Melanoma is the most lethal skin-cancer. The occurrence rate of malignant melanoma among the white population is rising faster than almost any other cancer in North America. Nearly all patients can be saved and cured by early detection and prompt surgical treatment. The important diagnostic and prognostic variables of melanoma are the vertical thickness, three- dimensional (3D) size and shape, and color of the lesion. The other characteristic features of early melanoma are irregularities in the boundary of the lesion, and the appearance of non-uniform pigmentation with a variety of color. The initial changes are subtle and cannot be observed with the naked eye. A novel optical instrument called the """"""""Nevoscope"""""""" has been developed to view the multiple images of the lesion in situ from several angles by transilluminating the skin-lesion. The multiple views have been digitized and 3D images of the lesion have been reconstructed to measure the thickness and 3D size. The objective of this research is to improve the design and performance of the Nevoscope, to perform analysis on thickness, 3D size, color, and margin, boundary and surface characteristics to detect early malignant lesions. The approach proposed here is based on the knowledge-based analysis and interpretation of the images of the transilluminated skin-lesion. The complete system will be tested in a clinical environment. The analysis and interpretation will be compared with the histology and the diagnosis made by a dermatologist for the lesions whose excisions have been planned.

Project Start
1988-09-01
Project End
1993-08-31
Budget Start
1989-09-01
Budget End
1990-08-31
Support Year
2
Fiscal Year
1989
Total Cost
Indirect Cost
Name
University of Cincinnati
Department
Type
Biomed Engr/Col Engr/Engr Sta
DUNS #
City
Cincinnati
State
OH
Country
United States
Zip Code
45221
Kini, P; Dhawan, A P (1994) Noninvasive imaging and analysis of skin lesions for early detection of cutaneous malignant melanoma. Biomed Instrum Technol 28:209-19
Dhawan, A P; Arata, L (1993) Segmentation of medical images through competitive learning. Comput Methods Programs Biomed 40:203-15
Dhawan, A P; Sicsu, A (1992) Segmentation of images of skin lesions using color and texture information of surface pigmentation. Comput Med Imaging Graph 16:163-77
Kini, P; Dhawan, A P (1992) Three-dimensional imaging and reconstruction of skin lesions. Comput Med Imaging Graph 16:153-61