Three-dimensional periodontal investigations using a prototype handheld ultrasound scanner with spatial positioning
Radu Chifor1, Mengxun Li2, Kim-Cuong T Nguyen3
1Department of Preventive Dentistry, University of Medicine and Pharmacy Iuliu Hatieganu, Cluj-Napoca, Romania Chifor Research SRL, Cluj-Napoca, Romania. raduchifor@yahoo.com.
Medical Ultrasonography
|March 3, 2021
Summary
This study shows that 3D ultrasound periodontal tissue reconstruction is feasible using standard 2D ultrasound and spatial positioning sensors. Deep learning algorithms enable practical clinical assessment of periodontal soft tissues.
Area of Science:
- Periodontology
- Medical Imaging
- Biomedical Engineering
Background:
- Periodontal disease assessment traditionally relies on invasive methods or less precise imaging.
- Advancements in ultrasound technology offer potential for non-invasive, detailed tissue visualization.
Purpose of the Study:
- To assess the feasibility of 3D periodontal tissue reconstruction using standard 2D ultrasound equipment.
- To evaluate the accuracy of 3D reconstructions generated from porcine mandible lateral areas.
Main Methods:
- A free-hand prototype system combining a 2D ultrasound scanner and spatial positioning sensors was utilized.
- Deep learning algorithms were employed for automated data processing of semi-automatically segmented ultrasound images.
- The system was tested on porcine mandible samples.
Main Results:
- Analysis confirmed non-parallel 2D frame acquisition during scanning.
- The accuracy of the 3D periodontal reconstructions ranged from 0.179 mm to 0.235 mm.
- The study demonstrated the diagnostic potential of the developed 3D reconstruction technique.
Conclusions:
- 3D ultrasound reconstruction using a free-hand 2D scanner with spatial positioning is feasible.
- Automated data processing with deep learning enhances the practicality for clinical periodontal soft tissue assessment.
