Featured Projects
The overall goal of our featured projects is to leverage artificial intelligence and data analytics to innovate in clinical and translational research and clinical and translational science while ensuring they are beneficially applied to underrepresented populations.
CANDO (Computational Analysis of Novel Drug Opportunities)
The Computational Analysis of Novel Drug Opportunities (CANDO) platform is a computational approach to make drug discovery faster and less expensive while also being safe and effective. CANDO is capable of drug discovery, repurposing of existing drugs to find novel uses, and also designing novel never-seen-before drug candidates. CANDO utilizes a holistic approach to understand the entire network of interactions between all drugs, molecules, and diseases simultaneously. Specific projects involving CANDO are to: integrate EHR data including results from clinical trials dynamically, enable precision medicine, perform simulated clinical trials, and incorporate further multiscale data and modeling to describe and exploit drug behavior. Contacts: Ram Samudrala, PhD; Zackary Falls, PhD.
Observational research/BMI Investigator
This interactive web-based tool enables users to perform automated retrospective research and also assists in planning prospective trials where the investigators can perform feasibility analyses for clinical trials to determine the number of potential participants available based on inclusion or exclusion criteria. In this way they can be practical in their study design and increase the likelihood of trials meeting recruitment goals. The system can also include genomic data for clinical genomic research. BMI investigator uses the OMOP common data model and adds Elastic Search indices using high definition natural language processing software to parse problems, procedures and devices in SNOMED CT, medications in RxNorm and laboratory test results in LOINC. Contact: Peter Elkin, MD.
Ontology development and use
The National Center for Ontological Research (NCOR) was established in Buffalo in 2005 with the goal of advancing the quality of ontological research and development and of establishing tools and measures for ontology evaluation and quality assurance. NCOR serves as a vehicle to coordinate, enhance, publicize, and seek funding for ontological research activities. It provides coordination, infrastructure, and independent review to organizations employing ontologies in fields such as defense and intelligence, management, healthcare and biomedical sciences. Contacts: Barry Smith, PhD; John Beverley, PhD.
Enhancing the power of SNOMED CT through the Basic Formal Ontology
SNOMED CT is a large concept-based terminology designed according to epistemic, semantic and pragmatic principles relevant to clinicians. Its goal is structured clinical reporting in electronic healthcare records (EHRs). The Basic Formal Ontology (BFO) is an ontology designed on the basis of types claimed to exist in reality based on a domain-independent ontological theory. Its goal is faithful representation of reality within that theory. The Ontology for General Medical Science (OGMS) extends the BFO by providing definitions for types relevant within the clinical domain. Combining SNOMED CT with the ontological rigor of BFO and OGMS might improve clinical reporting by, for instance, preventing data entry mistakes and inconsistencies, and make EHRs more comparable. To that end, we are developing a logical framework capable of exploiting what SNOMED CT offers terminologically and realism-based ontologies such as the BFO and the OGMS ontologically by means of bridging axioms compatible with the BFO, and expressed in the same CLIF-dialect as used in its axiomatization in first order logic. Contacts: Werner Ceusters, MD; Michael Rabenberg, PhD.
Referent Tracking
Terminological systems, including coding and classification systems, are used in electronic medical record systems to facilitate the interpretation of structured data by providing terms and codes with a relatively precise meaning. When a clinician selects a term or code from such system and enters it in the medical record of a patient, then, from an ontological perspective and as a consequence of how terminological systems are currently integrated in electronic medical record systems, an assertion has been made to the effect that the patient exhibits, or exhibited, some phenomenon of type T. It is however left unspecified which phenomenon in particular is of the designated type T. In other words: such records contain explicit references, i.e., the terms or codes, but the referents of these references are not explicitly identified. Because referents can be referenced in many different ways, types used as references can be about many referents, and referents may change so they become of a different type, data analytics applications which rely on types only are prone to drawing erroneous conclusions. Referent Tracking is a methodology for data management which allows assertions only to be made with explicit reference to the referents they are about. Contacts: Werner Ceusters, MD; Michael Rabenberg, PhD.
Magnetic Resonance Imaging Acquisition and Analysis Ontology (MRIO)
In collaboration with Drs. Alexander Bartnik and Michael Dwyer of the CTSI Center for Biomedical Imaging, we have created the Magnetic Resonance Imaging Acquisition and Analysis Ontology. MRIO provides well-reasoned classes and logical axioms for the acquisition of several MRI acquisition types and well-known, peer-reviewed analysis software, facilitating the use of MRI data. These classes provide a common language for the neuroimaging research process and help standardize the organization and analysis of MRI data for reproducible datasets. MRIO is used in conjunction with an MR image database and analysis system developed at the CTSI Center for Biomedical Imaging. Contacts: Alexander Diehl, PhD; Alexander Bartnik, PhD; Michael Dwyer, PhD.