The proposed study """"""""Algorithms for Pigmented Lesion Screening and Detection"""""""", addresses the need for better physician training by providing specific accurate clinical descriptions of pigmented lesions including malignant melanoma. In Phase I we propose to develop clinical algorithms to precisely describe malignant melanoma. In this phase, aims include: 1. Precise color description and quantization: location of early melanomas within the melanoma color band, definition of variegation. 2. Automatic border detection using global image features and automatic induction. 3. Analysis of asymmetry and irregularity: We will seek optimal measures for these and apply to pigmented lesion images. 4. Importance of features: What are the relative weights of critical features to be combined to generate a simple clinical rule for melanoma diagnosis? Phase II will add one more analytic feature: computer-determined visual texture. What texture measures best help to distinguish melanoma from irregular seborrheic keratoses? Analysis as in Phase I will be performed for benign pigmented lesions and will include development of software for computer-assisted instruction for improved clinical diagnosis of pigmented lesions. Future application of computer vision diagnostic assistance systems might enable low-cost screening of large groups for pigmented lesions. These could include routine hospital admissions, nursing home patients at annual intervals and public skin cancer screenings.
| Anantha, Murali; Moss, Randy H; Stoecker, William V (2004) Detection of pigment network in dermatoscopy images using texture analysis. Comput Med Imaging Graph 28:225-34 |
| Zhang, Z; Stoecker, W V; Moss, R H (2000) Border detection on digitized skin tumor images. IEEE Trans Med Imaging 19:1128-43 |
| Kjoelen, A; Umbaugh, S E; Zuke, M (1998) Compression of skin tumor images. IEEE Eng Med Biol Mag 17:73-80 |
