Intracranial aneurysms (ICAs) are an increasingly common finding, both from incidental discovery on imaging studies and on autopsy;it is estimated that anywhere from 1-6% of the American population will develop this problem. Unfortunately, while our ability to detect ICAs has grown, our fundamental understanding of this disease entity remains lacking and significant debate continues in regards to its treatment. Given the high degree of mortality and comorbidity associated with ruptured intracranial aneurysms, it is imperative that new insights and approaches be developed to inform medical decision making involving ICAs. Thus, the objective of this proposal is the creation of an informatics infrastructure to help elucidate the genesis, progression, and treatment of intracranial aneurysms. Building from our efforts from the previous R01, a set of technical developments is outlined to transform the array of information routinely collected from clinical as- sessment of ICA patients into a Bayesian belief network (BBN) that models the disease. First, we evolve the concept of a phenomenon-centric data model (PCDM) as the basis for (temporally) organizing clinically-derived observations, enabling the model to be associated with processing pipelines that can identify and transform targeted variables from the content of clinical data sources. Through these pipelines, specific values in free- text reports (radiology, surgery, pathology, discharge summaries) and imaging studies will be automatically extracted into a scientific-quality database. Second, the PCDM schema for ICAs is mapped to a Bayesian belief network: the linkage between the PCDM and BBN allows automatic updating of the network and its progressive refinement from a growing dataset. The BBN's topology will be determined by clinical experts and conditional probabilities computed from the extracted clinical data. A basic graphical user interface (GUI) will permit users to interact with the BBN, aiding in medical decision making tasks. The GUI will allow a clinician to pose questions from either a set of common clinical queries or to create new queries: loading a patient's medical record into this application will automatically populate BBN variables with extracted information (i.e., from the pipelines). Each technical component of this proposal will be evaluated in a laboratory setting. In addition, the BBN will be tested for its predictive capabilities and compared to other statistical models to assess its potential in guiding ICA treatment. This proposal leverages a clinical collaboration with the UCLA Division of Interventional Neuroradiology, a leader in ICA research and treatment. A combined dataset of 2,000 retrospective and prospective subjects will be used to create the ICA database and BBN. Data collection will encompass a comprehensive set of variables including clinical presentation, imaging assessment (morphology, hemodynamics), histopathology, gene expression, treatment, and outcomes. We will additionally leverage the NIH/NINDS Human Genetic DNA and Cell Line Repository for additional ICA-related data.
Intracranial aneurysms (ICAs) are an increasingly common finding on routine computed tomography (CT) and magnetic resonance (MR) neuro-imaging studies. The associated mortality rate and comorbidity resultant from ruptured ICAs are extreme: subarachnoid hemorrhage causes 50% of individuals to die within one month of rupture, and more than one third of survivors develop major neurological deficits. Hence, the focus of this re- search is the creation of a comprehensive research database for ICA patients, using the spectrum of data routinely acquired in the diagnosis and treatment of the problem;from this database, a new probabilistic model of ICAs will be created, providing new insights into the disease and its optimal treatment for a given individual.
|Shen, Shiwen; Han, Simon X; Petousis, Panayiotis et al. (2017) A Bayesian model for estimating multi-state disease progression. Comput Biol Med 81:111-120|
|Petousis, Panayiotis; Han, Simon X; Aberle, Denise et al. (2016) Prediction of lung cancer incidence on the low-dose computed tomography arm of the National Lung Screening Trial: A dynamic Bayesian network. Artif Intell Med 72:42-55|
|Nikkola, Elina; Laiwalla, Azim; Ko, Arthur et al. (2015) Remote Ischemic Conditioning Alters Methylation and Expression of Cell Cycle Genes in Aneurysmal Subarachnoid Hemorrhage. Stroke 46:2445-51|
|Hsu, William; Gonzalez, Nestor R; Chien, Aichi et al. (2015) An integrated, ontology-driven approach to constructing observational databases for research. J Biomed Inform 55:132-42|
|Bui, Alex A T; Hsu, William; Arnold, Corey et al. (2013) Imaging-based observational databases for clinical problem solving: the role of informatics. J Am Med Inform Assoc 20:1053-8|
|Hsu, William; Speier, William; Taira, Ricky K (2012) Automated extraction of reported statistical analyses: towards a logical representation of clinical trial literature. AMIA Annu Symp Proc 2012:350-9|
|Hsu, William; Taira, Ricky K; El-Saden, Suzie et al. (2012) Context-based electronic health record: toward patient specific healthcare. IEEE Trans Inf Technol Biomed 16:228-34|
|Wu, Juan Anna; Hsu, William; Bui, Alex At (2012) An Approach for Incorporating Context in Building Probabilistic Predictive Models. Proc IEEE Int Conf Healthc Inform Imaging Syst Biol 2012:96-105|
|Hsu, William; Taira, Ricky K; Viñuela, Fernando et al. (2011) A Case-based Retrieval System using Natural Language Processing and Population-based Visualization. Proc IEEE Int Conf Healthc Inform Imaging Syst Biol 2011:221-228|
|Watt, Emily; Sayre, James W; Bui, Alex A T (2011) Applying an Instance-specific Model to Longitudinal Clinical Data for Prediction. Proc IEEE Int Conf Healthc Inform Imaging Syst Biol 2011:81-88|
Showing the most recent 10 out of 17 publications