Smartphone Scanning and Machine Learning for Automated Presurgical 3D-printed Plate Fabrication From Cleft...

Area of Science:

• Medical Technology
• Digital Health
• Craniofacial Surgery

Background:

• Accurate palatal impressions are crucial for infant cleft lip and palate (CLP) early intervention.
• Traditional plaster casting methods for palatal impressions are difficult and uncomfortable for newborns.
• Existing digital solutions often require expensive equipment, limiting accessibility.

Purpose of the Study:

• To present an innovative, accessible digital workflow for capturing palatal impressions in infants with CLP.
• To automate the design of presurgical orthopedic plates using a machine learning tool.
• To enhance patient comfort and streamline presurgical treatment for CLP.

Main Methods:

• Utilizing an open-source smartphone application for high-resolution 3D palatal scanning at the point of care.
• Implementing a custom machine learning tool to process 3D models and design presurgical plates with nasal stents.
• Developing a protocol for infant positioning, scanning technique, and data exportation.

Main Results:

• The digital workflow successfully captures accurate 3D palatal scans suitable for presurgical plate generation.
• The machine learning tool automates the design process, reducing manual intervention.
• The method is demonstrated to be simple, portable, cost-effective, and enhances patient comfort.

Conclusions:

• This digital workflow offers a reproducible and accessible alternative to traditional methods for presurgical plate fabrication in CLP care.
• The approach is particularly beneficial for resource-limited settings, modernizing early cleft treatment.
• The protocol enhances efficiency and patient safety, setting a new standard in pediatric craniofacial interventions.