Automatic image segmentation of nuclear stained breast tissue sections using color active contour model and an improved watershed method

摘要: Abstract Automatic image segmentation of immunohistologically stained breast tissue sections helps pathologists to discover the cancer disease earlier. The detection real number nuclei in is a very tedious and time consuming task. Segmentation nuclei, especially touching presents many difficulties separate them by traditional algorithms. This paper new automatic scheme perform both classification order get total each class. Firstly, modified geometric active contour model used for multiple positive negative nuclear staining microscopic image. Secondly, method based on watershed algorithm concave vertex graph proposed accurate quantification different stains. Finally, benign are identified their morphological features they removed automatically from segmented assessment. schemes tested two datasets cell images containing level malignancy. experimental results show superiority methods when compared with other existing methods. On complete database, accuracy term over than 97%, reaching an improvement 3–4% earlier