Related Experiment Video
Updated: Jun 29, 2025
11:49
Cereal Crop Ear Counting in Field Conditions Using Zenithal RGB Images
Published on: February 2, 2019
9.3K
A Novel Ear Impression-Taking Method Using Structured Light Imaging and Machine Learning: A Pilot Proof of Concept
Kenneth Wei De Chua1, Hazel Kai Hui Yeo1, Charmaine Kai Ling Tan2
1Department of Otorhinolaryngology-Head and Neck Surgery, Allied Health, Audiology, Changi General Hospital, Singapore 529889, Singapore.
Journal of Clinical Medicine
|April 9, 2024
Summary
A new contactless ear scanning prototype shows promise for predicting ear canal information using structured light imaging and a deep neural network, potentially automating ear impression-taking for same-day 3D printing.
Area of Science:
- Biomedical Engineering
- Medical Imaging
- Artificial Intelligence
Background:
- Traditional ear impression-taking is invasive.
- Contactless ear scanning methods are underdeveloped.
- A correlation between external ear features and ear canal anatomy is proposed.
Purpose of the Study:
- Develop a prototype for contactless ear scanning.
- Create an algorithm to predict ear canal information.
- Establish proof of concept for automated ear impression-taking.
Main Methods:
- Structured light imaging prototype developed.
- Correlation analysis using existing ear impression data.
- Deep neural network for predictive algorithm creation.
- Comparative study with traditional ear impressions.
Main Results:
- Prototype successfully trialed; participants reported comfort.
- Partial ear canal matching achieved from external images.
- Predictive algorithm demonstrated a good standard of error.
- Proof of concept for the novel method established.
Conclusions:
- Further research needed to enhance algorithm predictive capabilities.
- Optimization of prototype imaging positions is required.
- Potential for automated ear impression-taking and same-day 3D printing of earmolds.
