Dental non-linear image registration and collection method with 3D reconstruction and change detection

Mark Rahmes; Dean Fagan; George Lemieux

doi:10.1117/12.2254817

9 March 2017 Dental non-linear image registration and collection method with 3D reconstruction and change detection

Mark Rahmes, Dean Fagan, George Lemieux

Proceedings Volume 10132, Medical Imaging 2017: Physics of Medical Imaging; 101325W (2017) https://doi.org/10.1117/12.2254817
Event: SPIE Medical Imaging, 2017, Orlando, Florida, United States

Abstract

The capability of a software algorithm to automatically align same-patient dental bitewing and panoramic x-rays over time is complicated by differences in collection perspectives. We successfully used image correlation with an affine transform for each pixel to discover common image borders, followed by a non-linear homography perspective adjustment to closely align the images. However, significant improvements in image registration could be realized if images were collected from the same perspective, thus facilitating change analysis. The perspective differences due to current dental image collection devices are so significant that straightforward change analysis is not possible. To address this, a new custom dental tray could be used to provide the standard reference needed for consistent positioning of a patient’s mouth. Similar to sports mouth guards, the dental tray could be fabricated in standard sizes from plastic and use integrated electronics that have been miniaturized. In addition, the x-ray source needs to be consistently positioned in order to collect images with similar angles and scales. Solving this pose correction is similar to solving for collection angle in aerial imagery for change detection. A standard collection system would provide a method for consistent source positioning using real-time sensor position feedback from a digital x-ray image reference. Automated, robotic sensor positioning could replace manual adjustments. Given an image set from a standard collection, a disparity map between images can be created using parallax from overlapping viewpoints to enable change detection. This perspective data can be rectified and used to create a three-dimensional dental model reconstruction.

Citation Download Citation

Mark Rahmes, Dean Fagan, and George Lemieux "Dental non-linear image registration and collection method with 3D reconstruction and change detection", Proc. SPIE 10132, Medical Imaging 2017: Physics of Medical Imaging, 101325W (9 March 2017); https://doi.org/10.1117/12.2254817

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available