Haoyu Lu

Haoyu Lu

Too late to quit,
too early to retire,
too young to die,
but too old to find a sugah daddee.

Publications

Multi-Annual Inventorying of Retrogressive Thaw Slumps Using Domain Adaptation

Published in Journal of Geophysical Research: Machine Learning and Computation, 2025

To enhance the model’s generalization ability, here we implemented and compared three domain adaptation methods, i.e., the classic supervised fine-tuning method and two proposed unsupervised methods: Image StyleTransfer Domain Adaptation (ISTDA) and the Tversky Adversarial Domain Adaptation (TADA) network. In our proposed ISTDA, we uniformed the contextual information of multi-temporal images by Cycle Generative Adversarial Network (CycleGAN). We introduced the Tversky loss and the automatic adjustment of weights for multiple loss functions to suppress false positives and to improve the generalization of TADA. Read more

Recommended citation: Lin, Y., Hu, X., Lu, H., Niu, F., Liu, G., Huang, L., et al. (2025). Multi-annual inventorying of retrogressive thaw slumps using domain adaptation. Journal of Geophysical Research: Machine Learning and Computation, 2, e2024JH000370. https://doi.org/10.1029/2024JH000370 https://doi.org/10.1029/2024JH000370

QUANTOF: multi-GPU-based list-mode fully quantitative TOF PET image reconstruction.

Published in 2024 IEEE Nuclear Science Symposium (NSS), Medical Imaging Conference (MIC) and Room Temperature Semiconductor Detector Conference (RTSD), 2024

This work introduces a multi-GPU-based list-mode fully quantitative Time-of-Flight (TOF) PET image reconstruction technique, named QUANTOF, that integrates all corrections during PET image iterative reconstruction on GPUs. Additionally, optimizations have been applied to the backprojection algorithm, resulting in a two-fold increase in speed compared to the pre-optimized version. Validation using experimental datasets from a Siemens Biograph Vision PET/CT scanner demonstrates that this technique significantly reduces reconstruction time costs while achieving exceptional quantitative accuracy. Read more

Recommended citation: Yuan, Z., Zhan, F., Lu, H., Li, C., Hou, Y., Wang, H., ... & Jiang, J. (2024, October). QUANTOF: multi-GPU-based list-mode fully quantitative TOF PET image reconstruction. In 2024 IEEE Nuclear Science Symposium (NSS), Medical Imaging Conference (MIC) and Room Temperature Semiconductor Detector Conference (RTSD) (pp. 1-1). IEEE. https://ieeexplore.ieee.org/abstract/document/10657018

A virtual-pinhole PET device for improving contrast recovery and enhancing lesion detectability of a one-meter-long PET scanner: a simulation study.

Published in Physics in Medicine & Biology, Volume 68, Number 14, 2023

This paper presents a simulation study to demonstrate that the contrast recovery coefficients (CRC) and detectability of small lesions of a one-meter-long positron emission tomography (PET) scanner can be further enhanced by the integration of high resolution virtual-pinhole (VP) PET devices. Read more

Recommended citation: Jiang, J., Hua, J., Wang, H., Yuan, Z., Meng, Y., Lu, H., ... & Tai, Y. C. (2023). "A virtual-pinhole PET device for improving contrast recovery and enhancing lesion detectability of a one-meter-long PET scanner: a simulation study." Physics in Medicine and Biology.. 68(14). https://iopscience.iop.org/article/10.1088/1361-6560/acdfaf