By Keely Holland

Background

Gastric cancer, also known as stomach cancer, is caused by a growth of mutated cells in the stomach (Mayo Clinic, 2018). Stomach cancer is the 4th most common form of cancer and can be treated if caught in its early stages, but it can be deadly if left unnoticed (Li et al., 2025). An endoscopy is what is performed to diagnose stomach cancer.

Technology

New technology has been developed called Endoangel—an AI assistive software that helps diagnose gastric cancer. Over time, humans develop lots of accurate endoscopy results. AI then receives these results and begins to analyze and learn from them, determining what constitutes cancer and what does not. Unfortunately, for humans and skilled medical professionals, gastric cancer is not always an easy diagnosis. When performing endoscopies, there are always spots that people can’t even see. Fortunately, AI can view these blind spots more clearly than humans can. 

Additionally, certain forms of gastric cancer may not look like typical cancer and will usually look like a small lump or imperfection. Most of the time, humans will pass this “small lump” by, due to barely even being able to see it in the first place or determine that it is cancer. However, AI technologies like Endoangel can tell if these imperfections are cancerous. Ultimately, this technology yields significantly more accurate results than humans alone.

        An Endoangel Device (ENDOANGEL Endoscope with Artificial Intelligence, 2022)

Applications

AI searches through images and then picks out major, noticeable features that help tell the difference between normal images and ones with cancerous tumors (Recio-Boiles et al., 2020). White light imagery is used to make things more easily recognizable for the AI. The AI can also zoom into images through what is called blue-laser imagery, which is  a term used to describe when AI magnifies a photo to search for details better. Through this process, the AI uses machine learning to learn from the doctor’s data, similar to how other AIs learn from their users. On the other hand, a process called deep learning could be used instead, where the AI creates links between data so it can “make decisions” and recognize data on its own without human assistance. Endoscopic AI utilizes both of these processes (Tham et al., 2025). 

AI is most commonly used by scanning through cameras and endoscopic results on computers in hospitals, medical centers, and cancer institutes. It has proved to be incredibly beneficial in the healthcare field, and with this evidence, it could be inferred that AI is not essentially inferior to human doctors.

Results/Conclusion

In the end, endoscopic AI has increased results compared to endoscopists alone. Accuracy has increased to 79.61% from 70.61%; sensitivity has increased to 82.11% from 75.95%; and specificity has increased to 77.73% from 66.44% (Dong et al., 2023). AI has developed far from just being able to recognize a part of the body to being able to detect minute details that humans would easily miss. Its accuracy has gone from what looks like a 50/50 guess to being more accurate and precise than trained endoscopists. However, despite all these promising results, AI will never be perfect. Nevertheless, the best results would come from endoscopists working together with AI to make the best gastric cancer diagnosis possible to save as xmany lives as possible.

Bibliography:

‌Dong, Z., Wang, J., Li, Y., Deng, Y., Zhou, W., Zeng, X., Gong, D., Liu, J., Pan, J., Shang, R., Xu, Y., Xu, M., Zhang, L., Zhang, M., Tao, X., Zhu, Y., Du, H., Lu, Z., Yao, L., & Wu, L. (2023). Explainable artificial intelligence incorporated with domain knowledge diagnosing early gastric neoplasms under white light endoscopy. Npj Digital Medicine, 6(1), 1–9. https://doi.org/10.1038/s41746-023-00813-y

‌Li, R., Li, J., Wang, Y., Liu, X., Xu, W., Sun, R., Xue, B., Zhang, X., Ai, Y., Du, Y., & Jiang, J. (2025). The artificial intelligence revolution in gastric cancer management: clinical applications. Cancer Cell International, 25(1). https://doi.org/10.1186/s12935-025-03756-4

Recio-Boiles, A., Waheed, A., & Babiker, H. M. (2020). Gastric cancer. PubMed; StatPearls Publishing. https://www.ncbi.nlm.nih.gov/books/NBK459142/

Stanford Department of Medicine. (2022, December 20). The Use of Artificial Intelligence in Gastric Cancer Screening – 2022 Gastric Cancer Summit. YouTube. https://www.youtube.com/watch?v=FxFCwsQpVRI

‌Stomach cancer – Symptoms and causes. (2025). Mayo Clinic; https://www.mayoclinic.org/diseases-conditions/stomach-cancer/symptoms-causes/syc-20352438 

Tham, C., Rea, D., & Tham, T. (2025). Artificial Intelligence in Endoscopy: A Narrative Review. The Ulster Medical Journal, 94(1), 16. https://pmc.ncbi.nlm.nih.gov/articles/PMC12042857/#b30

‌(2025). Radfarm.com.ua. https://radfarm.com.ua/image/cache/catalog/gastroscop/5-1200×800.JPG