CHANGES IN IMPAIRMENT AND DISABILITY FROM THE 3RD TO THE 6TH MONTH AFTER STROKE AND ITS RELATIONSHIP EVALUATED BY AN ARTIFICIAL NEURAL-NETWORK

Citation
S. Sonoda et al., CHANGES IN IMPAIRMENT AND DISABILITY FROM THE 3RD TO THE 6TH MONTH AFTER STROKE AND ITS RELATIONSHIP EVALUATED BY AN ARTIFICIAL NEURAL-NETWORK, American journal of physical medicine & rehabilitation, 76(5), 1997, pp. 395-400
Citations number
14
Categorie Soggetti
Rehabilitation
ISSN journal
08949115
Volume
76
Issue
5
Year of publication
1997
Pages
395 - 400
Database
ISI
SICI code
0894-9115(1997)76:5<395:CIIADF>2.0.ZU;2-K
Abstract
Functional recovery in a rather late stage after stroke was examined i n 70 stroke patients using the Stroke impairment Assessment Set (SIAS) and the Functional Independence Measure (FIM(SM)). The SIAS and the F IM(SM) Were administered at three and six months after the onset of st roke. Motor items and the abdominal manual muscle testing item improve d in more than 30 percent of patients. The motor subscore of the FIM(S M) changed from 60.8 to 73.4, and the cognitive subscore changed from 28.4 to 30.4. The relationship between impairment and disability was e valuated using the neural network method with the software, Skiltran. The change of the FIM(SM) motor subscore from three months to six mont hs was used as an output variable, and the change in the SIAS items an d the FIM(SM) motor subscore were included as input variables. As a re sult of the connection weight obtained from this network, the change i n the fifth motor item (one of the tone items, abdominal manual muscle testing) and the unaffected side grip as well as the FIM(SM) at three months had a strong connection to the change of the FIM(SM). It is co mpatible with ordinary experience that function of both the affected a nd unaffected side influences the level of disability. Contribution of the impairment to the disability indicates the importance of taking i nto consideration the impairment for predicting prognosis and selectin g adequate treatment when we carry out stroke rehabilitation. In concl usion, we described the relationship between the SIAS and the FIM(SM) using the neural network in stroke patients and proved the importance of the impairment to predict the outcome of disability.