Therapeutic Science for Rehabilitation (재활치료과학)

Korean Society of Neurological Occupational Therapy (대한신경계작업치료학회)
권호 표지
DOI QR코드

DOI QR Code

Predicting Functional Outcomes of Patients With Stroke Using Machine Learning: A Systematic Review

머신러닝을 활용한 뇌졸중 환자의 기능적 결과 예측: 체계적 고찰

  • Bae, Suyeong (Dept. of Occupational Therapy, Graduate School, Yonsei University) ;
  • Lee, Mi Jung (Dept. of Nutrition, Metabolism and Rehabilitation Sciences, School of Health Professions, University of Texas Medical Branch at Galveston) ;
  • Nam, Sanghun (Dept. of Occupational Therapy, Graduate School, Yonsei University) ;
  • Hong, Ickpyo (Dept. of Occupational Therapy, College of Software and Digital Healthcare Convergence, Yonsei University)
  • 배수영 (연세대학교 일반대학원 작업치료학과) ;
  • ;
  • 남상훈 (연세대학교 일반대학원 작업치료학과) ;
  • 홍익표 (연세대학교 소프트웨어디지털헬스케어융합대학 작업치료학과)
  • Received : 2022.06.24
  • Accepted : 2022.08.11
  • Published : 2022.11.30
Download PDF
⟨ Previous Next ⟩

Abstract

Objective : To summarize clinical and demographic variables and machine learning uses for predicting functional outcomes of patients with stroke. Methods : We searched PubMed, CINAHL and Web of Science to identify published articles from 2010 to 2021. The search terms were "machine learning OR data mining AND stroke AND function OR prediction OR/AND rehabilitation". Articles exclusively using brain imaging techniques, deep learning method and articles without available full text were excluded in this study. Results : Nine articles were selected for this study. Support vector machines (19.05%) and random forests (19.05%) were two most frequently used machine learning models. Five articles (55.56%) demonstrated that the impact of patient initial and/or discharge assessment scores such as modified ranking scale (mRS) or functional independence measure (FIM) on stroke patients' functional outcomes was higher than their clinical characteristics. Conclusions : This study showed that patient initial and/or discharge assessment scores such as mRS or FIM could influence their functional outcomes more than their clinical characteristics. Evaluating and reviewing initial and or discharge functional outcomes of patients with stroke might be required to develop the optimal therapeutic interventions to enhance functional outcomes of patients with stroke.

목적 : 본 연구는 뇌졸중 환자의 기능적 결과를 예측하기 위한 인구통계학적 및 임상학적 특징과 머신러닝의 사용을 체계적으로 분석하고 요약하기 위해 수행되었다. 연구방법 : PubMed, CINAHL과 Web of Science를 사용하여 2010년부터 2021년 사이에 게재된 연구를 검색하였다. 주요 검색어는 "machine learning OR data mining AND stroke AND function OR prediction OR/AND rehabilitation"을 사용하였다. 뇌 이미지 처리 기법만을 분석한 연구, 딥러닝만 적용한 연구와 전체 본문을 열람할 수 없는 연구는 제외되었다. 결과 : 검색한 결과, 총 9편의 국내외 논문을 선정했다. 선정된 논문에서 가장 많이 사용된 머신러닝 알고리즘은 서포트 벡터 머신(support vector machine, 19.05%)과 랜덤포레스트(random forest, 19.05%)였다. 9개 중 7개의 연구에서 뇌졸중 환자의 기능을 예측하기 위해 중요하다고 추출된 변수를 결과로 제시했다. 그 결과, 5개(55.56%)의 연구에서 뇌졸중 환자의 기능을 예측하기 위해 환자의 임상적 특성이 아닌 modified ranking scale (mRS) 및 functional independence measure (FIM)과 같은 초기 또는 퇴원 평가 점수가 중요하다고 도출되었다. 결론 : 이 연구는 mRS 및 FIM과 같은 뇌졸중 환자의 초기 또는 퇴원 평가 점수가 임상적 특성보다 기능적 결과에 더 많은 영향을 미칠 수 있음을 나타냈다. 따라서, 뇌졸중 환자의 기능적 결과를 향상시키기 위한 최적의 중재를 개발하고 적용하기 위해서는 뇌졸중 환자의 초기 및 퇴원 시 기능적 결과를 평가하고 검토하는 것이 필요하다.

Keywords

Acknowledgement

This work was supported by the Ministry of Education of the Republic of Korea and the National Research Foundation of Korea (NRF-2021S1A3A2A 02096338). In addition, this research was supported in part by grant# K12 HD055929 from the National Institutes of Health (NIH). Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the NIH.

References

  1. Al-Qazzaz, N. K., Ali, S. H., Ahmad, S. A., Islam, S., & Mohamad, K. (2014). Cognitive impairment and memory dysfunction after a stroke diagnosis: A post-stroke memory assessment. Neuropsychiatric Dsease and Treatment, 10, 1677-1691. http://doi.org/10.2147/NDT.S67184
  2. Alaka, S. A., Menon, B. K., Brobbey, A., Williamson, T., Goyal, M., Demchuk, A. M., Hill, M. D., & Sajobi, T. T. (2020). Functional outcome prediction in ischemic stroke: A comparison of machine learning algorithms and regression models. Frontiers in Neurology, 11. https://doi.org/10.3389/fneur.2020.00889
  3. American Occupational Therapy Association. (2020). Occupational therapy practice framework: Domain and process. American Journal of Occupational Therapy, 74(S2), 1-85. https://doi.org/10.5014/ajot.2020.74S2001
  4. Byeon, H. (2020). Is the Random Forest algorithm suitable for predicting Parkinson's disease with mild cognitive impairment out of Parkinson's disease with normal cognition? International Journal of Environmental Research and Public Health, 17(7), 2594-2608. https://doi.org/10.3390/ijerph17072594
  5. Caro, C. C., Costa, J. D., & da Cruz, D. M. C. (2018). Burden and quality of life of family caregivers of stroke patients. Occupational Therapy in Health Care, 32(2), 154-171. https://doi.org/10.1080/07380577.2018.1449046
  6. Cheong, M. J., Jeon, B., & Noh, S. E. (2020). A protocol for systematic review and meta-analysis on psychosocial factors related to rehabilitation motivation of stroke patients. Medicine, 99(52), e23727-e23727. http://doi.org/10.1097/MD.0000000000023727
  7. Clarke, D. J., & Forster, A. (2015). Improving post-stroke recovery: The role of the multidisciplinary health care team. Journal of Multidisciplinary Healthcare, 8, 433-442. http://doi.org/10.2147/JMDH.S68764
  8. Dworzynski, K., Ritchie, G., & Playford, E. D. (2015). Stroke rehabilitation: Long-term rehabilitation after stroke. Clinical Medicine, 15(5), 461-464. http://doi.org/10.7861/clinmedicine.15-5-461
  9. Elloker, T., Rhoda, A., Arowoiya, A., & Lawal, I. U. (2019). Factors predicting community participation in patients living with stroke, in the Western Cape, South Africa. Disability and Rehabilitation, 41(22), 2640-2647. https://doi.org/10.1080/09638288.2018.1473509
  10. Fishman, K. N., Ashbaugh, A. R., & Swartz, R. H. (2021). Goal setting improves cognitive performance in a randomized trial of chronic stroke survivors. Stroke, 52(2), 458-470. https://doi.org/10.1161/STROKEAHA.120.032131
  11. Harari, Y., O'Brien, M. K., Lieber, R. L., & Jayaraman, A. (2020). Inpatient stroke rehabilitation: Prediction of clinical outcomes using a machine-learning approach. Journal of NeuroEngineering and Rehabilitation, 17(1), 1-10. https://doi.org/10.1186/s12984-020-00704-3
  12. Heo, J., Yoon, J. G., Park, H., Kim, Y. D., Nam, H. S., & Heo, J. H. (2019). Machine learning-based model for prediction of outcomes in acute stroke. Stroke, 50(5), 1263-1265. https://doi.org/10.1161/STROKEAHA.118.024293
  13. Ij, H. (2018). Statistics versus machine learning. Nature Methods, 15(4), 233-234. https://doi.org/10.1038/nmeth.4642
  14. Iwamoto, Y., Imura, T., Tanaka, R., Imada, N., Inagawa, T., Araki, H., & Araki, O. (2020). Development and validation of machine learning-based prediction for dependence in the activities of daily living after stroke inpatient rehabilitation: A decision-tree analysis. Journal of Stroke and Cerebrovascular Diseases, 29(12), 1-6. https://doi.org/10.1016/j.jstrokecerebrovasdis.2020.105332
  15. Imura, T., Inoue, Y., Tanaka, R., Matsuba, J., & Umayahara, Y. (2021). Clinical features for identifying the possibility of toileting independence after convalescent inpatient rehabilitation in severe stroke patients: A decision tree analysis based on a nationwide Japan rehabilitation database. Journal of Stroke and Cerebrovascular Diseases, 30(2), 1-8. https://doi.org/10.1016/j.jstrokecerebrovasdis.2020.105483
  16. Jakkula, V. (2006). Tutorial on support vector machine (svm). School of EECS, Washington State University. https://course.ccs.neu.edu/cs5100f11/resources/jakkula.pdf
  17. Jordan, M. I., & Mitchell, T. M. (2015). Machine learning: Trends, perspectives, and prospects. Science, 349(6245), 255-260. http://doi.org/10.1126/science.aaa8415
  18. Korpershoek, C., van der Bijl, J., & Hafsteinsdottir, T. B. (2011). Self-efficacy and its influence on recovery of patients with stroke: A systematic review. Journal of Advanced Nursing, 67(9), 1876-1894. https://doi.org/10.1111/j.1365-2648.2011.05659.x
  19. Liao, W. W., Hsieh, Y. W., Lee, T. H., Chen, C. L., & Wu, C. Y. (2022). Machine learning predicts clinically significant health related quality of life improvement after sensorimotor rehabilitation interventions in chronic stroke. Scientific Reports, 12(1), 1-10. https://doi.org/10.1038/s41598-022-14986-1
  20. Lin, W., Chen, C., Tseng, Y. J., Tsai, Y. T., Chang, C. Y., Wang, H. Y., & Chen, C. K. (2018). Predicting post-stroke activities of daily living through a machine learning-based approach on initiating rehabilitation. International Journal of Medical Informatics, 111, 159-164. https://doi.org/10.1016/j.ijmedinf.2018.01.002
  21. Lin, C., Hsu, K., Johnson, K. R., Fann, Y. C., Tsai, C., Sun, Y., Lien, L., Chang, W., Chen, P., Lin, C., & Hsu, C. Y. (2020). Evaluation of machine learning methods to stroke outcome prediction using a nationwide disease registry. Computer Methods and Programs in Biomedicine, 190, 1-14. https://doi.org/10.1016/j.cmpb.2020.105381
  22. Maso, I., Pinto, E. B., Monteiro, M., Makhoul, M., Mendel, T., Jesus, P. A., & Oliveira-Filho, J. (2019). A simple hospital mobility scale for acute ischemic stroke patients predicts long-term functional outcome. Neurorehabilitation and Neural Repair, 33(8), 614-622. https://doi.org/10.1177/1545968319856894
  23. Mercier, L., Audet, T., Hebert, R., Rochette, A., & Dubois, M. F. (2001). Impact of motor, cognitive, and perceptual disorders on ability to perform activities of daily living after stroke. Stroke, 32(11), 2602-2608. https://doi.org/10.1161/hs1101.098154
  24. Meyer, D., & Wien, F. T. (2001). Support vector machines. R News, 1(3), 23-26.
  25. Platz, T. (2019). Evidence-based guidelines and clinical pathways in stroke rehabilitation-an international perspective. Frontiers in Neurology, 10, 1-7. https://doi.org/10.3389/fneur.2019.00200
  26. Schonlau, M., & Zou, R. Y. (2020). The random forest algorithm for statistical learning. Stata Journal, 20(1), 3-29. https://doi.org/10.1177/1536867X20909688
  27. Scrutinio, D., Ricciardi, C., Donisi, L., Losavio, E., Battista, P., Guida, P., Cesarelli, M., Pagano, G., & D'Addio, G. (2020). Machine learning to predict mortality after rehabilitation among patients with severe stroke. Scientific Reports, 10(1), 1-10. https://doi.org/10.1038/s41598-020-77243-3
  28. Siegert, R. J., & Taylor, W. J. (2004). Theoretical aspects of goal-setting and motivation in rehabilitation. Disability and Rehabilitation, 26(1), 1-8. https://doi.org/10.1080/09638280410001644932
  29. Singh, A., Thakur, N., & Sharma, A. (2016). A review of supervised machine learning algorithms. In 2016 3rd International Conference on Computing for Sustainable Global Development (INDIACom), 2016, 1310-1315.
  30. Sirsat, M. S., Ferme, E., & Camara, J. (2020). Machine learning for brain stroke: A review. Journal of Stroke and Cerebrovascular Diseases, 29(10), 1-17. https://doi.org/10.1016/j.jstrokecerebrovasdis.2020.105162
  31. Son, Y. J., Kim, H. G., Kim, E. H., Choi, S., & Lee, S. K. (2010). Application of support vector machine for prediction of medication adherence in heart failure patients. Healthcare Informatics Research, 16(4), 253-259. https://doi.org/10.4258/hir.2010.16.4.253
  32. Stylianou, N., Akbarov, A., Kontopantelis, E., Buchan, I., & Dunn, K. W. (2015). Mortality risk prediction in burn injury: Comparison of logistic regression with machine learning approaches. Burns, 41(5), 925-934. https://doi.org/10.1016/j.burns.2015.03.016
  33. Suzuki, M., Sugimura, S., Suzuki, T., Sasaki, S., Abe, N., Tokito, T., & Hamaguchi, T. (2020). Machine-learning prediction of self-care activity by grip strengths of both hands in poststroke hemiplegia. Medicine, 99(11). http://doi.org/10.1097/MD.0000000000019512
  34. Tozlu, C., Edwards, D., Boes, A., Labar, D., Tsagaris, K. Z., Silverstein, J., Lane, H. P., Subuncu, M. R., Liu, C., & Kuceyeski, A. (2020). Machine learning methods predict individual upper-limb motor impairment following therapy in chronic stroke. Neurorehabilitation and Neural Repair, 34(5), 428-439. https://doi.org/10.1177/1545968320909796
  35. Wang, W., Kiik, M., Peek, N., Curcin, V., Marshall, I. J., Rudd, A. G., Wang, Y., Douiri, A., Wolfe, C. D., & Bray, B. (2020). A systematic review of machine learning models for predicting outcomes of stroke with structured data. PLoS One, 15(6), 1-16. https://doi.org/10.1371/journal.pone.0234722
  36. Ward, N. S. (2017). Restoring brain function after stroke-bridging the gap between animals and humans. Nature Reviews Neurology, 13(4), 244-255. https://doi.org/10.1038/nrneurol.2017.34
  37. Woo, Y. C., Lee, S. Y., Choi, W., Ahn, C. W., & Baek, O. K. (2019). Trend of utilization of machine learning technology for digital healthcare data analysis. Electronics and Telecommunications Trends, 34(1), 98-110. https://doi.org/10.22648/ETRI.2019.J.340109
  38. Young, J., & Forster, A. (2007). Rehabilitation after stroke. British Medical Journal, 334, 86-90. https://doi.org/10.1136/bmj.39059.456794.68