• Select the area you would like to search.
  • ACTIVE INVESTIGATIONS Search for current projects using the investigator's name, institution, or keywords.
  • EXPERTS KNOWLEDGE BASE Enter keywords to search a list of questions and answers received and processed by the ADNI team.
  • ADNI PDFS Search any ADNI publication pdf by author, keyword, or PMID. Use an asterisk only to view all pdfs.
Principal Investigator  
Principal Investigator's Name: Anton Orlichenko
Institution: Tulane University
Department: Biomedical Engineering
Proposed Analysis: We seek to perform analysis on the effect of confounders with regards to diagnosis or phenotype prediction. In particular, we have recently identified fMRI endophenotypes to be highly predictive for race as well as sex in two datasets (PNC and BSNIP). We want to examine the effect of these confounders in other datasets, given that some datasets are somewhat heterogeneous (the above two mentioned) and some are highly homogenous (e.g. UKBB). We believe this bias may effect results of prediction/diagnosis, leading to suboptimal treatment, and want to design mitigating factors. https://www.techrxiv.org/articles/preprint/ImageNomer_developing_an_fMRI_and_omics_visualization_tool_to_detect_racial_bias_in_functional_connectivity/21992006 https://arxiv.org/abs/2302.00767 Our lab: https://www2.tulane.edu/~wyp/ Multiscale Bioimaging and Bioinformatics Laboratory In addition, we have plans to work on latent similarity/distance analysis of fMRI endophenotype. This is a graphical model that learns a metric based on response variables of interest. For future work, we want to combine with dictionary learning to efficiently summarize the relationship between the brain spaces as measured by fMRI. We believe this will facilitate clinical use by giving easy to understand explanations for model predictions (in terms of "this patient is very much like these other ones, because of these key summary features derived from endophenotypes"). A. Orlichenko et al., "Latent Similarity Identifies Important Functional Connections for Phenotype Prediction," in IEEE Transactions on Biomedical Engineering, doi: 10.1109/TBME.2022.3232964. If you have any concerns or questions, please reach out to me at the email address provided. Thank you!
Additional Investigators