Suryarini Widodo

Sample K-Means Clustering Method for Determining the Stage of Breast Cancer Malignancy Based on Cancer Size on Mammogram Image Basis

Breast cancer is a disease that arises due to the growth of breast tissue cells that are not normal. The detection of breast cancer malignancy level / stage relies heavily on the results of the analysis of the doctor. To assist the analysis, this research aims to develop a software that can determine the stage of breast cancer based on the size of the cancerous tissue. Steps of the research consist of mammogram image acquisition, determining the ROI (Region of Interest), using Region growing segmentation method, measuring the area of suspected cancer, and determine the stage classification of the area on the mammogram image by using Sample K-Means Clustering method. Based on 33 malignant (abnormal) mammogram sample images taken from the mini mammography database of MIAS, the proposed method can detect stage of breast cancer is in malignant group.

Handwritten Pattern Recognition Using Kohonen Neural Network Based on Pixel Character

Handwriting is the human way in communicating each other using written media. By the advancement in technology and development of science, there are a lot of changes of technology in terms of communication with computer through handwriting. Therefore, it is needed computer able to receive input in the form of handwriting data and able to recognize the handwriting input. Therefore, this research focuses on handwritten character recognition using Kohonen neural network. The purpose of this research is to find handwriting recognition algorithm which can receive handwriting input and recognize handwritten character directly inputted in computer using Kohonen neural network. This method studies the distribution of a set of patterns without any class information. The basic idea of this technique is understood from how human brain stores images/patterns that have been recognized through eyes, and then able to reveal the images/patterns back. This research has been successful in developing an application to recognize handwritten characters using Kohonen neural network method, and it has been tested. The application is personal computer based and using a canvas as input media. The recognition process consist of 3 stages layer: Input layer, Training Layer and Hidden Layer. The Kohonen neural network method on handwritten character recognition application has good similarity level of character patterns in character mapping process.