Xiaoxiao Sun
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Xiaoxiao Sun has a total of 15 co-authors. Here are the top co-authors by number of shared publications:
- Cynthia A. Thomson
Mel & Enid Zuckerman College of Public Health, University of Arizona, Tucson, Arizona
1 shared publication - David O. Garcia
University of Arizona
1 shared publication - Hehe Fan
Zhejiang University, Hangzhou, China
1 shared publication - Hsiao-Hui Sherry Chow
University of Arizona Cancer Center, University of Arizona, Tucson, AZ 85724, USA
1 shared publication - Jessica A. Martinez
School of Nutritional Sciences and Wellness, University of Arizona, Tucson, AZ, United States of America
1 shared publication - Jie Liang
OPPO Research Institute, Shenzhen, China
1 shared publication
Latest Publications
- Open Access
- 9 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 15069-15078
SmartMage: Dynamic Modality Orchestration for 3D Scene Understanding
- 9 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 15069-15078
Open AccessSmartMage: Dynamic Modality Orchestration for 3D Scene Understanding
- Open Access
- 1 January 2023
- Published by SAGE Publications in American Journal of Men's Health
- Vol. 17 (1)
Differences in Metabolomic Profiles by Birthplace in Mexican-Origin Hispanic Men Who Participated in a Weight Loss Lifestyle Intervention
Birthplace, as a proxy for environmental exposures (e.g., diet), may influence metabolomic profiles and influence risk of cancer. This secondary analysis investigated metabolomic profile differences between foreign and U.S.-born Mexican-origin (MO) Hispanic men to shed light on potential mechanisms through which foreign- and U.S.-born individuals experience differences in cancer risk and risk factors. Plasma samples from MO Hispanic men ( N = 42) who participated in a previous lifestyle intervention were collected pre-and post-intervention. Metabolomic profiles were characterized from samples using ultra performance liquid chromatography–quadrupole time of flight mass spectrometry (UPLC-QTOF). Models were visualized using supervised orthogonal projections to latent structures–discriminant analysis (OPLS-DA). Progenesis QI was used for peak integration and metabolite identification. Plasma metabolomic profiles differed between foreign- and U.S.-born pre-intervention (R2 = .65) and post-intervention (R2 = .62). Metabolomic profiles differed pre- versus post-intervention (R2 = .35 and R2 = .65) for the foreign- and U.S.-born group, respectively. Both endogenous metabolites and dietary components characterized differences between foreign- and U.S.-born participants pre- and post-intervention. Plasma metabolomic profiles from MO Hispanic men differed by birthplace. These results advance our understanding of relevant exposures that may affect cancer risk among MO Hispanic men born abroad or in the United States.
- 1 January 2023
- Published by SAGE Publications in American Journal of Men's Health
- Vol. 17 (1)
Open AccessDifferences in Metabolomic Profiles by Birthplace in Mexican-Origin Hispanic Men Who Participated in a Weight Loss Lifestyle Intervention
Birthplace, as a proxy for environmental exposures (e.g., diet), may influence metabolomic profiles and influence risk of cancer. This secondary analysis investigated metabolomic profile differences between foreign and U.S.-born Mexican-origin (MO) Hispanic men to shed light on potential mechanisms through which foreign- and U.S.-born individuals experience differences in cancer risk and risk factors. Plasma samples from MO Hispanic men ( N = 42) who participated in a previous lifestyle intervention were collected pre-and post-intervention. Metabolomic profiles were characterized from samples using ultra performance liquid chromatography–quadrupole time of flight mass spectrometry (UPLC-QTOF). Models were visualized using supervised orthogonal projections to latent structures–discriminant analysis (OPLS-DA). Progenesis QI was used for peak integration and metabolite identification. Plasma metabolomic profiles differed between foreign- and U.S.-born pre-intervention (R2 = .65) and post-intervention (R2 = .62). Metabolomic profiles differed pre- versus post-intervention (R2 = .35 and R2 = .65) for the foreign- and U.S.-born group, respectively. Both endogenous metabolites and dietary components characterized differences between foreign- and U.S.-born participants pre- and post-intervention. Plasma metabolomic profiles from MO Hispanic men differed by birthplace. These results advance our understanding of relevant exposures that may affect cancer risk among MO Hispanic men born abroad or in the United States.
- 12 June 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks and Learning Systems
- Vol. 31 (8) , 2832-2846
Self-Paced Balance Learning for Clinical Skin Disease Recognition
Class imbalance is a challenging problem in many classification tasks. It induces biased classification results for minority classes that contain less training samples than others. Most existing approaches aim to remedy the imbalanced number of instances among categories by resampling the majority and minority classes accordingly. However, the imbalanced level of difficulty of recognizing different categories is also crucial, especially for distinguishing samples with many classes. For example, in the task of clinical skin disease recognition, several rare diseases have a small number of training samples, but they are easy to diagnose because of their distinct visual properties. On the other hand, some common skin diseases, e.g., eczema, are hard to recognize due to the lack of special symptoms. To address this problem, we propose a self-paced balance learning (SPBL) algorithm in this paper. Specifically, we introduce a comprehensive metric termed the complexity of image category that is a combination of both sample number and recognition difficulty. First, the complexity is initialized using the model of the first pace, where the pace indicates one iteration in the self-paced learning paradigm. We then assign each class a penalty weight that is larger for more complex categories and smaller for easier ones, after which the curriculum is reconstructed by rearranging the training samples. Consequently, the model can iteratively learn discriminative representations via balancing the complexity in each pace. Experimental results on the SD-198 and SD-260 benchmark data sets demonstrate that the proposed SPBL algorithm performs favorably against the state-of-the-art methods. We also demonstrate the effectiveness of the SPBL algorithm's generalization capacity on various tasks, such as indoor scene image recognition and object classification.
- 12 June 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks and Learning Systems
- Vol. 31 (8) , 2832-2846
Self-Paced Balance Learning for Clinical Skin Disease Recognition
Class imbalance is a challenging problem in many classification tasks. It induces biased classification results for minority classes that contain less training samples than others. Most existing approaches aim to remedy the imbalanced number of instances among categories by resampling the majority and minority classes accordingly. However, the imbalanced level of difficulty of recognizing different categories is also crucial, especially for distinguishing samples with many classes. For example, in the task of clinical skin disease recognition, several rare diseases have a small number of training samples, but they are easy to diagnose because of their distinct visual properties. On the other hand, some common skin diseases, e.g., eczema, are hard to recognize due to the lack of special symptoms. To address this problem, we propose a self-paced balance learning (SPBL) algorithm in this paper. Specifically, we introduce a comprehensive metric termed the complexity of image category that is a combination of both sample number and recognition difficulty. First, the complexity is initialized using the model of the first pace, where the pace indicates one iteration in the self-paced learning paradigm. We then assign each class a penalty weight that is larger for more complex categories and smaller for easier ones, after which the curriculum is reconstructed by rearranging the training samples. Consequently, the model can iteratively learn discriminative representations via balancing the complexity in each pace. Experimental results on the SD-198 and SD-260 benchmark data sets demonstrate that the proposed SPBL algorithm performs favorably against the state-of-the-art methods. We also demonstrate the effectiveness of the SPBL algorithm's generalization capacity on various tasks, such as indoor scene image recognition and object classification.