Smartphone Scans of Newborns' Cleft Palate Models
Author Information
Author(s): Santos José Wittor de Macêdo, Mueller Andreas Albert, Benitez Benito K., Lill Yoriko, Nalabothu Prasad, Muniz Francisco Wilker Mustafa Gomes
Primary Institution: University of Basel
Hypothesis
Smartphone apps have inferior performance compared to intraoral scanners but attain sufficient accuracy for recognition by an automated presurgical plate generator.
Conclusion
KIRI Engine performed better in scanning UCLP models without a mirror, and the ML tool showed a high capability for morphology recognition and automated PSP generation.
Supporting Evidence
- Thirty 3D scans were acquired by each app, totaling 60 scans.
- KIRI scans without a mirror had a good performance comparable to the control group.
- The ML tool was able to predict landmarks and automatically generate plates, except in ICP models.
- KIRI scans' plates showed better performance compared to controls.
Takeaway
Researchers tested smartphone apps to scan models of newborns with cleft palates and found that one app worked really well, helping to create 3D models for surgery.
Methodology
A comparative analysis of two smartphone scanning apps on 15 cleft palate models was conducted, with scans compared to a control group using a professional intraoral scanner.
Potential Biases
Operator and phone movement during image capture could introduce inaccuracies.
Limitations
The findings are restricted to an in silico environment and cannot be generalized for clinical settings.
Participant Demographics
Fifteen newborns with cleft lip and palate were included, with five patients for each type of cleft.
Statistical Information
P-Value
p=0.653
Confidence Interval
95%
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website