Back To Projects

Dental Segmentation of The Mandible Using AI

Khushi G.

Dental segmentation can help with decision-support issues in medical diagnosis, such as human identification and maxillofacial surgery, as well as orthodontic therapy, implant planning, and other issues. This leads to my research question; “How do I automate the task required to diagnose a dental problem during a dental visit?”


Dental segmentation can help with decision-support issues in medical diagnosis, such as human identification and maxillofacial surgery, as well as orthodontic therapy, implant planning, and other issues. This leads to my research question; “How do I automate the task required to diagnose a dental problem during a dental visit?” The approach to solve this was to train an AI with images pre-segmented by experts and then test it. This process is also called image segmentation. The AI had eventually gotten a test accuracy score of 0.940 and test IoU score of 0.904. The closer to one that these scores are the better. Pixel accuracy and Iou score were surprisingly similar. I think that this is a good thing because it means both seem pretty accurate. The AI has its flaws but it can segment images when given images

Explore More!

Khushi G.
Joseph Vincent
Aerospace Engineering PhD Candidate at Stanford

Related Projects

Detecting Retinal Detachment using images with Machine Learning

This research is on how AI and machine learning can be used to detect retinal detachment using images.
Yi-Chen (.
Mentored by Samar Abu Hegly
workspace_premium
Machine Learning Approaches to Detect Brain Tumors from Magnetic Resonance Imaging Scans

Our study utilized a comprehensive dataset of brain magnetic resonance imaging (MRI) scans to compare and assess the performance of different baseline AI models.
Cherry (.
Mentored by Shreya Parchure
Predicting the Presence of Autism Spectrum Disorder Based on Eye-Tracking Scan Path Images

This research explores using eye-tracking data and machine learning to predict Autism Spectrum Disorder (ASD), achieving 74.6% accuracy with logistic regression, highlighting eye-tracking's potential for early, non-invasive ASD diagnosis and its applicability to other neurological conditions.
Sara C.
Mentored by Emily Broadhurst