36069185
pmcid: 9613533
doi: 10.1161/STROKEAHA.122.038956
Mehrholz, J., Wagner, K., Rutte, K., Meiβner, D. & Pohl, M. Predictive validity and responsiveness of the functional ambulation category in Hemiparetic patients after Stroke. Arch. Phys. Med. Rehabil. 88, 1314–1319 (2007).
pubmed: 17908575
doi: 10.1016/j.apmr.2007.06.764
Smith, M. C., Barber, P. A. & Stinear, C. M. The TWIST Algorithm Predicts Time to walking independently after stroke. Neurorehabil Neural Repair. 31, 955–964 (2017).
pubmed: 29090654
doi: 10.1177/1545968317736820
Campagnini, S. et al. Cross-validation of predictive models for functional recovery after post-stroke rehabilitation. J. Neuroeng. Rehabil. 19, 96 (2022).
pubmed: 36071452
pmcid: 9454118
doi: 10.1186/s12984-022-01075-7
Hirano, Y. et al. Prediction of independent walking ability for severely hemiplegic stroke patients at Discharge from a Rehabilitation Hospital. J. Stroke Cerebrovasc. Dis. Off J. Natl. Stroke Assoc. 25, 1878–1881 (2016).
doi: 10.1016/j.jstrokecerebrovasdis.2015.12.020
Ishiwatari, M. et al. Prediction of gait independence using the trunk impairment scale in patients with acute stroke. Ther. Adv. Neurol. Disord. 15, 17562864221140180 (2022).
pubmed: 36506941
pmcid: 9730005
doi: 10.1177/17562864221140180
Cecchi, F. et al. Predictors of recovering ambulation after hip fracture inpatient rehabilitation. BMC Geriatr. 18, 201 (2018).
pubmed: 30170554
pmcid: 6117966
doi: 10.1186/s12877-018-0884-2
Selves, C., Stoquart, G. & Lejeune, T. Gait rehabilitation after stroke: review of the evidence of predictors, clinical outcomes and timing for interventions. Acta Neurol. Belg. 120, 783–790 (2020).
pubmed: 32166723
doi: 10.1007/s13760-020-01320-7
Guralnik, J. et al. A short physical performance battery assessing lower extremity function: Association with Self-reported disability and prediction of mortality and nursing home admission. J. Gerontol. 49, M85–94 (1994).
pubmed: 8126356
doi: 10.1093/geronj/49.2.M85
Harrison, J. K., McArthur, K. S. & Quinn, T. J. Assessment scales in stroke: clinimetric and clinical considerations. Clin. Interv. Aging. 8, 201–211 (2013).
pubmed: 23440256
pmcid: 3578502
Veerbeek, J.M., Pohl, J., Held, J.P.O., Luft, A.R. External Validation of the Early Prediction of Functional Outcome After Stroke Prediction Model for Independent Gait at 3 Months After Stroke. Front Neurol. 13, 797791. https://doi.org/10.3389/fneur.2022.797791 (2022).
Kim, J. K., Choo, Y. J., Shin, H., Choi, G. S. & Chang, M. C. Prediction of ambulatory outcome in patients with corona radiata infarction using deep learning. Sci. Rep. 11, 7989 (2021).
pubmed: 33846472
pmcid: 8041824
doi: 10.1038/s41598-021-87176-0
Shin, H., Kim, J. K., Choo, Y. J., Choi, G. S. & Chang, M. C. Prediction of Motor Outcome of Stroke patients using a deep learning algorithm with Brain MRI as Input Data. Eur. Neurol. 85, 460–466 (2022).
pubmed: 35738236
doi: 10.1159/000525222
Kim, J. K., Lv, Z., Park, D. & Chang, M. C. Practical machine learning model to predict the Recovery of Motor Function in patients with stroke. Eur. Neurol. 85, 273–279 (2022).
pubmed: 35350014
doi: 10.1159/000522254
Ruksakulpiwat, S. et al. Associations between diagnosis with stroke, comorbidities, and activity of daily living among older adults in the United States. Chronic Dis. Transl. Med. 9, 164–176 (2023).
pubmed: 37305105
pmcid: 10249194