ACQUISITION AND ANALYSIS OF THE ENDOSCOPIC IMAGES

SUMA H.N.¹, ABHISHEK A.M.², SAHADEV R.^3
1Department of Medical Electronics, BMS College of Engineering Bangalore- 560 019, Karnataka, India.
²Department of Medical Electronics, BMS College of Engineering Bangalore- 560 019, Karnataka, India.
³Department of Surgical Gastroenterology, KIMS Hospital, Bangalore- 560 070, Karnataka, India.

Received : 31-07-2013     Accepted : 15-01-2014     Published : 30-04-2014
Volume : 2     Issue : 1       Pages : 25 - 31
Bioinfo Med Imag 2.1 (2014):25-31

Conflict of Interest : None declared
Acknowledgements/Funding : We acknowledge Dr. N.C Srinivasa Prabhu, Professor, Department of General Medicine, Kempegowda Institiute of Medical Science (KIMS), Bangalore for his valuable support. We also Thank B M S College of Engineering providing necessary infrastructure for exec

High incidence of cancer cases in Asia has triggered a need to develop an effective automated screening method for early detection. Currently a computer-based analysis for endoscopic diagnosis of suspected malignancies is not employed. The only method that is used is to take a tissue invasively from the infected parts (biopsy) for the confirmation of pathology which is expensive and time consuming. Therefore the objective of this paper is to develop a new technique for automated analysis of the endoscopic images based on color and texture information for stomach abnormalities. The proposed method involves endoscopic image processing for characterization using texture-associated features. It is used to classify endoscopic images to normal and potentially abnormal categories. This work proposes an automated way to differentiate between the images to compensate for invasive diagnostic measures. In this method the affected area is segmented and the features are found out from the segmented image that differentiate the abnormal image from the normal and then categorized into Tumour, Ulcer, Cancer or Polyp as individual classes. A computer-assisted method for automated endoscopic image analysis based on texture feature has been proposed. Initial results suggest the feasibility of the proposed method for stomach abnormalities. The proposed new computer-assisted endoscopic analysis method is devised to be beneficial for automatic classification of the images.