Prediction of Brain Biological Age and Acceleration of Aging in AD Using Large Datasets of Existing Neuroimaging Data and Deep Learning Approaches
This proposed research seeks to first predict physiological brain age (PBA) for individuals in healthy condition by leverage deep learning-based modeling with large size brain MRI datasets. The preliminary data showed that the current model is very accurate in this task with the mean absolute error (MAE) of 3 years. (i.e., the predicted age and the chronological age of a healthy subject are not different by more than 3 years). The model trained on healthy population will be applied to the cohorts of mild cognitive impairment (MCI) and AD in order to access the acceleration of aging, meaning that the model would show increased age in these cohorts by capturing the faster brain appearance changes on MRI due to MCI or AD. Our proposed deep-learning approach will be able to predict the voxel-wise PBA while mapping it on regional brain anatomy, which will allow the region specific aging pattern in MCI or AD (e.g., increased age of the hippocampus relative to other brain structures) and subsequently classify each patient from healthy or other disease groups (MCI / AD or other dementia group) using the disease specific pattern of PBA map. Finally, we will apply an event-based model (EBM) to the PBA map generated by the proposed deep learning algorithm. This will generate the sequence in which regional PBA(s) become abnormal regarding MCI or AD. In other words, starting from a specific region (e.g., hippocampus) to be abnormal at the earliest stage, one can observe how this brain abnormality propagates to other regions (e.g., prefrontal cortex) incrementally over different stages. The patients belonging to one stage are expected to probabilistically differ in clinical characteristics from those at other stages. New patients will be assigned to one of the stages in EBM based on the similarity of the PBA pattern and this process combined with a mathematical Hazard model will estimate the ‘survival’ probability explaining the remaining days in healthy status prior to the onset of MCI or AD, ultimately allowing disease-specific risk scoring as a clinical tool to be used in routine patient care.
The goal of this project is to predict individual physiological brain age (PBA), taking into account the complex nature of the human aging trajectory, we propose to leverage deep learning-based PBA modeling based upon multicontrast MRI databases of healthy and age-related disease populations.
There are two specific aims:
1. Develop multimodal MRI-based deep-learning methods to predict regional brain age in healthy subjects. To better understand the meaning of high PBA offsets, we will assess the multivariate association of the offset with putative risk variables including cognitive decline, health risk factors (biometrics), genetic influence, predisposed lifestyle and health factors (alcohol use, hypertension, diabetes, etc.).
2. Classification, staging and risk-scoring of dementias based on PBA patterns. Brain regional PBA pattern at a given age combined with putative aging variables is hypothesized to be unique for each of various age-related brain diseases and can be used as a “fingerprint”. Classification based on the voxel-wise PBA pattern together with putative aging variables will diagnose an individual into healthy, MCI, or AD.
About the Researcher
My research focuses on developing an analytic platform that assesses aging of brain structures and their structural and functional networks as well as predicting the eventual long-term outcome for neurodevelopment and quantifying the progression of neurodegeneration. To follow-up long-term brain structural modification associated with neurodevelopmental / neurodegenerative disorders, my group develops methods to quantify various aspects of brain anatomical and networking variability using longitudinally collected multi-contrast MRI. My technical expertise on surface-based morphology and texture modeling, network topology analysis, and multivariate statistical modeling consists of essential elements to develop a combination of techniques to accomplish the proposed specific aims. I have also applied various analytic frameworks, including cortical morphometry, voxel-based morphometry, deformation-based morphometry and structural / functional network analyses, to assessment of brain structures in healthy conditions as well as in pathological conditions that often present anatomical variations beyond the range of normal structures. Using a more advanced pattern analysis with machine learning and deep learning on innovative multicontrast MRI features, my current research seeks to understand the atypical structural and network alterations in various neurological diseases including epilepsy, dementia, and preterm birth and ultimately to predict neurological / brain functional outcome in the patients. In particular, the use of the deep-learning-based neural network (DNN) can reconstruct a smaller size of the high-order feature-set to enhance the performance in prediction as well as brain segmentation and artifact correction.
My ultimate research goal is to become a more established independent investigator in the field of brain image analysis science with a particular focus on neurodevelopment, neurodegeneration and disease progression in relation to the health of patients. This requires a broad range of expertise in computer science, including pattern recognition techniques, as well as thorough knowledge in neuroscience. To pursue an independent research career, I joined USC in Nov. 2016, and I am now entering a critical stage of my career. To build upon recent successful work, it is important to continue in a multi-disciplinary, intellectually rich environment with mentors and collaborators who possess a broad understanding of the various facets of my research, which ranges from engineering to physiology and medicine. The BrightFocus Award in Alzheimer’s Disease will provide a unique, unparalleled opportunity to further enhance and broaden my research skills, from development of brain image processing and machine learning techniques and translation of newly developed approaches to clinical studies of healthy aging and neurodegenerative disorders. My initial work on this project (~6 months) has already shown promising preliminary results (Fig 1-2) and demonstrated the capability to analyze this large dataset and interpret the results (e.g., the skillsets and resources provided b mentors and collaborators). Performing the proposed project under this award will build on this existing foundation to provide me with the complete range of skills and knowledge necessary to successfully analyze large neuroimaging datasets to address critical neurodegeneration questions as an independent investigator. As I have only recently begun researching brain aging, my immediate goals are to expand my knowledge of physiology, pathophysiology and genetics underlying brain aging and neurodegeneration, as well as learn how to develop relevant analytic framework, including advanced statistics and machine learning techniques, in order to formulate sound hypotheses and study designs for developing clinically useful early diagnostic tools and interpreting final results. Ultimately, the knowledge and experience gained through this project will be the key to expansion of my current research for more publications and R01 funding, thereby achieving my career goal of becoming a more established independent investigator.
Finally, I would love to indicate my sincere appreciation to the generous donors to the BrightFocus Foundation. Their donations and contributions to the funding and my project is critical to the completion of the current proposal and the continuation of my research career in aging study. Without them, I would not be able to gain the resources such as an experienced postdoctoral fellow, a performance computer and cost for trips to conferences where we exhibit our works and learn from other researchers, which play a pivotal role in implementing the research.
First published on: June 18, 2019
Last modified on: December 23, 2019