Clinical Application of Artificial Intelligence in New Borns With Cleft Lip and Palate

Official title

Effect of Automated AI Designed Presurgical Plate Therapy on Cleft Size Reduction in Newborns With Cleft Lip and Palate : A Comparative Study

Overview

The goal of this observational study is to evaluate whether passive plate therapy using automated AI-designed devices can reduce cleft size in newborns with cleft lip and palate. The main questions it aims to answer are: Does AI-designed passive plate therapy reduce the cleft size between birth and primary surgery? (Primary aim) How does cleft size at the time of surgery compare between infants who received passive plate therapy and those who did not? (Secondary aim) Researchers will compare infants who received AI-designed passive plate therapy with those who received no presurgical therapy to determine whether the intervention leads to a greater reduction in cleft width. Participants will: Undergo intraoral scans at birth and again at the time of primary surgery, around 4 months of age. Receive either no presurgical intervention or be treated with AI-designed passive plates, depending on site-specific clinical practices

Detailed description

This observational study investigates the impact of passive plate therapy using an automated, artificial intelligence (AI)-driven design workflow on presurgical cleft size in infants with unilateral cleft lip and palate. Cleft lip and palate represent one of the most common congenital craniofacial anomalies, and early presurgical interventions such as passive plates aim to reduce the cleft size, support feeding, and facilitate better surgical outcomes. However, access to such interventions is often limited by the need for specialized staff, complex workflows, and reliance on intraoral impressions.

Recent technological advancements have enabled the integration of digital workflows and AI into presurgical cleft care. In particular, a pipeline has been developed that uses intraoral scanning, and AI-assisted modeling to design individualized passive plates. These plates are manufactured via 3D printing and do not require invasive impressions nor extensive laboratory work making them potentially safer and more scalable in low or medium resource settings.

This study specifically evaluates the clinical effectiveness of such AI-designed passive plates compared to standard care without any presurgical orthopedic therapy. Infants are enrolled from two sites in India (Chennai and Hyderabad), where clinical practices differ with respect to the use of passive plates. As this is a non-randomized, observational study, group assignment is based on local standard of care at each site.

The primary objective is to assess the percentage reduction in cleft width from birth to the time of primary surgery (typically around 4 months of age) in infants treated with AI-designed passive plates. The secondary objective is to compare cleft width at the time of surgery between infants who received the plates and those who did not, offering insight into the relative anatomical outcomes of the intervention.

All participating infants will undergo standardized intraoral scans at baseline (within the first two weeks of life) and again just prior to primary surgical repair. The scans are used to measure the anterior-posterior cleft width and calculate percentage change over time. Cleft measurements are obtained digitally from 3D scan data using validated image processing software.

The AI-assisted design of the passive plates is performed using a custom plugin within Blender software, which automatically detects anatomical landmarks and generates the plate geometry with minimal user input. The digital files are subsequently exported for 3D printing using biocompatible materials. Infants in the intervention group will wear the plate continuously from the time of fitting until primary surgery, under the supervision of trained clinical teams.

Data from the two cohorts (plate vs. no plate) will be compared using appropriate statistical methods to assess differences in cleft size reduction and absolute cleft width at surgery. This study does not include randomization or blinding, as the intervention is assigned based on institutional practice. However, efforts will be made to ensure consistency in scanning methods, measurement protocols, and outcome assessment across both sites.

This study is part of a larger initiative to evaluate the generalizability and effectiveness of AI-based digital workflows in cleft care across diverse healthcare settings. The findings are expected to inform future integration of automated design technologies in presurgical treatment planning, especially in regions where access to skilled cleft teams is limited.

No additional interventions, medications, or behavioral changes are introduced as part of this study. Participation involves routine procedures and follow-up through the standard cleft treatment timeline.

Interventions

• Other AI-Designed Passive Plate Therapy
  A removable passive orthopedic plate designed using an AI-assisted digital workflow based on intraoral scan data. The design process is automated via a Blender-based software plugin that identifies anatomical landmarks and generates a custom-fit plate. The plate is 3D printed using biocompatible material and fitted shortly after birth. It is worn continuously until primary surgical repair, typically around 4 months of age. The intervention aims to reduce cleft width prior to surgery without requ

Primary outcome measures

• Percent Reduction in Cleft Width from Birth to Primary Surgery [Time frame: From birth to time of primary cleft surgery (approximately 4 months of age)]

Secondary outcome measures (1)

• Cleft Width at Time of Primary Surgery [Time frame: At the time of primary cleft surgery (approximately 4 months of age)]

Eligibility criteria

Inclusion criteria

Infants diagnosed with unilateral cleft lip and palate Age at enrollment: within the first 14 days of life (for treatment group) Age at cleft surgery: approximately 4 months Medically stable and fit to undergo intraoral scanning and cleft surgery Parent or legal guardian has provided written informed consent

Exclusion criteria

Syndromic cleft lip and palate or other craniofacial syndromes Bilateral cleft lip and palate Significant comorbidities affecting feeding, growth, or surgery (e.g., cardiac anomalies) Premature infants (<37 weeks gestational age at birth) Infants who have already undergone any presurgical orthopedic intervention elsewhere Guardians/ Parents unwilling or unable to comply with follow-up or study procedures

Criteria are shown verbatim from the registry (in English). Final eligibility is always assessed by the study center.

Healthy volunteers: No

Study design

Allocation

Non-randomized

Model

Parallel assignment

Masking

Open label

Primary purpose

Treatment

Study locations

India · 2 centers

• Saveetha Medical College — Chennai
• GSR Hospital Hyderabad — Hyderabad

Publications

• Nalabothu P, Benitez BK, Dalstra M, Verna C, Mueller AA. Three-Dimensional Morphological Changes of the True Cleft under Passive Presurgical Orthopaedics in Unilateral Cleft Lip and Palate: A Retrospective Cohort Study. J Clin Med. 2020 Mar 31;9(4):962. doi: 10.3390/jcm9040962. PMID 32244361
• Benitez BK, Brudnicki A, Surowiec Z, Wieprzowski L, Rasadurai A, Nalabothu P, Lill Y, Mueller AA. Digital impressions from newborns to preschoolers with cleft lip and palate: A two-centers experience. J Plast Reconstr Aesthet Surg. 2022 Nov;75(11):4233-4242. doi: 10.1016/j.bjps.2022.08.015. Epub 2022 Aug 22. PMID 36154982
• Zarean P, Zarean P, Thieringer FM, Mueller AA, Kressmann S, Erismann M, Sharma N, Benitez BK. A Point-of-Care Digital Workflow for 3D Printed Passive Presurgical Orthopedic Plates in Cleft Care. Children (Basel). 2022 Aug 20;9(8):1261. doi: 10.3390/children9081261. PMID 36010151
• Meyer S, Benitez BK, Thieringer FM, Mueller AA. Three-Dimensional Printable Open-Source Cleft Lip and Palate Impression Trays: A Single-Impression Workflow. Plast Reconstr Surg. 2024 Feb 1;153(2):462-465. doi: 10.1097/PRS.0000000000010684. Epub 2023 May 15. PMID 37184464
• Schnabel TN, Gozcu B, Gotardo P, Lingens L, Dorda D, Vetterli F, Emhemmed A, Nalabothu P, Lill Y, Benitez BK, Mueller AA, Gross M, Solenthaler B. Automated and data-driven plate computation for presurgical cleft lip and palate treatment. Int J Comput Assist Radiol Surg. 2023 Jun;18(6):1119-1125. doi: 10.1007/s11548-023-02858-6. Epub 2023 Apr 2. PMID 37009952

Identifiers

NCT: NCT06970158 · placeholder unique ID

Primary sources (government registries)

• ClinicalTrials.gov (US NIH)

View this study on ClinicalTrials.gov ↗