DEEP LEARNING-BASED DETERMINATION OF INTRAOCULAR LENS POSITION IN CATARACT SURGERY PATIENTS
Main Article Content
Keywords
Deep Learning, Cataract Surgery, Intraocular Lens, Artificial Intelligence, IOL Position, Ophthalmology.
Abstract
Background: Cataract surgery using intraocular lens (IOL) implantation is one of the most common ophthalmic surgeries studied, yet accurate evaluation of the location of the IOL after surgery is the most important to the optimal vision.
Purpose: To quantify the effectiveness of a deep learning-trained model to predict the intraocular lens position of patients who have undergone cataract surgery and the effect it has on the post-surgical visual outcomes.
Methods: This analytical observation study was conducted at Prince Abdul-Aziz Bin Mosaad Hospital Arar, KSA during the period of September, 2025 to February, 2025 and with a sample size of 300 cataract surgery patients. Based on the preoperative biometric information, intraoperative surgical videos, and postoperative imaging, a convolutional neural network (CNN) framework was constructed. Predicted IOL positions were compared with the real measurements obtained with the anterior segment OCT and statistical tests of accuracy, sensitivity and specificity were used. Pearson correlation was used.
Findings: The deep learning model performed well having an accuracy of 92.6, sensitivity of 90.8, and specificity of 94.2. The actual and predicted IOL positions had a close positive relationship (r = 0.81, p < 0.001). The model was also validated in terms of clinical reliability in the detection of IOL decentration and tilt.
Conclusion: Deep learning provides a viable and accurate approach to determine the position of intraocular lens, which can significantly enhance the accuracy and postoperative results of surgery in cataract patients.
References
2. Kiuchi, G., Tanabe, M., Nagata, K., Ishitobi, N., Tabuchi, H., & Oshika, T. (2022). Deep learning-based system for preoperative safety management in cataract surgery. Journal of Clinical Medicine, 11(18), 5397.
3. Ghamsarian, N. (2023). Deep-learning-assisted analysis of cataract surgery videos. arXiv preprint arXiv:2312.05900.
4. Gatinel, D., Debellemanière, G., Saad, A., Dubois, M., & Rampat, R. (2021). Determining the theoretical effective lens position of thick intraocular lenses for machine learning–based IOL power calculation and simulation. Translational Vision Science & Technology, 10(4), 27.
5. Lu, S., Ba, L., Wang, J., Zhou, M., Huang, P., Zhang, X., Pan, S., Zhou, X., Wen, K., & Sun, J. (2025). Deep learning-driven approach for cataract management: Towards precise identification and predictive analytics. Frontiers in Cell and Developmental Biology, 13, 1611216.
6. Yeh, H. H., Jain, A. M., Fox, O., Sebov, K., & Wang, S. Y. (2023). PhacoTrainer: Deep learning for cataract surgical videos to track surgical tools. Translational Vision Science & Technology, 12(3), 23.
7. Baumann, C. (n.d.). Automatic quantification of toric intraocular lens orientation based on visual computing and deep learning (Doctoral dissertation, Technische Universität Wien).
8. Ghamsarian, N., El-Shabrawi, Y., Nasirihaghighi, S., Putzgruber-Adamitsch, D., Zinkernagel, M., Wolf, S., Schoeffmann, K., & Sznitman, R. (2024). Cataract-1k dataset for deep-learning-assisted analysis of cataract surgery videos. Scientific Data, 11(1), 373.
9. Mori, Y., Yamauchi, T., Tokuda, S., Minami, K., Tabuchi, H., & Miyata, K. (2021). Machine learning adaptation of intraocular lens power calculation for a patient group. Eye and Vision, 8(1), 42.
10. Tham, Y. C., Goh, J. H., Anees, A., Lei, X., Rim, T. H., Chee, M. L., Wang, Y. X., Jonas, J. B., Thakur, S., Teo, Z. L., & Cheung, N. (2022). Detecting visually significant cataract using retinal photograph-based deep learning. Nature Aging, 2(3), 264–271.
11. Yeh, H. H., Jain, A. M., Fox, O., & Wang, S. Y. (2021). PhacoTrainer: A multicenter study of deep learning for activity recognition in cataract surgical videos. Translational Vision Science & Technology, 10(13), 23.
12. Yuan, D. Q., Tang, F. N., Wang, Y., Zhang, H., Zhang, W. W., Gu, L. W., & Liu, Q. H. (2025). Prediction of postoperative vault after implantable collamer lens implantation with deep learning. International Journal of Ophthalmology, 18(7), 1197.
13. Gutierrez, L., Lim, J. S., Foo, L. L., Ng, W. Y., Yip, M., Lim, G. Y., Wong, M. H., Fong, A., Rosman, M., Mehta, J. S., & Lin, H. (2022). Application of artificial intelligence in cataract management: Current and future directions. Eye and Vision, 9(1), 3.
14. Yang, Y., Long, Z., Lei, B., Liu, W., & Ye, J. (2024). Clinical decision support system based on deep learning for evaluating implantable collamer lens size and vault after implantable collamer lens surgery: A retrospective study. BMJ Open, 14(2), e081050.
15. Ottonelli, G., De Rosa, G., Celada Ballanti, J., Gaeta, A., Vinciguerra, P., & Di Maria, A. (2025). Does DeepSeek provide clinically acceptable intraocular lens (IOL) power predictions in cataract surgery? A proof-of-concept study. Journal of Clinical Medicine, 14(24), 8870.
16. Liu, Z., Zhang, Z., Chen, W., Wang, Y., Cai, H., & Chen, X. (2024). A novel learning-based keypoint matching framework for toric intraocular lens navigation during cataract surgery. In Optoelectronic Imaging and Multimedia Technology XI (Vol. 13239, pp. 45–56). SPIE.
17. Junayed, M. S., Islam, M. B., Sadeghzadeh, A., & Rahman, S. (2021). CataractNet: An automated cataract detection system using deep learning for fundus images. IEEE Access, 9, 128799–128808.
Article Sidebar
Article Details
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
All articles published in JPTCP are licensed under Copyright Creative Commons Attribution-NonCommercial 4.0 International License.
Zafar Iqbal
FRCS, Consultant Ophthalmologist, Dr. Ziauddin Hospital / Ex-Head of Department, Prince Abdulaziz Bin Mosaad Hospital, Arar, KSA
Sidra Zafar Iqbal
MD, Johns Hopkins Hospital, Baltimore, USA
