Related Experiment Video
Updated: Oct 18, 2025
14:56
An Experimental Paradigm for the Prediction of Post-Operative Pain PPOP
Published on: January 27, 2010
21.5K
Deep Learning-Based Prediction of Paresthesia after Third Molar Extraction: A Preliminary Study
Byung Su Kim1, Han Gyeol Yeom2, Jong Hyun Lee3
1Department of Oral and Maxillofacial Surgery, Daejeon Dental Hospital, Wonkwang University College of Dentistry, Daejeon 35233, Korea.
Diagnostics (Basel, Switzerland)
|September 28, 2021
Summary
Convolutional neural networks (CNNs) can predict inferior alveolar nerve paresthesia from panoramic X-rays before mandibular third molar extraction. This AI tool shows promise in reducing nerve damage risks associated with the procedure.
Area of Science:
- Dentistry
- Medical Imaging
- Artificial Intelligence
Background:
- Inferior alveolar nerve paresthesia is a potential complication of mandibular third molar extraction.
- Predictive tools are needed to mitigate risks associated with this common dental procedure.
Purpose of the Study:
- To evaluate the efficacy of convolutional neural networks (CNNs) in predicting inferior alveolar nerve paresthesia.
- To assess the use of panoramic radiographic images for AI-driven risk prediction before mandibular third molar extraction.
Main Methods:
- A dataset of 300 panoramic radiographic images from patients undergoing mandibular third molar extraction was utilized.
- Images were classified into two groups: those with post-extraction paresthesia (n=100) and those without (n=200).
- Deep learning models, specifically SSD300 and ResNet-18 CNNs, were trained and validated on the dataset.
Main Results:
- The CNN models achieved an average accuracy of 0.827, sensitivity of 0.84, specificity of 0.82, and area under the curve of 0.917.
- These results indicate a strong predictive capability of the CNNs for identifying patients at risk of paresthesia.
Conclusions:
- Convolutional neural networks show significant potential as an assistive tool for predicting inferior alveolar nerve paresthesia.
- Utilizing preoperative panoramic radiographic images with CNNs can aid clinicians in assessing and potentially mitigating risks before mandibular third molar extraction.
