Abstract
Objective: Reading small bowel capsule endoscopy (SBCE) is a tedious task. We aim to evaluate the performance of a Convolutional Neural Network (CNN) algorithm (SmartScan)-assisted reading in the setting of real-life clinical care. Design: A total of 5,825 SBCE examinations (comprising 295,314,067 images) performed in 51 Chinese medical centres were collected. 2,927 SBCE examinations from 29 medical centres were used to train SmartScan to detect 17 CEST findings. SmartScan was later validated in 2,898 SBCE examinations collected from 22 medical centres. All 2,898 SBCE examinations were read with both Conventional Reading (CR) and SmartScan-Assisted Reading (SSAR) by eight specialist gastroenterologists. If finding(s)/results by both types of reading in a patient were concordant, the result was taken as the Combined Agreed Comparator (CAC) for that case. In discordant cases, the consensus result by three expert readers was taken as the CAC for the case. Results: SSAR achieved an overall higher detection rate (79.30%) for CEST findings compared to CR (70.67%); from a total of 6,084 SB findings, SSAR detected 95.89%, significantly higher compared to CR (76.10%). Furthermore, SSAR achieved overall higher sensitivity (98.80%), for the 17 CEST subtypes of findings, 10.75% higher than CR (88.05%). With SSAR, the number of images requiring review per SB CE video was reduced to 779.17 ±337.18, with reading time was shortened to 5.37 ±1.51min. C Conclusion: A commercially available CNN algorithm (SmartScan) can increase the detection rate of SB findings while reducing SB CE reading times. Trial Registration: This clinical trial study was registered in the Chinese Clinical Trial Registry (reg no. CHiCTR2100042455). Funding: This work was supported by the National Key Research and Development Program (NO. 2016YFC0107000) and Special project of National Health Committee (NO. 201502013). Declaration of Interest: The authors declare no competing interests. Ethical Approval: The whole study was performed in accordance with guidelines approved by Medical Ethics Committee of Second Affiliated Hospital of the Third Medical University (NO.2020-yan-134-01).
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
