Improvement of Predictive Accuracies of Functional Outcomes After Subacute Stroke Inpatient Rehabilitation by Machine Learning Models

Overview

Journal PLoS One

Specialties General Medicine
Science

Date 2023 May 26

PMID 37235575

Authors

Yuta Miyazaki

Michiyuki Kawakami

Kunitsugu Kondo

Masahiro Tsujikawa

Kaoru Honaga

Kanjiro Suzuki

Tetsuya Tsuji

Affiliations

Soon will be listed here.

Abstract

Objectives: Stepwise linear regression (SLR) is the most common approach to predicting activities of daily living at discharge with the Functional Independence Measure (FIM) in stroke patients, but noisy nonlinear clinical data decrease the predictive accuracies of SLR. Machine learning is gaining attention in the medical field for such nonlinear data. Previous studies reported that machine learning models, regression tree (RT), ensemble learning (EL), artificial neural networks (ANNs), support vector regression (SVR), and Gaussian process regression (GPR), are robust to such data and increase predictive accuracies. This study aimed to compare the predictive accuracies of SLR and these machine learning models for FIM scores in stroke patients.

Methods: Subacute stroke patients (N = 1,046) who underwent inpatient rehabilitation participated in this study. Only patients' background characteristics and FIM scores at admission were used to build each predictive model of SLR, RT, EL, ANN, SVR, and GPR with 10-fold cross-validation. The coefficient of determination (R2) and root mean square error (RMSE) values were compared between the actual and predicted discharge FIM scores and FIM gain.

Results: Machine learning models (R2 of RT = 0.75, EL = 0.78, ANN = 0.81, SVR = 0.80, GPR = 0.81) outperformed SLR (0.70) to predict discharge FIM motor scores. The predictive accuracies of machine learning methods for FIM total gain (R2 of RT = 0.48, EL = 0.51, ANN = 0.50, SVR = 0.51, GPR = 0.54) were also better than of SLR (0.22).

Conclusions: This study suggested that the machine learning models outperformed SLR for predicting FIM prognosis. The machine learning models used only patients' background characteristics and FIM scores at admission and more accurately predicted FIM gain than previous studies. ANN, SVR, and GPR outperformed RT and EL. GPR could have the best predictive accuracy for FIM prognosis.

Citing Articles

Logistic regression analysis and machine learning for predicting post-stroke gait independence: a retrospective study.

Miyazaki Y, Kawakami M, Kondo K, Hirabe A, Kamimoto T, Akimoto T Sci Rep. 2024; 14(1):21273.

PMID: 39261645 PMC: 11390880. DOI: 10.1038/s41598-024-72206-4.

Prognostic Equations and Accuracy of a Total Score of Functional Independence Measure at Discharge for Different Diseases in a Convalescent Rehabilitation Ward.

Mikayama S, Kubo T, Tahara T, Nakamura M, Oku F, Kenmochi K Cureus. 2024; 16(8):e66509.

PMID: 39252717 PMC: 11382432. DOI: 10.7759/cureus.66509.

Leveraging Emerging Technologies to Expand Accessibility and Improve Precision in Rehabilitation and Exercise for People with Disabilities.

Willingham T, Stowell J, Collier G, Backus D Int J Environ Res Public Health. 2024; 21(1).

PMID: 38248542 PMC: 10815484. DOI: 10.3390/ijerph21010079.

References

Sonoda S, Chino N, Domen K, Saitoh E . Changes in impairment and disability from the third to the sixth month after stroke and its relationship evaluated by an artificial neural network. Am J Phys Med Rehabil. 1997; 76(5):395-400. DOI: 10.1097/00002060-199709000-00010. View

Shahid N, Rappon T, Berta W . Applications of artificial neural networks in health care organizational decision-making: A scoping review. PLoS One. 2019; 14(2):e0212356. PMC: 6380578. DOI: 10.1371/journal.pone.0212356. View

Nawabi J, Kniep H, Elsayed S, Friedrich C, Sporns P, Rusche T . Imaging-Based Outcome Prediction of Acute Intracerebral Hemorrhage. Transl Stroke Res. 2021; 12(6):958-967. PMC: 8557152. DOI: 10.1007/s12975-021-00891-8. View

Mainali S, Darsie M, Smetana K . Machine Learning in Action: Stroke Diagnosis and Outcome Prediction. Front Neurol. 2021; 12:734345. PMC: 8685212. DOI: 10.3389/fneur.2021.734345. View

Moon S, Ahmadnezhad P, Song H, Thompson J, Kipp K, Akinwuntan A . Artificial neural networks in neurorehabilitation: A scoping review. NeuroRehabilitation. 2020; 46(3):259-269. DOI: 10.3233/NRE-192996. View

Lee S, Adans-Dester C, OBrien A, Vergara-Diaz G, Black-Schaffer R, Zafonte R . Predicting and Monitoring Upper-Limb Rehabilitation Outcomes Using Clinical and Wearable Sensor Data in Brain Injury Survivors. IEEE Trans Biomed Eng. 2020; 68(6):1871-1881. PMC: 8723794. DOI: 10.1109/TBME.2020.3027853. View

Chumney D, Nollinger K, Shesko K, Skop K, Spencer M, Newton R . Ability of Functional Independence Measure to accurately predict functional outcome of stroke-specific population: systematic review. J Rehabil Res Dev. 2010; 47(1):17-29. DOI: 10.1682/jrrd.2009.08.0140. View

Lancashire L, Lemetre C, Ball G . An introduction to artificial neural networks in bioinformatics--application to complex microarray and mass spectrometry datasets in cancer studies. Brief Bioinform. 2009; 10(3):315-29. DOI: 10.1093/bib/bbp012. View

Gomez-Moreno H, Gil-Jimenez P, Lafuente-Arroyo S, Lopez-Sastre R, Maldonado-Bascon S . A "salt and pepper" noise reduction scheme for digital images based on Support Vector Machines classification and regression. ScientificWorldJournal. 2014; 2014:826405. PMC: 4151368. DOI: 10.1155/2014/826405. View

10.

Carino-Escobar R, Valdes-Cristerna R, Carrillo-Mora P, Rodriguez-Barragan M, Hernandez-Arenas C, Quinzanos-Fresnedo J . Prognosis of stroke upper limb recovery with physiological variables using regression tree ensembles. J Neural Eng. 2021; 18(4). DOI: 10.1088/1741-2552/abfc1e. View

11.

Lee S, Mortazavi B, Hoffman H, Lu D, Li C, Paak B . A Prediction Model for Functional Outcomes in Spinal Cord Disorder Patients Using Gaussian Process Regression. IEEE J Biomed Health Inform. 2014; 20(1):91-9. DOI: 10.1109/JBHI.2014.2372777. View

12.

Fujiwara T, Liu M, Tsuji T, Sonoda S, Mizuno K, Akaboshi K . Development of a new measure to assess trunk impairment after stroke (trunk impairment scale): its psychometric properties. Am J Phys Med Rehabil. 2004; 83(9):681-8. DOI: 10.1097/01.phm.0000137308.10562.20. View

13.

Sale P, Ferriero G, Ciabattoni L, Cortese A, Ferracuti F, Romeo L . Predicting Motor and Cognitive Improvement Through Machine Learning Algorithm in Human Subject that Underwent a Rehabilitation Treatment in the Early Stage of Stroke. J Stroke Cerebrovasc Dis. 2018; 27(11):2962-2972. DOI: 10.1016/j.jstrokecerebrovasdis.2018.06.021. View

14.

Yamada S, Liu M, Hase K, Tanaka N, Fujiwara T, Tsuji T . Development of a short version of the motor FIM for use in long-term care settings. J Rehabil Med. 2006; 38(1):50-6. DOI: 10.1080/16501970510044034. View

15.

Tsuji T, Sonoda S, Domen K, Saitoh E, Liu M, Chino N . ADL structure for stroke patients in Japan based on the functional independence measure. Am J Phys Med Rehabil. 1995; 74(6):432-8. DOI: 10.1097/00002060-199511000-00007. View

16.

Sargent D . Comparison of artificial neural networks with other statistical approaches: results from medical data sets. Cancer. 2001; 91(8 Suppl):1636-42. DOI: 10.1002/1097-0142(20010415)91:8+<1636::aid-cncr1176>3.0.co;2-d. View

17.

Heo J, Yoon J, Park H, Kim Y, Nam H, Heo J . Machine Learning-Based Model for Prediction of Outcomes in Acute Stroke. Stroke. 2019; 50(5):1263-1265. DOI: 10.1161/STROKEAHA.118.024293. View

18.

Lin W, Chen C, Tseng Y, Tsai Y, Chang C, Wang H . Predicting post-stroke activities of daily living through a machine learning-based approach on initiating rehabilitation. Int J Med Inform. 2018; 111:159-164. DOI: 10.1016/j.ijmedinf.2018.01.002. View

19.

Deo R . Machine Learning in Medicine. Circulation. 2015; 132(20):1920-30. PMC: 5831252. DOI: 10.1161/CIRCULATIONAHA.115.001593. View

20.

Belliveau T, Jette A, Seetharama S, Axt J, Rosenblum D, Larose D . Developing Artificial Neural Network Models to Predict Functioning One Year After Traumatic Spinal Cord Injury. Arch Phys Med Rehabil. 2016; 97(10):1663-1668.e3. DOI: 10.1016/j.apmr.2016.04.014. View