Sa1256 ARTIFICIAL INTELLIGENCE USING CONVOLUTIONAL NEURAL NETWORK SHOWS HIGH DIAGNOSTIC PERFORMANCE OF MICROVESSELS ON SUPERFICIAL ESOPHAGEAL SQUAMOUS CELL CARCINOMA SIMILAR TO EXPERTS

2019 ◽  
Vol 89 (6) ◽  
pp. AB190-AB191
Author(s):  
Ryotaro Uema ◽  
Yoshito Hayashi ◽  
Minoru Kato ◽  
Keiichi Kimura ◽  
Takanori Inoue ◽  
...  
2020 ◽  
Vol 08 (03) ◽  
pp. E234-E240
Author(s):  
Yoichiro Ono ◽  
Yasuhiro Takaki ◽  
Kenshi Yao ◽  
Satoshi Ishikawa ◽  
Masaki Miyaoka ◽  
...  

Abstract Background and study aims Magnifying endoscopy with narrow-band imaging (M-NBI) is reported to be useful in diagnosing invasion depth of superficial esophageal squamous cell carcinoma (SCC), but accurate diagnosis of deep submucosal invasion (SM2) has remained difficult. However, we discovered that irregularly branched microvessels observed with M-NBI are detected in SM2 cancers with high prevalence. Thus, this retrospective study aimed to investigate the diagnostic performance of irregularly branched microvessels as visualized by M-NBI for predicting SM2 cancers. Patients and methods Patients with superficial esophageal SCC lesions that were endoscopically or surgically resected at our hospital between September 2005 and December 2014 were included. Endoscopic findings by M-NBI of these lesions were presented to an experienced endoscopist who was unaware of the histopathological diagnosis and who then judged whether irregularly branched microvessels were present. Using the invasion depth according to postoperative histopathological diagnosis as the gold standard, we determined the diagnostic performance of the presence of irregularly branched microvessels as an indicator for SM2 cancers. Results A total of 302 superficial esophageal SCC lesions (228 patients) were included in the analysis. When irregularly branched microvessels were used as an indicator of SM2 cancers, the diagnostic accuracy was 94.0 % (95 % confidence interval [CI]: 91.1–96.1 %), sensitivity was 79.4 % (95 % CI: 66.6–88.4 %), specificity was 95.9 % (95 % CI: 94.3–97.0 %), positive predictive value was 71.1 % (95 % CI: 59.6–79.1 %), and negative predictive value was 97.3 % (95% CI: 95.7–98.5 %). Conclusions Irregularly branched microvessels may be a reliable M-NBI indicator for the diagnosis of cancers with deep submucosal invasion.


2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Sho Shiroma ◽  
Toshiyuki Yoshio ◽  
Yusuke Kato ◽  
Yoshimasa Horie ◽  
Ken Namikawa ◽  
...  

AbstractDiagnosis using artificial intelligence (AI) with deep learning could be useful in endoscopic examinations. We investigated the ability of AI to detect superficial esophageal squamous cell carcinoma (ESCC) from esophagogastroduodenoscopy (EGD) videos. We retrospectively collected 8428 EGD images of esophageal cancer to develop a convolutional neural network through deep learning. We evaluated the detection accuracy of the AI diagnosing system compared with that of 18 endoscopists. We used 144 EGD videos for the two validation sets. First, we used 64 EGD observation videos of ESCCs using both white light imaging (WLI) and narrow-band imaging (NBI). We then evaluated the system using 80 EGD videos from 40 patients (20 with superficial ESCC and 20 with non-ESCC). In the first set, the AI system correctly diagnosed 100% ESCCs. In the second set, it correctly detected 85% (17/20) ESCCs. Of these, 75% (15/20) and 55% (11/22) were detected by WLI and NBI, respectively, and the positive predictive value was 36.7%. The endoscopists correctly detected 45% (25–70%) ESCCs. With AI real-time assistance, the sensitivities of the endoscopists were significantly improved without AI assistance (p < 0.05). AI can detect superficial ESCCs from EGD videos with high sensitivity and the sensitivity of the endoscopist was improved with AI real-time support.


2020 ◽  
Author(s):  
Sho Shiroma ◽  
Toshiyuki Yoshio ◽  
Yusuke Kato ◽  
Yoshimasa Horie ◽  
Ken Namikawa ◽  
...  

Abstract Diagnosis using artificial intelligence (AI) with deep learning could be useful in endoscopic examinations. We investigated the ability of AI to detect superficial esophageal squamous cell carcinoma (ESCC) from esophagogastroduodenoscopy (EGD) videos. We retrospectively collected 8428 EGD images of esophageal cancer to develop a convolutional neural network through deep learning. We evaluated the detection accuracy of the AI diagnosing system compared with that of 18 endoscopists. We used 144 EGD videos for the two validation sets. First, we used 64 EGD observation videos of ESCC using both white light imaging (WLI) and narrow-band imaging (NBI). We then evaluated the system using 80 EGD videos from 40 patients (20 with superficial ESCC and 20 with non-ESCC). In the first set, the AI system correctly diagnosed 100% ESCCs. In the second set, it correctly detected 85% (17/20) ESCCs. Of these, 75% (15/20) and 55% (11/22) were detected by WLI and NBI, and the positive predictive value was 36.7%. The endoscopists correctly detected 45% (25-70%) ESCCs. With AI real-time assistance, the sensitivities of the endoscopists were significantly improved without AI assistance (p<0.05). AI can detect superficial ESCC from EGD videos with high sensitivity and improve endoscopists’ detection of ESCC with real-time support.


Sign in / Sign up

Export Citation Format

Share Document