Artificial intelligence in preoperative assessment of gallbladder polyps.

Dr Prince  Pankaj; Dr. Amjad Zia  Mallik; Dr. Shambhu Kumar  Singh

doi:10.51168/sjhrafrica.v7i2.2700

Authors

Dr .Prince Pankaj Assistant Professor, Department of General Surgery, Katihar Medical College and Hospital Katihar Bihar.
Dr. AmjadZia Mallik Professor, Department of General Surgery, Katihar Medical College and Hospital, Katihar, Bihar.
Dr . Shambhu Kumar Singh Associate Professor, Department of General Surgery, Katihar Medical College and Hospital, Katihar, Bihar.

DOI:

https://doi.org/10.51168/sjhrafrica.v7i2.2700

Keywords:

Artificial Intelligence, Gallbladder Polyps, Preoperative Assessment, Convolutional Neural Networks, Diagnostic Accuracy.

Abstract

Background

Gallbladder polyps (GBPs) are increasingly detected due to widespread use of ultrasonography. Although most polyps are benign, certain characteristics can indicate malignancy, necessitating surgical intervention. Differentiating benign from malignant polyps preoperatively remains a diagnostic challenge, often leading to unnecessary cholecystectomies. Artificial Intelligence (AI), particularly machine learning (ML) and deep learning (DL) algorithms, has emerged as a promising tool in enhancing diagnostic accuracy.

Objective

This study aims to evaluate the role of AI-based imaging analysis in the preoperative assessment of gallbladder polyps to improve diagnostic precision and reduce unnecessary surgical procedures.

Methods

A retrospective dataset of 420 patients with diagnosed gallbladder polyps was analyzed. Ultrasonographic and radiologic images were processed using a convolutional neural network (CNN) trained on annotated cases classified by histopathological outcomes. The model evaluated polyp size, echogenicity, base attachment, and growth patterns. Statistical comparisons were made between AI prediction outcomes and actual histopathology reports. Sensitivity, specificity, and accuracy were calculated.

Results

The AI model demonstrated a sensitivity of 91.3% and specificity of 87.6% in differentiating neoplastic from non-neoplastic polyps. It achieved an overall diagnostic accuracy of 89.2%, outperforming human radiologists whose average diagnostic accuracy was 74.5%. Notably, the AI algorithm reduced false positives in polyps <10mm, which traditionally lead to overtreatment. ROC curve analysis yielded an AUC of 0.93, indicating high diagnostic reliability. Integration of clinical metadata (e.g., age, BMI, and lipid profile) further improved performance metrics.

Conclusion

AI-driven analysis of gallbladder polyp imaging provides a valuable adjunct in preoperative decision-making. Its superior diagnostic performance compared to conventional radiology holds promise for minimizing unnecessary surgeries and optimizing patient outcomes. Future multicentric studies and prospective validation are necessary before clinical implementation.

Author Biographies

Dr .Prince Pankaj, Assistant Professor, Department of General Surgery, Katihar Medical College and Hospital Katihar Bihar.

is Assistant Professor in the Department of General Surgery at Katihar Medical College and Hospital, Katihar, Bihar, with academic and clinical interests in gastrointestinal and hepatobiliary surgery.

Dr. AmjadZia Mallik, Professor, Department of General Surgery, Katihar Medical College and Hospital, Katihar, Bihar.

is Professor in the Department of General Surgery at Katihar Medical College and Hospital, Katihar, Bihar, with extensive experience in surgical education and advanced general surgery.

Dr . Shambhu Kumar Singh, Associate Professor, Department of General Surgery, Katihar Medical College and Hospital, Katihar, Bihar.

is Associate Professor in the Department of General Surgery at Katihar Medical College and Hospital, Katihar, Bihar, with research interests in clinical surgery and perioperative patient care.

References

Yuan K, Zhang X, Yang Q, Deng X, Deng Z, Liao X, Si W. Risk prediction and analysis of gallbladder polyps with deep neural network. Comput Assist Surg. 2024;29(1):2331774. Doi:10.1080/24699322.2024.2331774.

Ward TM, Hashimoto DA, Ban Y, Rosman G, Meireles OR. Artificial intelligence prediction of cholecystectomy operative course from automated identification of gallbladder inflammation. Surg Endosc. 2022;36(9):6832–6840. Doi:10.1007/s00464-021-08957-y.

Ward TM, Hashimoto DA, Ban Y, Rosman G, Meireles OR. Artificial intelligence prediction of cholecystectomy operative course from automated identification of gallbladder inflammation. Surg Endosc. 2022;36(9):6832–6840. doi:10.1007/s00464-021-08957-y

Jeong Y, Kim JH, Chae HD, Park SJ, Bae JS, Joo I, Han JK. Deep learning-based decision support system for the diagnosis of neoplastic gallbladder polyps on ultrasonography: preliminary results. Sci Rep. 2020;10(1):7700. Doi:10.1038/s41598-020-64601-0.

Li Q, Zhang J, Cai Z, Jia P, Wang X, Geng X, Zhang D. A Bayesian network prediction model for gallbladder polyps with malignant potential based on preoperative ultrasound. Surg Endosc. 2023;37(1):518–527. Doi:10.1007/s00464-022-09613-4.

Han S, Liu Y, Li X, Jiang X, Li B, Zhang C, Zhang J. Development and validation of a preoperative nomogram for predicting benign and malignant gallbladder polypoid lesions. Front Oncol. 2022;12:800449. Doi:10.3389/fonc.2022.800449.

Yuan HX, Yu QH, Zhang YQ, Yu Q, Zhang Q, Wang WP. Ultrasound radiomics effective for preoperative identification of true and pseudo gallbladder polyps based on spatial and morphological features. Front Oncol. 2020;10:1719. Doi:10.3389/fonc.2020.01719.

Bai M, Guo R, Zhao Q, Li Y. Artificial intelligence-based CT images in analysis of postoperative recovery of patients undergoing laparoscopic cholecystectomy under balanced anesthesia. Sci Program. 2021;2021:1125573. doi:10.1155/2021/1125573.

McGivern K, Drake T, Knight S, Lucocq J, Bernabeu MO, Clark N, Harrison EM. Applying artificial intelligence to big data in hepatopancreatic and biliary surgery: a scoping review. Artif Intell Surg. 2023;3(1):27–47. doi:10.1007/s43681-023-00228-1.

Liu J, Qian Y, Yang F, Huang S, Chen G, Yu J, Huang G. Value of prediction model in distinguishing gallbladder adenoma from cholesterol polyp. J Gastroenterol Hepatol. 2022;37(10):1893–1900. Doi:10.1111/jgh.15969.

Li Q, Dou M, Zhang J, Jia P, Wang X, Lei D, Zhang D. A Bayesian network model to predict neoplastic risk for patients with gallbladder polyps larger than 10 mm based on preoperative ultrasound features. Surg Endosc. 2023;37(7):5453–5463. Doi:10.1007/s00464-023-10329-5.

Abbing JR, Voskens FJ, Gerats BG, Egging RM, Milletari F, Broeders IA. Towards an AI-based assessment model of surgical difficulty during early phase laparoscopic cholecystectomy. Comput Methods Biomech Biomed Eng Imaging Vis. 2023;11(4):1299–1306. Doi:10.1080/21681163.2023.2208880.

Wu S, Chen Z, Liu R, Li A, Cao Y, Wei A, Wang X. SurgSmart: an artificial intelligent system for quality control in laparoscopic cholecystectomy: an observational study. Int J Surg. 2023;109:1105–1114. Doi:10.1016/j.ijsu.2023.08.008.

Foley KG, Riddell Z, Coles B, Roberts SA, Willis BH. Risk of developing gallbladder cancer in patients with gallbladder polyps detected on transabdominal ultrasound: a systematic review and meta-analysis. Br J Radiol. 2022;95(1137):20220152. Doi:10.1259/bjr.20220152.

Khalid MU, Laplante S, Masino C, Alseidi A, Jayaraman S, Zhang H, Madani A. Use of artificial intelligence for decision-support to avoid high-risk behaviors during laparoscopic cholecystectomy. Surg Endosc. 2023;37(12):9467–9475. doi:10.1007/s00464-023-10385-1.