AUTHOR=Yuan Peng , Bai Ruxue , Yan Yan , Li Shijie , Wang Jing , Cao Changqi , Wu Qi 

TITLE=Subjective and objective quality assessment of gastrointestinal endoscopy images: From manual operation to artificial intelligence

JOURNAL=Frontiers in Neuroscience

VOLUME=Volume 16 - 2022

YEAR=2023

URL=https://www.frontiersin.org/journals/neuroscience/articles/10.3389/fnins.2022.1118087

DOI=10.3389/fnins.2022.1118087

ISSN=1662-453X

ABSTRACT=Gastrointestinal endoscopy has been identified as an important tool for cancer diagnosis and therapy, particularly for patients with early gastric cancer (EGC). It is well known that the quality of gastroscope images is a prerequisite for a high detection rate of gastrointestinal lesions. Due to the manual operation for gastroscope detection, in practice, it very possibly introduces motion blur and produces low-quality gastroscope images during the imaging process. Hence, the quality assessment of gastroscope images is the key process of gastrointestinal endoscopy detection. In this paper, we first present a novel gastroscope image motion blur (GIMB) database, which includes 1,050 images generated by imposing 15 distortion levels of motion blur on 70 lossless images and the associated subjective scores produced with manual operation of 15 viewers. Then, we come up with a new artificial intelligence-based gastroscope image quality evaluator (GIQE), which leverages the newly proposed semi-full combination subspace to learn multiple kinds of human visual system (HVS) inspired features for providing objective quality scores. Results of experiments conducted on the GIMB database confirm that the proposed GIQE has attained more effective performance as compared with the state-of-the-art peers.