Prospective clinical validation of a novel artificial intelligence system for real-time detection of solid pancreatic masses during endoscopic ultrasonography

Abstract

Background

Endoscopic ultrasonography (EUS) is the most sensitive modality for accurately establishing a tissue diagnosis in patients with solid pancreatic masses. However, small lesions can be challenging to detect, particularly for less experienced endosonographers. Therefore, outcomes of EUS are operator dependent. We validated the performance of novel artificial intelligence (AI)-enhanced EUS for detection of solid pancreatic lesions.

Methods

In this single-center, prospective, nonrandomized, comparative study, high-risk patients aged ≥18 years referred for pancreatic cancer screening or with suspected (solid and cystic) pancreatic lesions owing to symptoms, radiological, or laboratory findings were evaluated in real time using AI-EUS software. The model included 32 713 EUS frames (training/testing phases) of normal, solid, and >10-mm cystic pancreatic lesions from 202 patients. Clinical validation was conducted prospectively when EUS findings were evaluated concurrently in real time by two independent expert examiners, one using conventional EUS and another with AI-EUS, both blinded to the alternative assessments. The primary outcome was detection of solid pancreatic masses.

Results

308 patients were evaluated (January–July 2024). AI-EUS performance was not significantly different to that of conventional EUS performed by experts (97.1% vs. 100%; risk difference 2.9%, 95%CI –1.2 to 6.8; P = 0.25). Final pathology of 105 pancreatic solid masses revealed neoplasia in 93 (88.6%) and benign lesions in 12 (11.4%).

Conclusion

The performance of AI-EUS was not significantly different to that of experienced endosonographers for detection and segmentation of solid pancreatic masses. By standardizing performance, AI-EUS may have the potential to optimize clinical outcomes in pancreatic cancer.

Publication History

Received: 06 March 2025

Accepted after revision: 12 September 2025

Accepted Manuscript online:
15 September 2025

Article published online:
13 October 2025

• References

• 1 Siegel RL, Miller KD, Wagle NS. et al. Cancer statistics 2023. CA Cancer J Clin 2023; 73: 17-48
• 2 National Cancer Institute. Cancer stat facts: pancreatic cancer. Accessed June 03, 2024 at: https://seer.cancer.gov/statfacts/html/pancreas.html
• 3 Blackford AL, Canto MI, Klein AP. et al. Recent trends in the incidence and survival of stage 1a pancreatic cancer: a surveillance, epidemiology and end result analysis. J Natl Cancer Inst 2020; 112: 1162-69
• 4 Mazer BL, Lee JW, Roberts NJ. et al. Screening for pancreatic cancer has the potential to save lives, but is it practical?. Expert Rev Gastroenterol Hepatol 2023; 17: 555-574
• 5 Karlson BM, Ekbom A, Lindgren PG. et al. Abdominal US for diagnosis of pancreatic tumor: prospective cohort analysis. Radiology 1999; 213: 107-111
• 6 Bronstein YL, Loyer EM, Kaur H. et al. Detection of small pancreatic tumors with multiphasic helical CT. AJR Am J Roentgenol 2004; 182: 619-623
• 7 Harinck F, Konings ICAW, Klujit I. et al. A multicentre comparative prospective blinded analyses of EUS and MRI for screening of pancreatic cancer in high-risk individuals. Gut 2016; 65: 1505-1513
• 8 Canto MI, Hruban RH, Fishman EK. et al. Frequent detection of pancreatic lesions in asymptomatic high-risk individuals. Gastroenterology 2012; 142: 796-780
• 9 Kurita Y, Hara K, Kobayashi N. et al. Detection rate of endoscopic ultrasound and computed tomography in diagnosing pancreatic neuroendocrine neoplasms including small lesions: a multicenter study. J Hepatobiliary Pancreat Sci 2022; 29: 950-959
• 10 Bang JY, Hebert-Magee S, Navaneethan U. et al. EUS-guided fine needle biopsy of pancreatic masses can yield true histology. Gut 2018; 67: 2081-2084
• 11 Bang JY, Krall K, Jhala N. et al. Comparing needles and methods of endoscopic ultrasound-guided fine-needle biopsy to optimize specimen quality and diagnostic accuracy for patients with pancreatic masses in a randomized trial. Clin Gastroenterol Hepatol 2021; 19: 825-835
• 12 Tempero MA, Malafa MP, Chiorean EG. et al. Pancreatic adenocarcinoma, version 1.2019. J Natl Compr Canc Netw 2019; 17: 202-210
• 13 Mertz H, Gautam S. The learning curve for EUS-guided FNA of pancreatic cancer. Gastrointest Endosc 2004; 59: 33-37
• 14 Eloubeidi MA, Tamhane A. EUS-guided FNA of solid pancreatic masses: a learning curve with 300 consecutive procedures. Gastrointest Endosc 2005; 61: 700-708
• 15 Wani S, Keswani R, Hall M. et al. A prospective multicenter study evaluating learning curves and competence in endoscopic ultrasound and endoscopic retrograde cholangiopancreatography among advanced endoscopy trainees: the rapid assessment of trainee endoscopy skills study. Clin Gastroenterol Hepatol 2017; 15: 1758-1767
• 16 Li J, Lu J, Yan J. et al. Artificial intelligence can increase the detection rate of colorectal polyps and adenomas: a systematic review and meta-analysis. Eur J Gastroenterol Hepatol 2021; 33: 1041-1048
• 17 Wu HL, Yao LW, Wu LL. et al. Validation of a real-time biliopancreatic endoscopic ultrasonography analytical device in China: a prospective, single-centre, randomized controlled trial. Lancet Digit Health 2023; 5: e812-e820
• 18 Udristoiu AL, Cazacu IM, Gruionu LG. et al. Real-time computer-aided diagnosis of focal pancreatic masses from endoscopic ultrasound imaging based on a hybrid convolutional and long short-term memory neural network model. PLoS One 2021; 16: e0251701
• 19 BangJYSaftoiu A, Udristoiu A. et al. Real-time AI-enhanced analysis of EUS movies used for detection and segmentation of cystic pancreatic lesions and focal pancreatic masses. Gastrointest Endosc 2024; 99: AB38
• 20 Hawes RH, Varadarajulu S, Fockens P. How to perform an endoscopic ultrasonography examination of the pancreas, bile duct, and liver. In: Varadarajulu S, Fockens P, Hawes RH. , ed. Endosonography. 5th edn. Amsterdam: Elsevier; 2023: 120-131
• 21 Vege SS, Ziring B, Jain R. et al. American Gastroenterological Association Institute guideline on the diagnosis and management of asymptomatic neoplastic pancreatic cysts. Gastroenterology 2015; 148: 819-822
• 22 Bang JY, Navaneethan U, Hasan MK. et al. Endoscopic ultrasound-guided specimen collection and evaluation techniques affect diagnostic accuracy. Clin Gastroenterol Hepatol 2018; 16: 1820-1828
• 23 Dumonceau JM, Polkowski M, Larghi A. et al. Indications, results, and clinical impact of endoscopic ultrasound (EUS)-guided sampling in gastroenterology: European Society of Gastrointestinal Endoscopy (ESGE) clinical guideline. Endoscopy 2011; 43: 897-909
• 24 Kuwahara T, Hara K, Mizuno N. et al. Artificial intelligence using deep learning analysis of endoscopic ultrasonography images for the differential diagnosis of pancreatic masses. Endoscopy 2023; 55: 140-149
• 25 Marya N, Powers P, Chari S. et al. Utilization of artificial intelligence for the development of an EUS-convolutional network model trained to enhance the detection of autoimmune pancreatitis. Gut 2021; 70: 1335-1344
• 26 Cui H, Zhao Y, Xiong S. et al. Diagnosing solid lesions in the pancreas with multimodal artificial intelligence: a randomized crossover trial. JAMA Netw Open 2024; 7: e2422454