Instructor in Medicine; Investigator
Harvard Medical School, Department of Radiology
Center for Clinical Investigation, Brigham and Women's Hospital
Contact
Office Phone: 617-732-6467
Medical Research Building 208C
75 Francis Street, Boston MA 02115
Email: zhou.joe.lan@gmail.com, zlan@bwh.harvard.edu
Overview
I am a computational and statistical investigator at Brigham and Women’s Hospital and Harvard Medical School, with strong interests in biomedical data, particularly brain imaging, genetics, and cardiovascular data. My work can be viewed in two major components.
The first component focuses on building end-to-end pipelines that transform raw, large-scale datasets into reliable resources for clinical discovery. Leveraging these resources, I develop statistical and machine learning methodologies to address challenges such as high dimensionality, spatial correlation, and complex data structures. Reproducibility and accessibility are central to my work: each paper is accompanied by well-documented Python/R/C/C++ software (available at https://github.com/lanzhouBWH), enabling others to apply and extend my methods.
The second component centers on providing statistical contributions—including survival analysis, causal inference, and predictive modeling—for studies that utilize diverse biomedical data sources such as electronic health records, clinical trial data, and population-based cohorts. This involves design of experiments, hands-on analysis, statistical plan drafting, and manuscript/grant preparation.
My work has appeared in leading statistical journals (Technometrics, Biometrics, Bayesian Analysis, Journal of the Royal Statistical Society: Series A, Data Science in Science), as well as in top neuroimaging, genetics, bioinformatics, and medical journals (Imaging Neuroscience, NeuroImage: Clinical, Bioinformatics, JAMIA Open, The New England Journal of Medicine). In addition, as a statistical consultant, I have co-authored more than 50 collaborative papers in high-impact biomedical and clinical journals, including the Journal of the American Heart Association, Circulation, and Radiology.
Currently, I am expanding my work to incorporate artificial intelligence for biomedical discovery and clinical translation.
Research
Innovative Statistical Methodologies for Diffusion MRI:
The structural connections within the brain’s white matter are essential for its function. Diffusion MRI tractography allows for the in-vivo reconstruction of white matter fiber pathways. Diffusion MRI poses significant challenges due to their high dimensionality, spatial correlation, and complex signal structures. My research focuses on developing and applying cutting-edge statistical and computational methods to unlock the full potential of such data. The motivating data includes HCP Young Adult - -Connectome, Adolescent Brain Cognitive Development (ABCD) Study, and Alzheimer's Disease Neuroimaging Initiative (ADNI). The key collaborators and mentors in this research avenue includes Lauren J. O'Donnell from Brigham and Women's Hospital at Harvard Medical School, Brian J Reich from Department of Statistics, North Carolina State University, Arkaprava Roy from Department of Biostatistics, University of Florida, and Zhengwu Zhang from Department of Statistics and Operations Research, UNC Chapel Hill.- Lan, Z., Chen, Y., Rushmore, J., Zekelman, L., Makris, N., Rathi, Y., Golby, A.J., Zhang, F. and O’Donnell, L.J., 2025. Fiber microstructure quantile (FMQ) regression: A novel statistical approach for analyzing white matter bundles from periphery to core. Imaging Neuroscience, 3, p.imag_a_00569.
- Roy, A., Lan, Z. and Zhang, Z., 2024. Nonparametric Modeling of Diffusion MRI Signal in Q-Space. Data Science in Science, 3(1), p.2412017.
- Lan, Z., Reich, B.J., Guinness, J., Bandyopadhyay, D., Ma, L. and Moeller, F.G., 2022. Geostatistical modeling of positive‐definite matrices: An application to diffusion tensor imaging. Biometrics, 78(2), pp.548-559.
- Yan, L., Zhang, X., Lan, Z., Bandyopadhyay, D., and Wu, Y., 2024. Variable Screening and Spatial Smoothing in Fréchet Regression with Application to Diffusion Tensor Imaging. Annals of Applied Statistics.
Robust Statistical Methodologies and Theoretical Frameworks for Matrix-Valued Data:
I focus on developing robust statistical methodologies and theoretical frameworks tailored to the complexities of real-world biomedical and imaging data. My work emphasizes extending theoretical foundations to accommodate challenges such as high dimensionality, spatial correlation, and intricate dependency structures, ensuring the developed methods are both robust and generalizable. The key collaborators and mentors in this research avenue includes Arkaprava Roy from Department of Biostatistics, University of Florida.- Lan, Z. and Roy, A., 2025. Spatial von-Mises Fisher Regression for Directional Data. Technometrics. (Accepted)
- Roy, A. and Lan, Z., 2024. Double soft-thresholded model for multi-group scalar on vector-valued image regression. Bayesian Analysis, 1(1), pp.1-30.
- Lan, Z., 2024. Correlated Wishart matrices classification via an expectation-maximization composite likelihood-based algorithm. Statistics and Its Interface, 17(2), 173–185.
Innovative Statistical Methodologies for Magnetic Resonance Spectroscopy (MRS):
MRS is a critical tool for exploring the neurometabolic underpinnings of brain function, providing unique insights into the biochemical composition of tissues. My research has focused on developing and applying advanced methodologies to enhance MRS data processing, analysis, and interpretation, particularly in the context of neurological disorders such as functional neurological disorders (FND). The key collaborators and mentors in this research avenue includes Alexander Lin from Center for Clinical Spectroscopy, Department of Radiology, Harvard Medical School, Brigham and Women's Hospital.- Lan, Z., Foster, S., Charney, M., van Grinsven, M., Breedlove, K., Kozlowska, K. and Lin, A., 2025. Neurometabolic network (NMetNet) for functional neurological disorder in children and adolescents. NeuroImage: Clinical, 46, p.103767..
- Beroukhim, B., McComas, S., Joyce, J.M., Schuhmacher, L.S., Koerte, I., Lan, Z. and Lin, A., 2025. A novel automated pipeline to assess MR spectroscopy quality control: Comparing current standards and manual assessment. Journal of Neuroimaging, 35(1), p.e13246.
Epidemiological/Clinical Studies:
One component of my research is epidemiological/clinical studies. This includes addressing challenges in data quality, integration, and predictive modeling to improve public health.Recent Updates
- June 24, 2025: Presented an invited talk at the International Conference on Statistics and Data Science, titled "Fiber Microstructure Quantile (FMQ) Regression: A Novel Statistical Approach for Analyzing White Matter Bundles from Periphery to Core."
- June 2, 2025: Presented a local invited talk at the Ferenc Jolesz First Monday Research Seminar Series, titled "Neurometabolic Network (NMetNet): A Novel Conceptual Brain Network for Studying Functional Neurological Disorder."
- November 7, 2024: Attended the 2024 International Biometric Conference in Atlanta, GA.
- August 7, 2024: Attended the 2024 Joint Statistical Meetings (JSM) in Portland, OR.
- September 7, 2023: Selected as a committee member for the ASA Mental Health Statistics Section (MHSS) Student Paper Competition.
- June 11–14, 2023: Organized and chaired the invited session "Mathematics and Statistics in Medical Imaging" at the 2023 ICSA Applied Statistics Symposium.
- April 19, 2023: Participated as a mentored biostatistics investigator at the Early Career Biostatistics Faculty Collaborating on Mental Health and Aging workshop in Tucson, AZ.