Jianliang Lu

Title: Jianliang Lu and Medical Image Compression Innovation.

Introduction

Jianliang Lu is an inventor associated with Hong Kong, CN. His patent work includes a generative adversarial network-based approach for producing thin-cut CT images from thick-cut CT inputs, addressing a challenge in cross-sectional medical imaging.

Latest Patents

The provided information lists one patent for Lu: “Generative adversarial network-based lossless image compression model for cross-sectional imaging.” The invention describes a process that first registers continuous thin-cut CT images and combines five or more of them by maximizing pixel-wise intensity to create thick-cut images. An encoder-decoder generator then converts each thick-cut image into a low-dimensional representation and decodes it into multiple thin-cut images. A discriminator distinguishes original thin-cut images from generated ones. Through adversarial training, feedback from the discriminator helps the generator improve the quality of its synthetic images while also strengthening the discriminator’s ability to distinguish real from generated images. The stated patent count is 1 patent.

Career Highlights

The supplied career information associates Lu with Versitech, Inc. and The Education University of Hong Kong. These affiliations connect his profile with organizations involved in technology and education.

Collaborations

Lu’s listed coworkers include Man Fung Yuen and Gilbert Chiu Sing Lui. Their names are provided as professional collaborators, though specific joint projects are not detailed in the supplied information.

Conclusion

Jianliang Lu’s listed patent focuses on using adversarial machine learning to generate thin-cut CT images from thick-cut inputs, reflecting an application of computational innovation to medical imaging.