Multiple imputation integrated to machine learning: predicting post-stroke recovery of ambulation after intensive inpatient rehabilitation
Abstract 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 wi...
Saved in:
| Main Authors: | , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2024-10-01
|
| Series: | Scientific Reports |
| Subjects: | |
| Online Access: | https://doi.org/10.1038/s41598-024-74537-8 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850204226009956352 |
|---|---|
| author | Alice Finocchi Silvia Campagnini Andrea Mannini Stefano Doronzio Marco Baccini Bahia Hakiki Donata Bardi Antonello Grippo Claudio Macchi Jorge Navarro Solano Michela Baccini Francesca Cecchi |
| author_facet | Alice Finocchi Silvia Campagnini Andrea Mannini Stefano Doronzio Marco Baccini Bahia Hakiki Donata Bardi Antonello Grippo Claudio Macchi Jorge Navarro Solano Michela Baccini Francesca Cecchi |
| author_sort | Alice Finocchi |
| collection | DOAJ |
| description | Abstract 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. |
| format | Article |
| id | doaj-art-21fcf14e78574b85adbb44e80da1e770 |
| institution | OA Journals |
| issn | 2045-2322 |
| language | English |
| publishDate | 2024-10-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj-art-21fcf14e78574b85adbb44e80da1e7702025-08-20T02:11:20ZengNature PortfolioScientific Reports2045-23222024-10-0114111410.1038/s41598-024-74537-8Multiple imputation integrated to machine learning: predicting post-stroke recovery of ambulation after intensive inpatient rehabilitationAlice Finocchi0Silvia Campagnini1Andrea Mannini2Stefano Doronzio3Marco Baccini4Bahia Hakiki5Donata Bardi6Antonello Grippo7Claudio Macchi8Jorge Navarro Solano9Michela Baccini10Francesca Cecchi11IRCCS Fondazione Don Carlo Gnocchi onlusIRCCS Fondazione Don Carlo Gnocchi onlusIRCCS Fondazione Don Carlo Gnocchi onlusIRCCS Fondazione Don Carlo Gnocchi onlusIRCCS Fondazione Don Carlo Gnocchi onlusIRCCS Fondazione Don Carlo Gnocchi onlusIRCCS Fondazione Don Carlo Gnocchi onlusIRCCS Fondazione Don Carlo Gnocchi onlusIRCCS Fondazione Don Carlo Gnocchi onlusIRCCS Fondazione Don Carlo Gnocchi onlusDepartment of Statistics, Computer Science, Applications, University of FlorenceIRCCS Fondazione Don Carlo Gnocchi onlusAbstract 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.https://doi.org/10.1038/s41598-024-74537-8AmbulationMachine learningMultiple imputationPredictionRehabilitationStroke |
| spellingShingle | Alice Finocchi Silvia Campagnini Andrea Mannini Stefano Doronzio Marco Baccini Bahia Hakiki Donata Bardi Antonello Grippo Claudio Macchi Jorge Navarro Solano Michela Baccini Francesca Cecchi Multiple imputation integrated to machine learning: predicting post-stroke recovery of ambulation after intensive inpatient rehabilitation Scientific Reports Ambulation Machine learning Multiple imputation Prediction Rehabilitation Stroke |
| title | Multiple imputation integrated to machine learning: predicting post-stroke recovery of ambulation after intensive inpatient rehabilitation |
| title_full | Multiple imputation integrated to machine learning: predicting post-stroke recovery of ambulation after intensive inpatient rehabilitation |
| title_fullStr | Multiple imputation integrated to machine learning: predicting post-stroke recovery of ambulation after intensive inpatient rehabilitation |
| title_full_unstemmed | Multiple imputation integrated to machine learning: predicting post-stroke recovery of ambulation after intensive inpatient rehabilitation |
| title_short | Multiple imputation integrated to machine learning: predicting post-stroke recovery of ambulation after intensive inpatient rehabilitation |
| title_sort | multiple imputation integrated to machine learning predicting post stroke recovery of ambulation after intensive inpatient rehabilitation |
| topic | Ambulation Machine learning Multiple imputation Prediction Rehabilitation Stroke |
| url | https://doi.org/10.1038/s41598-024-74537-8 |
| work_keys_str_mv | AT alicefinocchi multipleimputationintegratedtomachinelearningpredictingpoststrokerecoveryofambulationafterintensiveinpatientrehabilitation AT silviacampagnini multipleimputationintegratedtomachinelearningpredictingpoststrokerecoveryofambulationafterintensiveinpatientrehabilitation AT andreamannini multipleimputationintegratedtomachinelearningpredictingpoststrokerecoveryofambulationafterintensiveinpatientrehabilitation AT stefanodoronzio multipleimputationintegratedtomachinelearningpredictingpoststrokerecoveryofambulationafterintensiveinpatientrehabilitation AT marcobaccini multipleimputationintegratedtomachinelearningpredictingpoststrokerecoveryofambulationafterintensiveinpatientrehabilitation AT bahiahakiki multipleimputationintegratedtomachinelearningpredictingpoststrokerecoveryofambulationafterintensiveinpatientrehabilitation AT donatabardi multipleimputationintegratedtomachinelearningpredictingpoststrokerecoveryofambulationafterintensiveinpatientrehabilitation AT antonellogrippo multipleimputationintegratedtomachinelearningpredictingpoststrokerecoveryofambulationafterintensiveinpatientrehabilitation AT claudiomacchi multipleimputationintegratedtomachinelearningpredictingpoststrokerecoveryofambulationafterintensiveinpatientrehabilitation AT jorgenavarrosolano multipleimputationintegratedtomachinelearningpredictingpoststrokerecoveryofambulationafterintensiveinpatientrehabilitation AT michelabaccini multipleimputationintegratedtomachinelearningpredictingpoststrokerecoveryofambulationafterintensiveinpatientrehabilitation AT francescacecchi multipleimputationintegratedtomachinelearningpredictingpoststrokerecoveryofambulationafterintensiveinpatientrehabilitation |