Page 58 - Read Online

George et al. Mini-invasive Surg 2024;8:4 https://dx.doi.org/10.20517/2574-1225.2023.102 Page 19 of 23

alone does not paint a comprehensive picture of model performance in medical applications. For diagnostic
models, maintaining a low rate of false negatives is crucial to ensure no diagnoses are missed. While false
positives may cause unnecessary worry and additional testing, false negatives can lead to delayed treatment
with potentially severe consequences. Additionally, the current body of research is primarily conducted
retrospectively, introducing the risk of investigator bias. Hence, future prospective multicentre research on
this topic is required.

CONCLUSION AND FUTURE DIRECTIONS
This narrative review provides a comprehensive synthesis on the literature relating to AI in WCE. While
integrating AI into capsule endoscopy shows immense promise in reading time reduction and accuracy
improvement, there is a potential possibility that the system could independently read images in the future.
This path, though, must be navigated carefully, bearing in mind the unique challenges associated with
medical data and the specific requirements of diagnostic models. The potential of ViTs is yet to be fully
exploited in this field. We anticipate an exciting progression in the coming years as more refined and
accurate models are developed.

DECLARATIONS
Authors’ contributions
Study conception and design: Singh R
Data collection: George AA, Tan JL, Kovoor JG, Singh R
Analysis and interpretation of results: George AA, Tan JL, Kovoor JG, George B, Lee A, Stretton B, Gupta
AK, Bacchi S, Singh R
Draft manuscript preparation: George AA, Tan JL, Kovoor JG, Singh R

Availability of data and materials
Not applicable.

Financial support and sponsorship
None.

Conflicts of interest
All authors declared that there are no conflicts of interest.

Ethical approval and consent to participate
Not applicable.

Consent for publication
Not applicable.

Copyright
© The Author(s) 2024.

REFERENCES
1. Muñoz-Navas M. Capsule endoscopy. World J Gastroenterol 2009;15:1584-6. DOI PubMed PMC
2. Beg S, Card T, Sidhu R, Wronska E, Ragunath K; UK capsule endoscopy users’ group. The impact of reader fatigue on the accuracy
of capsule endoscopy interpretation. Dig Liver Dis 2021;53:1028-33. DOI PubMed
3. Giritharan B, Yuan X, Liu J, Buckles B, Oh JH, Tang SJ. Bleeding detection from capsule endoscopy videos. Annu Int Conf IEEE Eng
Med Biol Soc 2008;2008:4780-3. DOI PubMed
4. Pan G, Yan G, Song X, Qiu X. BP neural network classification for bleeding detection in wireless capsule endoscopy. J Med Eng