Application of artificial intelligence using a convolutional neural network for diagnosis of early gastric cancer based on magnifying endoscopy with narrow‐band imaging

Background and Aim Magnifying endoscopy with narrow‐band imaging (ME‐NBI) has made a huge contribution to clinical practice. However, acquiring skill at ME‐NBI diagnosis of early gastric cancer (EGC) requires considerable expertise and experience. Recently, artificial intelligence (AI), using deep l...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Journal of gastroenterology and hepatology Ročník 36; číslo 2; s. 482 - 489
Hlavní autoři: Ueyama, Hiroya, Kato, Yusuke, Akazawa, Yoichi, Yatagai, Noboru, Komori, Hiroyuki, Takeda, Tsutomu, Matsumoto, Kohei, Ueda, Kumiko, Matsumoto, Kenshi, Hojo, Mariko, Yao, Takashi, Nagahara, Akihito, Tada, Tomohiro
Médium: Journal Article
Jazyk:angličtina
Vydáno: Australia Wiley Subscription Services, Inc 01.02.2021
John Wiley and Sons Inc
Témata:
Accuracy
Artificial intelligence
convolutional neural network
Deep learning
Diagnosis
early gastric cancer
Endoscopy
Gastric cancer
Gastritis
Intestine
magnifying endoscopy
Mucosa
narrow‐band imaging
Neural networks
Regular
magnifying endoscopy
convolutional neural network
artificial intelligence
narrow-band imaging
early gastric cancer
ISSN:0815-9319, 1440-1746, 1440-1746
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Background and Aim Magnifying endoscopy with narrow‐band imaging (ME‐NBI) has made a huge contribution to clinical practice. However, acquiring skill at ME‐NBI diagnosis of early gastric cancer (EGC) requires considerable expertise and experience. Recently, artificial intelligence (AI), using deep learning and a convolutional neural network (CNN), has made remarkable progress in various medical fields. Here, we constructed an AI‐assisted CNN computer‐aided diagnosis (CAD) system, based on ME‐NBI images, to diagnose EGC and evaluated the diagnostic accuracy of the AI‐assisted CNN‐CAD system. Methods The AI‐assisted CNN‐CAD system (ResNet50) was trained and validated on a dataset of 5574 ME‐NBI images (3797 EGCs, 1777 non‐cancerous mucosa and lesions). To evaluate the diagnostic accuracy, a separate test dataset of 2300 ME‐NBI images (1430 EGCs, 870 non‐cancerous mucosa and lesions) was assessed using the AI‐assisted CNN‐CAD system. Results The AI‐assisted CNN‐CAD system required 60 s to analyze 2300 test images. The overall accuracy, sensitivity, specificity, positive predictive value, and negative predictive value of the CNN were 98.7%, 98%, 100%, 100%, and 96.8%, respectively. All misdiagnosed images of EGCs were of low‐quality or of superficially depressed and intestinal‐type intramucosal cancers that were difficult to distinguish from gastritis, even by experienced endoscopists. Conclusions The AI‐assisted CNN‐CAD system for ME‐NBI diagnosis of EGC could process many stored ME‐NBI images in a short period of time and had a high diagnostic ability. This system may have great potential for future application to real clinical settings, which could facilitate ME‐NBI diagnosis of EGC in practice.
Bibliografie:Author contribution
Ueyama H, Nagahara A, and Tada T conceived and designed the study and wrote, edited, and reviewed the manuscript. Yao T performed all histopathological diagnosis. Akazawa Y, Yatagai N, Komori H, Takeda T, Matsumoto K, Ueda K, Matsumoto K, and Hojo M gathered ME‐NBI images and patients' clinical information. Kato Y provided valuable advice regarding the technical information and managed the AI‐assisted CNN‐CAD system and analyzed the data in this manuscript. All authors gave final approval for publication. Ueyama H, Nagahara A, and Tada T take full responsibility for the work as a whole, including the study design, access to data, and the decision to submit and publish the manuscript.
No author has a financial relationship relevant to this publication. Aoyama K and Kato Y are technical staff of AI Medical Service. Tada T is a shareholder of AI Medical Service.
Declaration of conflict of interest
ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
Declaration of conflict of interest: No author has a financial relationship relevant to this publication. Aoyama K and Kato Y are technical staff of AI Medical Service. Tada T is a shareholder of AI Medical Service.
Author contribution: Ueyama H, Nagahara A, and Tada T conceived and designed the study and wrote, edited, and reviewed the manuscript. Yao T performed all histopathological diagnosis. Akazawa Y, Yatagai N, Komori H, Takeda T, Matsumoto K, Ueda K, Matsumoto K, and Hojo M gathered ME‐NBI images and patients' clinical information. Kato Y provided valuable advice regarding the technical information and managed the AI‐assisted CNN‐CAD system and analyzed the data in this manuscript. All authors gave final approval for publication. Ueyama H, Nagahara A, and Tada T take full responsibility for the work as a whole, including the study design, access to data, and the decision to submit and publish the manuscript.
ISSN:0815-9319
1440-1746
1440-1746
DOI:10.1111/jgh.15190