Deep Learning in Image Processing: Part 2—Image Enhancement, Reconstruction and Registration

Research output: Contribution to book/anthology/report/proceedingBook chapterResearch

Abstract

The digital dental workflow involves the use of several types of images, including two-dimensional (radiographs, photographs), 3D tomographic (computed tomography, cone-beam computed tomography, magnetic resonance imaging, ultrasound), and 3D mesh-type (intra-oral scans, facial scans) data. While the combined use of these various imaging modalities augments the diagnostic and treatment process, it can be challenging to make optimal use of them. Deep learning (DL) has shown tremendous potential for image processing tasks and is progressively being integrated within the clinical workflow. The previous chapter described neural networks with particular use in image processing, as well as applications in image segmentation. This chapter looks at applications of DL in image enhancement, image reconstruction and image registration. Some of the applications in this chapter are somewhat more exploratory, in the sense that higher-level diagnostic efficacy remains to be assessed. This field is expected to continue evolving in the following years; whether the optimal approach is to have DL methods operate fully independently or in combination with existing image processing techniques remains to be seen.
Original languageEnglish
Title of host publicationArtificial Intelligence in Dentistry
EditorsK. Orhan, R. Jagtap
Place of publicationCham
PublisherSpringer
Publication dateJan 2024
Pages317-351
Chapter18
ISBN (Print)978-3-031-43826-4, 978-3-031-43829-5
ISBN (Electronic)978-3-031-43827-1
DOIs
Publication statusPublished - Jan 2024

Fingerprint

Dive into the research topics of 'Deep Learning in Image Processing: Part 2—Image Enhancement, Reconstruction and Registration'. Together they form a unique fingerprint.

Cite this