ATD Angle Palmprint Feature Extraction And Its Correlation With Breast Cancer

A large number of clinical investigation shows that the breast cancer is the highest incidence for women malignant tumors and one of the most important causes of death in women. The key to improve the breast cancer’s survival probability is early detection and treatment. Currently, the widely used susceptibility estimation of breast cancer based on gene has the disadvantages of expensive, invasive and low detection rate. Research on breast cancer pathology and genetics demonstrates that the etiology of breast cancer and human palm print are all influenced by genetic and environment. At the embryonic period, breast and palm prints are origned from embryo’s ectoderm. Moreover, a large number of dermatoglyphics research reveals the palm print’s ATD angle has great correlation with the risk of breast cancer.The current methods of breast cancer related features extraction is manual by using the eyes detect qualitative features and coarsly describe the features. Thus method is time consuming and low efficiency, poor reliability and consistency, and the accuracy varies depending on the extraction method. In this paper, based on the theory of dermatoglyphy and breast cancer, a method is proposed to accurately and automatically extract the ATD angle of palm prints more, which applies non physical contact method for palm sampling through a digital camera directly, and then correctly locates the A, T, D point in a palm, finally calculates the ATD angle. The width of palm skin grain is only0.4mm or so, the texture distribution of palm is similar fingerprint singular point. But the pressure sensor of fingerprints sampling can not be applied to extract uneven palm prints. Therefore, we propose to paint carbon ink on the related area to over up the effect of skin and distinct the skin ridge and ditch more clearly. By using this method, we can get high quality palm image, which is very similar to the image quality received by fingerprint pressure sensor, through high-resolution digital camera. The sampling method is highly helpful for the ATD angle location and calculation.The proposed ATD angle extraction,first perform preprocessing to palm images, including gray scale expansion, erosion, open and close operation..Subsequently, according to the characteristics of the palm print image, butterworth filter is used to enhance the image. We proposed a improved Canny edge detection algorithm to extract the edge of the palm print texture. Then a,improved algorithm of the direction field is proposed for precisely locating the feature point, that is A, T and D. The error analysis of A, T, D location. is carried out on the final result, which shows the error between the feature points located by the proposed algorithm and the manual extraction is within0.3mm, a half width of the stripes. The error between the ATD angle computed by our method and the one of manual extraction is less than0.2degrees, which the accuracy can meet the requirements for dermatoglyph analysis needs. Finally, we sample data of breast cancer patients and normal people and analyze the relationship between the ATD angle of palm print and breast cancer susceptibility by using the statistical method. The result show ATD angle is related to breast cancer and can be used as a palm feature for breast cancer susceptibility..