Interoperative Biopsy Site Localization in Endoscope for Gastro Intestinal Tract using Deep Learning Technology

Abstract

Endoscopic biopsy is a common procedure used to obtain tissue samples from the gastrointestinal tract for diagnostic purposes. However, accurately localizing the biopsy site can be challenging, which can lead to errors in diagnosis and treatment. In this study, we propose a method for inter operative biopsy site localization in the gastrointestinal tract using Deep Learning Technology (DLT). The proposed method involves training a deep learning model on a dataset of endoscopic images and their corresponding biopsy locations. The model is then used to predict the location of the biopsy site in real-time during the biopsy procedure. The model is designed to be integrated with the endoscope's imaging system, allowing for real-time localization of the biopsy site. Overall, the proposed method has the potential to improve the accuracy and efficiency of endoscopic biopsy procedures, thereby improving patient outcomes. The deep learning model will be refined further and tested on a bigger dataset of endoscopic pictures in the future.