Oesophagus Achalasia Diagnosis from Esophagoscopy Based on a Serial Multi-scale Network

ABSTRACT

Oesophageal achalasia is a primary oesophageal motility disorder disease. To diagnose oesophagus achalasia, physicians recommend endoscopic evaluation of the oesophagus. However, a low sensitivity still accompanies esophagoscopy on oesophagus achalasia diagnosis. Thus, a quantitative diagnosis system is needed to support physicians diagnose achalasia from the esophagoscopy video. This paper proposes a Serial Multi-scale Network for classifying achalasia images from the esophagoscopy video. The proposed method contains two main components, a Dense-pooling Net, and a Serial Multi-scale Dilated encoder. We construct the Dense-pooling Net using a convolution neural network with dense mixed-pooling connections to extract features. We design the Serial Multi-scale Dilated encoder based on a residual-style dilated encoder. We combine the dilated encoder and spatial attention modules to focus on features we need. We trained and evaluated our method with a dataset that was extracted from several esophagoscopy videos of achalasia patients. The evaluation results reveal a state-of-the-art accuracy of achalasia diagnosis. Furthermore, we developed a real-time computer-aided achalasia diagnosis system with the trained network. In the real-time test, the achalasia diagnosis system can stably output the diagnosis results in only $0.138$ seconds. The extended experiments demonstrate that the constructed diagnosis system can diagnose achalasia from esophagoscopy videos.

KEYWORDS:

• Oesophageal achalasia
• oesophagoscopy
• computer-aided diagnosis
• deep learning

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

Parts of this work was supported by the JST CREST under Grant JPMJCR20D5; the MEXT/JSPS KAKENHI under Grant 26108006, 17H00867, 17K20099, 21K19898; the JSPS Bilateral International Collaboration Grants; the AMED under Grant 19hs0110006h0003; the Hori Sciences & Arts Foundation; and the J-CASE Research Grant.

Notes on contributors

Masahiro Oda

Masahiro Oda (Ph.D. Nagoya University '09) is an associate professor at the Nagoya University, Japan. His research is in medical image processing and mainly concerns computer-aided diagnosis and computer-assisted surgery in many application areas. He has (co-)authored over 200 peer-reviewed full papers in international conferences and journals and is the recipient of the Joint MICCAI 2022 Workshop on AE-CAI, CARE, and OR 2.0 Outstanding Paper Award and the RSNA Certificate of Merit (2009, 2014, 2019) Awards.

Hironari Shiwaku

Dr. Shiwaku is a surgeon and endoscopist specializing in endoscopic diagnosis and treatment of diseases of the gastrointestinal tract using EMR, ESD, POEM, POET and EFTR. Dr. Shiwaku graduated from Fukuoka University School of Medicine in 2003, and after a residency in Fukuoka University Hospital and its affiliated hospitals, he joined the Digestive Disease Center, Showa University Northern Yokohama Hospital, to study endoscopic diagnosis and treatment from 2008–2009. In 2010, he returned to Fukuoka University and completed his PhD degree. Dr. Shiwaku subsequently accepted a position as Associate professor in the Department of Gastroenterological Surgery of Fukuoka University in 2021. He is also a part-time lecturer in the Digestive Diseases Center, Showa University, Koto Toyosu Hospital. Dr. Shiwaku is board-certified and a Senior Fellow, Councilor, and member of the POEM Guidelines Committee of the Japan Gastroenterological Endoscopy Society (JGES). He is also a Fellow of the American Society for Gastrointestinal Endoscopy (ASGE) and JGES.

Masashi Misawa

Dr. Misawa is an associate professor at Showa University Northern Yokohama Hospital. His PhD thesis, awarded in 2015, examined the efficacy of magnifying narrow-band imaging for colorectal cancer. He maintains an active research interest in: artificial intelligence, computer-aided diagnosis; endoscopic imaging and techniques; endocytoscopy. Dr. Misawa is a member of Japan Gastroenterological Endoscopy Society, The American Society for Gastrointestinal Endoscopy, The Japanese Society of Gastroenterology, etc.

Kensaku Mori

Kensaku Mori has received the B.S degree in electronics engineering, the M.S degree in Information Engineering, and the Ph.D in Information Engineering from Nagoya University, Japan, under the supervision of Prof. Jun-ichiro Toriwaki, in 1992, 1994 and 1996 respectively. He was a research fellow of the Japanese Society for the Promotion of Science (JSPS) from 1994 to 1997, Research Associate at Department of Computational Science and Engineering, Nagoya University from 1997 to 2000, an assistant professor in 2000. He was Associate Professor at the Research Center for Advance Waste and Emission Management of Nagoya University from 2001 to 2003. He was also Visiting Associate Professor at Department of Neurosurgery, Stanford University from 2001 to 2002. He was Associate Professor of the Graduate School of Information Science, Nagoya University from 2003 to 2009. He has been Professor of Information and Communications Headquarters, Nagoya University since 2009. In 2016, he started his services as the Directory of Information Technology Center, Nagoya University. In 2017, he was appointed as the Vice Chair of Information and Information and Communication, Nagoya University. He has been a Professor of Graduate School of Informatics, Nagoya University since 2017. His current research interests include three-dimensional image processing, computer graphics, virtual reality, and their applications to medical image. Computer aided diagnosis and surgery are his major active research fields. Recently AI-based colonoscope diagnosis system developed in his laboratory has obtained an official certificate as a medical device. He has received many awards from several institutes includes the Awards for Science and Technology, the Ministry of Education, Culture, Sports, the best paper awards from the Japanese Society of Medical Imaging Technology. He served as the general chair of the 16th international conference of Medical Image Computing and Computer Assisted Intervention (MICCAI 2013). He was elected as the President of the International Society of Computer Aided Surgery (ISCAS) in 2018. He was the Chair of SPIE Medical Imaging CAD conference in 2018 and 2019. He is a member of the Institute of Electronics, Information and Communication Engineering of Japan, Japanese Society of Biomedical Engineering, and Japanese Society of Medical Imaging Technology, MICCAI, IEEE, and SPIE. He is an Affiliate member of the Science Council of Japan.