Publications

1. Cao  C. (2024). Integration of ten years of daily weather, traffic, and air pollution data across the six largest Norwegian cities." Scientific Data – Nature Portfolio

2. Cao  C. (2024). How to better predict the effect of urban traffic and weather on air pollution? Norwegian evidence from machine learning approaches. Journal of Economic Behavior & Organization, 221, 544-569.

3. Cao C. (2024). Rich Data Variables: Integration of Ten Years of Daily Weather, Traffic, and Air Pollution Data Across Six Largest Norwegian Cities. V3. Science Data Bank. https://doi.org/10.57760/sciencedb.17721.

4. Cao C, Dyrstad JM, Green CP.(2025) Modeling impacts of traffic, air pollution, and weather conditions on cardiopulmonary disease mortality. Scandinavian Journal of Public Health. 2025;0(0). doi:10.1177/14034948241290852

5. Cao, C., Debnath, R., & Alvarez, R. M. (2025). Physics-based machine learning for predicting urban air pollution using decadal time series data. Environmental Research Communications, 7(5), 051009.

6. Cao, C., Debnath, R., & Alvarez, R. M. (2025). Interaction between Climate Factors and Air Quality in Three Norwegian Cities: A machine learning analysis. Atmospheric Environment: X, 100366.

Revise and resubmitted

1. Zhang CC, Cao C, Zhang PY, Alvarez RM, Debnath R, “Measuring the Interaction of Sociodemographic and Building Characteristics on Energy Efficiency and Environmental Impact in London Using Machine Learning" [VIEW] Energy Policy

2. Cao C. Challenges and Recent Advances in Feedback Modalities for Teleoperated Robot-Assisted Surgery, npj Robotics 

Submitted and Under Review

1. Cao C, Hu R, Breen EC, Shih R, Sehl ME, Palella F, Mimiaga M, Martinson J, Brown T, Weiser SD,  Breen EC, Alvarez RM, Jamieson BD, Ramirez CM, "Using Causal Machine Learning to Explore the Role of Climate Change and Air Pollution in the Aging of People with HIV." Full paper available upon request.
   We analyze data from men with and without HIV to assess how environmental factors contribute to aging. Our study employs causal machine-learning techniques, such as random forest and K-means clustering, to investigate the impact of air quality and weather factors on aging. The results indicate a significant impact of health on aging, with air quality and smoking having a minor effect.

2. Cao C, Hu R, Breen EC, Shih R, Sehl ME, Palella F, Mimiaga M, Martinson J, Brown T, Weiser SD,  Breen EC, Alvarez RM, Jamieson BD, Ramirez CM, "Machine Learning to Identify Key Factors in Aging and Health Inequalities for People with HIV." Full paper available upon request.
   This study examines the effects of climate change and air pollution on aging, particularly in people living with HIV (PLWH). We use machine learning techniques to analyze existing data on aging rates and correlate these with historical air pollution and weather data. Our goal is to provide insights that inform climate and public health policies aimed at reducing the impact of climate change on aging in PLWH.

3. Cao C, R Debnath, Alvarez RM, “Contribution of PM2.5 and Climate to the Expansion of the West Nile Virus in the US." Full paper available upon request.
   This study investigates the relationship between West Nile virus (WNV) outbreaks and air quality indicators (PM₂.₅ and AQI) in California from 2016 to 2022. We employ Generalized Additive Models (GAM) and Causal Machine Learning (CML) to find a marginally significant link between PM₂.₅ levels and neuroinvasive WNV cases. Additionally, we use Poisson Event Weighted Moving Average (PEWMA) and Seasonal Autoregressive Integrated Moving Average (SARIMA) for forecasting, revealing seasonal trends in WNV outbreaks during warmer months. These findings underscore the importance of air quality in managing WNV transmission.

4. Cao C, McGraw K, Chervu NL, Delong MR, Alvarez RM, Ramirez CM, "Uncovering Plastic Surgery Risk Factors: An Exploration of Fuzzy Forest and NSQIP." Full paper available upon request.
   This study explores the key risk factors for plastic surgery using the fuzzy forest algorithm, which is suitable for complex datasets with limited samples. We analyze NSQIP data from 2016 to 2018 and identify significant correlations in various health indicators, including renal and liver function, BMI, and anesthesia factors. Our results highlight critical factors for predicting complications and readmissions, offering valuable insights for improving risk prediction models in plastic surgery.

5. Cao C, McGraw K, Chervu NL, Delong MR, Alvarez RM, Ramirez CM, "Predicting Plastic Surgery Risks: An Explainable Machine learning Algorithm Using NSQIP Data."Full paper available upon request.
   This paper introduces a new tree-based machine learning algorithm to enhance the prediction of complications in plastic surgery using data from the American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP). We compare the performance of various models, including Random Forest (RF), Fuzzy Forest (FF), and a novel model called Fuzzy Forest GLM, to existing methods like the ACS-NSQIP surgical risk calculator. Our results demonstrate superior predictive accuracy of the machine learning models, particularly the Fuzzy Forest Group (FFG), in forecasting plastic surgery outcomes.

6. Cao C, Alvarez RM, R Shilpa. "Wintertime Blues: Impacts of Air Pollution on Cardiorespiratory Medical Expenditures in Oslo." Under review. Full paper available upon request.

7. Cao C, Green CP, "Off-Premises Alcohol Availability and Traffic Accidents: Evidence from the Extension of the Norwegian Wine Monopoly" [VIEW]

Working in progress

                    Physics-Informed Disturbance Learning: Integrating PINNs with Interference Networks for Modeling Unknown Physical Processes