Generative AI for osteoarthritis imaging: CycleGAN-based synthesis and U-net3D segmentation

Osteoarthritis (OA) is one of the leading diseases in aging people. However, we still don’t know what causes the disease and the efficient treatment for it. In this thesis, we deploy a generative artificial intelligence, which has become a tool and a new trend in AI. We design generative AI models for medical imaging, particularly for lowering data imbalance and increasing diagnostic accuracy. This thesis investigates the application of CycleGAN and Unet3D in two medical imaging tasks: hand osteoarthritis (HOA) and knee effusion. The experiment results demonstrate promising findings in improving the accuracy of image classification and segmentation tasks.

Author dc:creator

• Cao, Zhen

Advisor dc:contributor.advisor

• Zhang, Ming

• Computer science
• CycleGAN
• Generative artificial intelligence
• Medical imaging
• Osteoarthritis
• Unet3D

Statement dc:rights

• Attribution 4.0 International

Licence dc:rights.uri

• http://creativecommons.org/licenses/by/4.0/

Language dc:language.iso

en_US

Handle dc:identifier.uri

https://hdl.handle.net/2144/53162

OAI identifier oai:identifier

oai:open.bu.edu:2144/53162