Indonesia Conference Directory

Gram-negative bacteria are one of the causes of nosocomial in Indonesia. This bacterium is a cause of resistance so that the disease is difficult to cure with antibiotic treatment. In the process of antibiotic therapy, these nosocomial bacteria must be removed first before handling the main bacteria that cause disease. This bacterial observation is carried out using image processing to replace visual observation. The process consists of four stages, namely pre-processing, segmentation, feature extraction, and identification. At the segmentation stage, the bacterial image object is selected that best suits the expert representation, in this case, the medical analyst. Feature extraction is done to get pixel information to be processed. At the classification stage, the use of Extreme learning machine is chosen because the training process time is shorter than other algorithms based on artificial neural networks. At the stage of modeling, 2 different bacteria were used, namely nosocomial bacterium and Gram-negative bacterium. In this research trial, each consists of 420 images of training data, validation and testing so that the total amount of data used is 2520 images with a pixel size of 256x256. Accuracy results obtained at 97% of the training process