Computerized diagnosis of skin cancer using neural networks
1998; Elsevier BV; Volume: 16; Linguagem: Inglês
10.1016/s0923-1811(98)83685-2
ISSN1873-569X
AutoresPeter Altmeyer, Guido Pott, M. Happe, R. Husomann, Andrés Muñoz, Laurent Eckert, S. Tölg, Nashan, Thomas Schwarz, Dreier, Peter J. Frosch, Karin Scharffetter‐Kochanek, Heiko Neumann, Ortonne Jp, Lucio Andreassi, R.M. MACKIE, Julia Newton‐Bishop, J.-P. Cesarini, U. Zoras, A. Galvez Garcia, R. Lindskoy, F.R. CHRISTENSEN, Klaus H. Hoffmann,
Tópico(s)AI in cancer detection
ResumoSkin cancer has reached the highest rate of increase among all types of cancer. On the other hand, even the malignant melanoma can be cured. The key is early detection and the key to early detection is regular screening. There exists a clear demand to improve both, the quantity and quality of skin cancer screening. This requirement, which will even increase in the future, cannot be met to the desired extend by current methods alone, as they are too costly and require too many highly-skilled physicians. Computerized techniques, that support diagnostic procedures, promises to be a solution to this problem. Therefore, our primary objective is to proceede developing a prototype named DANAOS (Diagnostic And Neuronal Analysis Of Skin cancer) into a standart medical device. This tool is intended to be used by dermatologically untrained assistants (i.e. nurses, technicians, etc.) to pre-screen patients for melanoma. Transfering major burden of the work to less paid personnel, DANAOS provides not only improved health care but also helps to decrease costs of the health care system.
Referência(s)