Haihong Hu

Haihong Hu

Auto-generated profile, based on publicly available information

Haihong Hu

Mid-Atlantic Permanente Research Institute, Kaiser Permanente Mid-Atlantic States, Washington, DC

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
13 September 2024 in Cancer Epidemiology, Biomarkers & Prevention
Verified
Journal Article
Closed access
PDF file
KLKamaria L. Lee
Kamaria L. Lee
VSVarada Sarovar
Varada Sarovar
JLJennifer O. Lam
Jennifer O. Lam
WLWendy A. Leyden
Wendy A. Leyden
SAStacey Alexeeff
Stacey Alexeeff
MSMichael J. Silverberg
Michael J. Silverberg
Background: With extended lifespans for people with human immunodeficiency virus (PWH), there is a corresponding increased burden of chronic illnesses, including cancer. Our objective was to estimate the excess mortality among PWH with cancer compared with people without HIV (PWoH), accounting for the higher background mortality in the general PWH population. Methods: We identified 39,000 PWH and 387,767 demographically matched PWoH in three integrated healthcare systems from 2000 to 2016. We estimated excess mortality among PWH with cancer, computed as the cancer mortality rate difference-in-difference comparing PWH and PWoH. We evaluated five cancer groups: any cancer; virus-, human papillomavirus–, and Epstein–Barr virus –related cancers; virus-unrelated cancers, and common individual cancers. We fitted a multivariable additive Poisson model to estimate excess mortality among PWH with cancer. Results: PWH with any cancer had excess mortality compared with PWoH [41.3/1,000 person-years (py), 95% confidence interval (CI), 34.0–48.7]. The highest excess mortality was observed for Epstein–Barr virus–related cancers (63.2/1,000 py, 95% CI, 47.8–78.7), lung cancer (147.7/1,000 py, 95% CI, 41.1–254.3), and non–Hodgkin lymphoma (70.5/1,000 py, 95% CI, 51.4–89.6). Excess mortality among PWH was attenuated from 2009 to 2016, and PWH with cancer had no excess mortality 5 years after diagnosis. Conclusion: PWH in care may have excess mortality from certain cancer types, although disparities may have attenuated over time and do not persist beyond 5 years after diagnosis. Impact: Findings may guide improved clinical practice and suggest further research is needed to investigate whether cancer treatment or other factors contribute to mortality disparities for PWH with cancer.

10 September 2020 in arXiv
Verified
Preprint
Open access
PDF file
Wei ChenWei Chen
Wei Chen
Chuang NiuChuang Niu
Chuang Niu
Yiping TangYiping Tang
Yiping Tang
YWYue Wang
Yue Wang
HHHaihong Hu
Haihong Hu
JLJimin Liang
Jimin Liang
Neural architecture search (NAS) is a promising method for automatically design neural architectures. NAS adopts a search strategy to explore the predefined search space to find outstanding performance architecture with the minimum searching costs. Bayesian optimization and evolutionary algorithms are two commonly used search strategies, but they suffer from computationally expensive, challenge to implement or inefficient exploration ability. In this paper, we propose a neural predictor guided evolutionary algorithm to enhance the exploration ability of EA for NAS (NPENAS) and design two kinds of neural predictors. The first predictor is defined from Bayesian optimization and we propose a graph-based uncertainty estimation network as a surrogate model that is easy to implement and computationally efficient. The second predictor is a graph-based neural network that directly outputs the performance prediction of the input neural architecture. The NPENAS using the two neural predictors are denoted as NPENAS-BO and NPENAS-NP respectively. In addition, we introduce a new random architecture sampling method to overcome the drawbacks of the existing sampling method. Extensive experiments demonstrate the superiority of NPENAS. Quantitative results on three NAS search spaces indicate that both NPENAS-BO and NPENAS-NP outperform most existing NAS algorithms, with NPENAS-BO achieving state-of-the-art performance on NASBench-201 and NPENAS-NP on NASBench-101 and DARTS, respectively.

2 December 2024
Verified
Other
Open access
KLKamaria L. Lee
Kamaria L. Lee
VSVarada Sarovar
Varada Sarovar
JLJennifer O. Lam
Jennifer O. Lam
WLWendy A. Leyden
Wendy A. Leyden
SAStacey Alexeeff
Stacey Alexeeff
MSMichael J. Silverberg
Michael J. Silverberg
Unadjusted and adjusted mortality rate differences and excess mortality rate for cancer groups, stratified by sex and race/ethnicity

2 December 2024
Verified
Other
Open access
KLKamaria L. Lee
Kamaria L. Lee
VSVarada Sarovar
Varada Sarovar
JLJennifer O. Lam
Jennifer O. Lam
WLWendy A. Leyden
Wendy A. Leyden
SAStacey Alexeeff
Stacey Alexeeff
MSMichael J. Silverberg
Michael J. Silverberg
Number of prevalent and incident cancers by HIV status

2 December 2024
Verified
Other
Open access
KLKamaria L. Lee
Kamaria L. Lee
VSVarada Sarovar
Varada Sarovar
JLJennifer O. Lam
Jennifer O. Lam
WLWendy A. Leyden
Wendy A. Leyden
SAStacey Alexeeff
Stacey Alexeeff
MSMichael J. Silverberg
Michael J. Silverberg
Unadjusted mortality rate differences and excess mortality rate for common individual cancers, stratified by time from cancer diagnosis

31 October 2020 in arXiv
Verified
Preprint
Open access
PDF file
CWChen Wei
Chen Wei
Yiping TangYiping Tang
Yiping Tang
Chuang NiuChuang Niu
Chuang Niu
HHHaihong Hu
Haihong Hu
YWYue Wang
Yue Wang
JLJimin Liang
Jimin Liang
Recently proposed neural architecture search (NAS) algorithms adopt neural predictors to accelerate the architecture search. The capability of neural predictors to accurately predict the performance metrics of neural architecture is critical to NAS, and the acquisition of training datasets for neural predictors is time-consuming. How to obtain a neural predictor with high prediction accuracy using a small amount of training data is a central problem to neural predictor-based NAS. Here, we firstly design a new architecture encoding scheme that overcomes the drawbacks of existing vector-based architecture encoding schemes to calculate the graph edit distance of neural architectures. To enhance the predictive performance of neural predictors, we devise two self-supervised learning methods from different perspectives to pre-train the architecture embedding part of neural predictors to generate a meaningful representation of neural architectures. The first one is to train a carefully designed two branch graph neural network model to predict the graph edit distance of two input neural architectures. The second method is inspired by the prevalently contrastive learning, and we present a new contrastive learning algorithm that utilizes a central feature vector as a proxy to contrast positive pairs against negative pairs. Experimental results illustrate that the pre-trained neural predictors can achieve comparable or superior performance compared with their supervised counterparts with several times less training samples. We achieve state-of-the-art performance on the NASBench-101 and NASBench201 benchmarks when integrating the pre-trained neural predictors with an evolutionary NAS algorithm.