Exploring the Feasibility of a Deep Learning Algorithm for Postoperative Outcome Assessment in Unilateral Cleft Lip Repair: A Pilot Study.
Uloženo v:
| Název: | Exploring the Feasibility of a Deep Learning Algorithm for Postoperative Outcome Assessment in Unilateral Cleft Lip Repair: A Pilot Study. |
|---|---|
| Autoři: | Daiem M; Department of Plastic Surgery, CLAPP Hospital, Lahore, Pakistan.; Operation Smile Incorporated, Virginia Beach, VA.; Department of Plastic Surgery, Aga Khan University, Karachi., Fayyaz GQ; Department of Plastic Surgery, CLAPP Hospital, Lahore, Pakistan., Irfan S; Department of Plastic Surgery, Aga Khan University, Karachi., Ali RN; Department of Plastic Surgery, College of Physicians and Surgeons Multan., Ali RZ; Department of Computer Science, COMSATS University., Bashir MM; Department of Plastic Surgery, King Edward Medical University, Lahore, Pakistan., Turk M; Division of Plastic and Reconstructive Surgery, Keck School of Medicine, University of Southern California, Los Angeles, CA., Pontell ME; Department of Plastic Surgery, Vanderbilt University Medical Center, Nashville, TN., Miles MG; Division of Plastic and Reconstructive Surgery, Lehigh Valley Health Network, Allentown, PA., Nolte J; Division of Plastic and Reconstructive Surgery, Amsterdam UMC, Amsterdam, The Netherlands., Breugem C; Division of Plastic and Reconstructive Surgery, Amsterdam UMC, Amsterdam, The Netherlands. |
| Zdroj: | The Journal of craniofacial surgery [J Craniofac Surg] 2026 Mar-Apr 01; Vol. 37 (3-4), pp. 767-773. Date of Electronic Publication: 2025 Nov 19. |
| Způsob vydávání: | Journal Article |
| Jazyk: | English |
| Informace o časopise: | Publisher: Lippincott Williams & Wilkins Country of Publication: United States NLM ID: 9010410 Publication Model: Print-Electronic Cited Medium: Internet ISSN: 1536-3732 (Electronic) Linking ISSN: 10492275 NLM ISO Abbreviation: J Craniofac Surg Subsets: MEDLINE |
| Imprint Name(s): | Publication: <2014-> : Hagerstown, MD : Lippincott Williams & Wilkins Original Publication: Burlington, Ont. : B.C. Decker, c1990- |
| Výrazy ze slovníku MeSH: | Cleft Lip*/surgery , Outcome Assessment, Health Care*/methods , Deep Learning* , Algorithms*, Humans ; Pilot Projects ; Feasibility Studies ; Infant ; Female ; Male ; Reoperation |
| Abstrakt: | Primary surgical repair for a cleft lip is often performed around 3 to 4 months of age. Traditional outcome assessments rely heavily on subjective clinical judgment and structured rating tools such as the Cleft Aesthetic Rating Scale (CARS), both of which are limited by inter-rater variability and lack of scalability. This pilot study explores the feasibility of a deep learning (DL) algorithm to assist in postoperative outcome assessment and identify patients at increased risk for revision following unilateral cleft lip repair. The authors developed a convolutional neural network based on the EfficientNet-B1 architecture, trained on a class-balanced data set of 500 standardized postoperative facial photographs labeled using real-world revision outcomes. Rigorous data preprocessing, augmentation, and validation protocols were used to ensure model robustness. The model achieved an accuracy of 74% and an area under the ROC curve (AUC) of 0.79 on an independent test set, with a recall of 76% for identifying patients needing revision. Confidence score distribution and t-SNE feature space visualization demonstrated reliable class separation and interpretability. Our findings suggest that DL algorithms, when trained on structured clinical data, can offer meaningful support in cleft care. This model serves as an early prototype for AI-assisted triage systems that could aid in identifying revision candidates, particularly in resource-limited or outreach settings. While preliminary, this study highlights the potential for integrating artificial intelligence into routine cleft lip postoperative workflows to improve equity and consistency of care. (Copyright © 2025 by Mutaz B. Habal, MD.) |
| Competing Interests: | The authors report no conflicts of interest. |
| References: | Crockett DJ, Goudy SL. Cleft lip and palate. Facial Plast Surg Clin North Am 2014;22:573–586. Farronato G, Cannalire P, Martinelli G, et al. Cleft lip and/or palate: review. Minerva Stomatol 2014;63:111–126. Taub PJ, Piccolo P. Cleft lip repair: through the looking glass. J Craniofac Surg 2016;27:2031–2035. Mulliken JB, LaBrie RA. Fourth-dimensional changes in nasolabial dimensions following rotation-advancement repair of unilateral cleft lip Multan. Plast Reconstr Surg 2012;129:491–498. Vyas RM, Warren SM. Unilateral cleft lip repair. Clin Plast Surg 2014;41:165–177. Roussel LO, Myers RP, Girotto JA. The Millard rotation-advancement cleft lip repair: 50 years of modification. Cleft Palate Craniofac J 2015;52:e188–e195. Kim SK, Kim MH, Kwon YS, et al. Long-term results in the bilateral cleft lip repair by Mulliken’s method. J Craniofac Surg 2009;20:1455–1461. Brauer RO, Cronin TD. The Tennison lip repair revisited. Plast Reconstr Surg 1983;71:633–642. Fisher DM. Unilateral cleft lip repair: an anatomical subunit approximation technique. Plast Reconstr Surg 2005;116:61–71. Mohler LR. Unilateral cleft lip repair. Plast Reconstr Surg 1987;80:511–517. Sadove AM, Eppley BL. Correction of secondary cleft lip and nasal deformities. Clin Plast Surg 1993;20:793–801. Plonkowski AT, Choi DG, Turk M, et al. The rate of revision surgery for secondary deformities of the lip in cleft lip patients: a systematic review. J Plast Reconstr Aesthet Surg 2025;109:99–107. Sitzman TJ, Mara CA, Long RE Jr, et al. The Americleft Project: burden of care from secondary surgery. Plast Reconstr Surg Glob Open 2015;3:e442. Wells-Durand E, Buchel A, Tuen YJ, et al. What does cleft lip and palate care cost? The time and economic-associated burden of care from birth to maturity. Plast Surg (Oakv) 2025;33:139–148. Asher-McDade C, Roberts C, Shaw WC, et al. Development of a method for rating nasolabial appearance in patients with clefts of the lip and palate. Cleft Palate Craniofac J 1991;28:385–391. Mosmuller DGM, Mennes LM, Prahl C, et al. The development of the cleft aesthetic rating scale: a new rating scale for the assessment of nasolabial appearance in complete unilateral cleft lip and palate patients. Cleft Palate Craniofac J 2017;54:555–561. Ritter K, Trotman CA, Phillips C. Validity of subjective evaluations for the assessment of lip scarring and impairment. Cleft Palate Craniofac J 2002;39:587–596. Klein GM, Dhawan A, Laskowski RA, et al. Comparison of the basal view and a previously standardized cleft lip rating scale. Cleft Palate Craniofac J 2018;55:1277–1281. Tan RA, Mulder FJ, Schwirtz RMF, et al. Atypical outcomes of nasal and lip appearance after unilateral cleft lip repair: judgment by professionals, patients, and laypeople. Cleft Palate Craniofac J 2021;58:1226–1235. Zhang H, Botler M, Kooman JP. Deep learning for image analysis in kidney care. Adv Kidney Dis Health 2023;30:25–32. Jiang Y, Yang M, Wang S, et al. Emerging role of deep learning-based artificial intelligence in tumor pathology. Cancer Commun (Lond) 2020;40:154–166. Dildar M, Akram S, Irfan M, et al. Skin cancer detection: a review using deep learning techniques. Int J Environ Res Public Health 2021;18:5479. Esteva A, Robicquet A, Ramsundar B, et al. A guide to deep learning in healthcare. Nat Med 2019;25:24–29. Jiang F, Jiang Y, Zhi H, et al. Artificial intelligence in healthcare: past, present and future. Stroke Vasc Neurol 2017;2:230–243. Baig R, Bibi M, Hamid A, et al. Deep learning approaches towards skin lesion segmentation and classification from dermoscopic images—a review. Curr Med Imaging 2020;16:513–533. Nogueira R, Eguchi M, Kasmirski J, et al. Machine learning, deep learning, artificial intelligence and aesthetic plastic surgery: a qualitative systematic review. Aesthetic Plast Surg 2025;49:389–399. Eldaly AS, Avila FR, Torres-Guzman RA, et al. Simulation and artificial intelligence in rhinoplasty: a systematic review. Aesthetic Plast Surg 2022;46:2368–2377. Huqh MZU, Abdullah JY, Wong LS, et al. Clinical applications of artificial intelligence and machine learning in children with cleft lip and palate—a systematic review. Int J Environ Res Public Health 2022;19:10860. Rosero K, Salman AN, Harrison LM, et al. Deep learning-based assessment of lip symmetry for patients with repaired cleft lip. Cleft Palate Craniofac J 2025;62:289–299. Tan M, Le Q. EfficientNet: rethinking model scaling for convolutional neural networks. Proceedings of the 36th International Conference on Machine Learning 2019;97:6105–6114. Al-Namankany A, Alhubaishi A. Effects of cleft lip and palate on children’s psychological health: a systematic review. J Taibah Univ Med Sci 2018;13:311–318. Hunt O, Burden D, Hepper P, et al. The psychosocial effects of cleft lip and palate: a systematic review. Eur J Orthod 2005;27:274–285. Salinero LK, Romeo DJ, Pontell ME, et al. Psychosocial status and self-perception in patients with cleft lip and/or palate. Cleft Palate Craniofac J 2025;62:1079–1086. Garland K, Matic D. Current approaches to cleft lip revision. Curr Opin Otolaryngol Head Neck Surg 2019;27:287–293. Naidu P, Plonkowski AT, Yao CA, et al. A summary of the existing data on cleft surgical outcomes: what do we not know? Plast Reconstr Surg Glob Open 2025;13:e6660. Weidler EM, Britto MT, Sitzman TJ. Facilitators and barriers to implementing standardized outcome measurement for children with cleft lip and palate. Cleft Palate Craniofac J 2021;58:7–18. Trotman CA, Phillips C, Faraway JJ, et al. Association between subjective and objective measures of lip form and function: an exploratory analysis. Cleft Palate Craniofac J 2003;40:241–248. Christofides E, Potgieter A, Chait L. A long term subjective and objective assessment of the scar in unilateral cleft lip repairs using the Millard technique without revisional surgery. J Plast Reconstr Aesthet Surg 2006;59:380–386. Almoammar KA. Harnessing the power of artificial intelligence in cleft lip and palate: an in-depth analysis from diagnosis to treatment, a comprehensive review. Children (Basel) 2024;11:140. Dhillon H, Chaudhari PK, Dhingra K, et al. Current applications of artificial intelligence in cleft care: a scoping review. Front Med (Lausanne) 2021;8:676490. Lannon CM, Nuchovich N, Louis C, et al. Factors associated with lack of long-term follow-up data after global cardiac surgery missions. World J Pediatr Congenit Heart Surg 2024;15:325–331. Velin L, Lantz A, Ameh EA, et al. Systematic review of low-income and middle-income country perceptions of visiting surgical teams from high-income countries. BMJ Glob Health 2022;7:e008791. Tong CYM, Koh RYV, Lee ES. A scoping review on the factors associated with the lost to follow-up (LTFU) amongst patients with chronic disease in ambulatory care of high-income countries (HIC). BMC Health Serv Res 2023;23:883. Lucas C, Torres-Guzman R, James AJ, et al. Machine learning for automatic detection of velopharyngeal dysfunction: a preliminary report. J Craniofac Surg 2025;36:816–819. |
| Contributed Indexing: | Keywords: Artificial intelligence; Cleft Aesthetic Rating Scale (CARS); cleft lip revision; deep learning; postoperative outcome assessment; unilateral cleft lip |
| Entry Date(s): | Date Created: 20251120 Date Completed: 20260306 Latest Revision: 20260306 |
| Update Code: | 20260307 |
| DOI: | 10.1097/SCS.0000000000012199 |
| PMID: | 41263442 |
| Databáze: | MEDLINE |
Nájsť tento článok vo Web of Science | Abstrakt: | Primary surgical repair for a cleft lip is often performed around 3 to 4 months of age. Traditional outcome assessments rely heavily on subjective clinical judgment and structured rating tools such as the Cleft Aesthetic Rating Scale (CARS), both of which are limited by inter-rater variability and lack of scalability. This pilot study explores the feasibility of a deep learning (DL) algorithm to assist in postoperative outcome assessment and identify patients at increased risk for revision following unilateral cleft lip repair. The authors developed a convolutional neural network based on the EfficientNet-B1 architecture, trained on a class-balanced data set of 500 standardized postoperative facial photographs labeled using real-world revision outcomes. Rigorous data preprocessing, augmentation, and validation protocols were used to ensure model robustness. The model achieved an accuracy of 74% and an area under the ROC curve (AUC) of 0.79 on an independent test set, with a recall of 76% for identifying patients needing revision. Confidence score distribution and t-SNE feature space visualization demonstrated reliable class separation and interpretability. Our findings suggest that DL algorithms, when trained on structured clinical data, can offer meaningful support in cleft care. This model serves as an early prototype for AI-assisted triage systems that could aid in identifying revision candidates, particularly in resource-limited or outreach settings. While preliminary, this study highlights the potential for integrating artificial intelligence into routine cleft lip postoperative workflows to improve equity and consistency of care.<br /> (Copyright © 2025 by Mutaz B. Habal, MD.) |
|---|---|
| ISSN: | 1536-3732 |
| DOI: | 10.1097/SCS.0000000000012199 |