The goal of this project is to develop technology for machine vision
based psoriasis area assessment.
The principle is to use segmentation to differentiate the psoriasis plaques
from healthy skin and the
background in medical images. The final classification of the segments
affected by psoriasis is carried
out manually.
The CIA system is based on segmentation with the previously developed
hierarchical connected
components (HCC) analyzis. In order to calculate the percentage area of
psoriasis involvement, the skin
surface area of the whole body is determined by extracting the image with
the OhtaI2 color feature
and using gray level thresholding.
Photography system for psoriasis skin image application
![]()
Image after:
Ohta I2 extraction Gray
level thresholding HCC
segmentation
User
interface for final area assessment
Publications:
Kontinen J. (1994) The assessment of psoriasis area with
digital image processing. Diploma thesis. University of Oulu,
Department of Electrical Engineering, Oulu.
Röning J. and Kontinen J. (1996) Measurement of the
area of involvement in skin disease. Proc. SPIE Vol. 2904
Intelligent Robots and Computer Vision XV: Algorithms, Techniques, and
Materials, November 19-21, Boston,
Mass., USA, pp. 382-388.
Savolainen L., Kontinen J., Röning J. & Oikarinen A. (1997) Application of machine vision to assess involved surface in patients with psoriasis. British Journal of Dermatology 137, pp. 395-400.
Kontinen J, Savolainen L., Oikarinen A. & Röning
J. (1997) Computer image analyzing system for psoriasis area assessment.
Proc. IASTED International Conference Signal and Image Processing (SIP-97),
December 4-6, New Orleans, Louisiana, 145-149.