Cancer detection from pathological images using Higher-order Local Autocorrelation feature

Pathological diagnosis, plays a very crucial role in cancer examination. However, because of the shortage of pathologists, increasing working burden has been imposed on pathologists. Aiming to reduce pathologist´s burden and improve the reliability of cancer diagnosis, an innovative cancer detection method for pathological images and its extended approach for advanced ability is presented in this paper. The experiments demonstrate the effectiveness of the proposed method in cancer detection and cancer location estimation.