A computer assisted image analysis system for diagnosing movement disorders
Video image analysis is able to provide quantitative data on postural and movement abnormalities and thus has an important application in neurological diagnosis and management. The conventional techniques require patients to be videoed while wearing markers in a highly structured laboratory environment. This restricts the utility of video in routine clinical practice. We have begun development of intelligent software able to extract complete human profiles from video frames, to fit skeletal frameworks to the profiles and derive joint angles and local curvatures. By this means a given posture is reduced to a set of basic parameters that can provide input to a neural network classifier.
To test the system's performance, we videoed patients with dopa-responsive Parkinson's and age matched normals during several gait cycles, to yield 61 patient and 49 normal postures. These postures were reduced to their basic parameters and fed to the neural network classifier in various combinations. The optimal parameter sets (consisting of both swing distances and joint angles) yielded successful classification of normals and patients with an accuracy above 90%. This result demonstrated the feasibility of the approach. The technique has the potential to guide clinicians on the relative sensitivity of specific postural /gait features in diagnosis.
Unable to display preview. Download preview PDF.
- 1.Embry, D. G.,Yates, L. and Mott, D. H.: Effects of Neur-Developmental Treatment and Orthoses on Knee Flexion During Gait: A Single Subject Design. Physical Therapy 70(10): 1990.Google Scholar
- 6.Martinez-Martin, P., Bermejo-Parejo, F.: Rating Scales in Parkinson's Disease.In: Jankovic Joseph and Tolosa Eduardo, ed. Parkinson's Disease and Movement Disorders. Baltimore-Munich: Urban and Schwarzenberg, pp.235–242, 1988.Google Scholar
- 7.Nuzzo, R.M., Jllly, J. and Langrana, NA.: contralateral Compensation with Knee Impairment.Clenicl Orthopaedics and Related Research No.223:225–236, 1987.Google Scholar
- 8.Gonzlez, R. C., Woods, R. E.: Digital Image Processing No.148–156; No.532–533. Addison-Wesley, New York, U.S.A. 1992Google Scholar
- 9.Sonka, M., Hlavac, V., Boyle, R.: Image Processing, analysis and Machine Vision No. 129–131. University Press, Cambridge, U.K. 1993.Google Scholar
- 10.Jang, B., Chin, R. T.: Analysis of Thinning Algorithms Using Mathematical Morphology, IEEE Transactions on Pattern Analysis and Machine Intelligence,Vo1.12, No. 6, June 1990.Google Scholar
- 11.Masters, R. T.: Practical Neural Network Recipes in C++, Academic Press, Boston, U.S.A. 1993.Google Scholar