Data Science Commons^™

Text Classification Of Cancer Clinical Trial Eligibility Criteria, Yumeng Yang, Soumya Jayaraj, Ethan Ludmir, Kirk Roberts

Faculty, Staff and Student Publications

Automatic identification of clinical trials for which a patient is eligible is complicated by the fact that trial eligibility are stated in natural language. A potential solution to this problem is to employ text classification methods for common types of eligibility criteria. In this study, we focus on seven common exclusion criteria in cancer trials: prior malignancy, human immunodeficiency virus, hepatitis B, hepatitis C, psychiatric illness, drug/substance abuse, and autoimmune illness. Our dataset consists of 764 phase III cancer trials with these exclusions annotated at the trial level. We experiment with common transformer models as well as a new pre-trained …

Sensitive Data Detection With High-Throughput Machine Learning Models In Electrical Health Records, Kai Zhang, Xiaoqian Jiang

Faculty, Staff and Student Publications

In the era of big data, there is an increasing need for healthcare providers, communities, and researchers to share data and collaborate to improve health outcomes, generate valuable insights, and advance research. The Health Insurance Portability and Accountability Act of 1996 (HIPAA) is a federal law designed to protect sensitive health information by defining regulations for protected health information (PHI). However, it does not provide efficient tools for detecting or removing PHI before data sharing. One of the challenges in this area of research is the heterogeneous nature of PHI fields in data across different parties. This variability makes rule-based …

Annotation And Information Extraction Of Consumer-Friendly Health Articles For Enhancing Laboratory Test Reporting, Zhe He, Shubo Tian, Arslan Erdengasileng, Karim Hanna, Yang Gong, Zhan Zhang, Xiao Luo, Mia Liza A Lustria

Faculty, Staff and Student Publications

Viewing laboratory test results is patients' most frequent activity when accessing patient portals, but lab results can be very confusing for patients. Previous research has explored various ways to present lab results, but few have attempted to provide tailored information support based on individual patient's medical context. In this study, we collected and annotated interpretations of textual lab result in 251 health articles about laboratory tests from AHealthyMe.com. Then we evaluated transformer-based language models including BioBERT, ClinicalBERT, RoBERTa, and PubMedBERT for recognizing key terms and their types. Using BioPortal's term search API, we mapped the annotated terms to concepts in …

Experiences Of Parents With Opioid Use Disorder During Their Attempts To Seek Treatment: A Qualitative Analysis, Christine Bakos-Block, Angela J Nash, A Sarah Cohen, Tiffany Champagne-Langabeer

Faculty, Staff and Student Publications

In the U.S., 12.3% of children live with at least one parent who has a substance use disorder. Prior research has shown that men are more likely to seek treatment than women and that the barriers are different; however, there is limited research focusing specifically on opioid use disorder (OUD). We sought to understand the barriers and motivators for parents with OUD. We conducted a qualitative study by interviewing parents with OUD who were part of an outpatient treatment program. Interviews followed a semi-structured format with questions on access to and motivation for treatment. The interviews were recorded and transcribed …

High-Frequency Ultrasound In Patients With Seronegative Rheumatoid Arthritis, Junkui Wang, Miao Wang, Qinghua Qi, Zhibin Wu, Jianguo Wen

Faculty, Staff and Student Publications

This study aimed to investigate the value of high-frequency ultrasound (HFUS) in differentiation of the seronegative rheumatoid arthritis (SNRA) and osteoarthritis (OA) and in the diagnosis of SNRA. 83 patients diagnosed with SNRA (SNRA group) and 40 diagnosed with OA (OA group) who received HFUS were retrospectively analyzed. The grayscale (GS) scores, power Doppler (PD) scores, and bone erosion (BE)scores were recorded, and added up to calculate the total scores of US variables. The correlations of the total scores of US variables with the 28-joint disease activity score (DAS28), erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) were analyzed. The …

Open-Source Benchmarking Of Ibd Segment Detection Methods For Biobank-Scale Cohorts, Kecong Tang, Ardalan Naseri, Yuan Wei, Shaojie Zhang, Degui Zhi

Faculty, Staff and Student Publications

In the recent biobank era of genetics, the problem of identical-by-descent (IBD) segment detection received renewed interest, as IBD segments in large cohorts offer unprecedented opportunities in the study of population and genealogical history, as well as genetic association of long haplotypes. While a new generation of efficient methods for IBD segment detection becomes available, direct comparison of these methods is difficult: existing benchmarks were often evaluated in different datasets, with some not openly accessible; methods benchmarked were run under suboptimal parameters; and benchmark performance metrics were not defined consistently. Here, we developed a comprehensive and completely open-source evaluation of …

Intraoperative Localization And Preservation Of Reading In Ventral Occipitotemporal Cortex, Oscar Woolnough, Kathryn M Snyder, Cale W Morse, Meredith J Mccarty, Samden D Lhatoo, Nitin Tandon

Faculty, Staff and Student Publications

OBJECTIVE: Resective surgery in language-dominant ventral occipitotemporal cortex (vOTC) carries the risk of causing impairment to reading. Because it is not on the lateral surface, it is not easily accessible for intraoperative mapping, and extensive stimulation mapping can be time-consuming. Here the authors assess the feasibility of using task-based electrocorticography (ECoG) recordings intraoperatively to help guide stimulation mapping of reading in vOTC.

METHODS: In 11 patients undergoing extraoperative, intracranial seizure mapping, the authors recorded induced broadband gamma activation (70-150 Hz) during a visual category localizer. In 2 additional patients, whose pathologies necessitated resections in language-dominant vOTC, task-based functional mapping was …

Privacy-Aware Estimation Of Relatedness In Admixed Populations, Su Wang, Miran Kim, Wentao Li, Xiaoqian Jiang, Han Chen, Arif Harmanci

Faculty, Staff and Student Publications

BACKGROUND: Estimation of genetic relatedness, or kinship, is used occasionally for recreational purposes and in forensic applications. While numerous methods were developed to estimate kinship, they suffer from high computational requirements and often make an untenable assumption of homogeneous population ancestry of the samples. Moreover, genetic privacy is generally overlooked in the usage of kinship estimation methods. There can be ethical concerns about finding unknown familial relationships in third-party databases. Similar ethical concerns may arise while estimating and reporting sensitive population-level statistics such as inbreeding coefficients for the concerns around marginalization and stigmatization.

RESULTS: Here, we present SIGFRIED, which makes …

A Comprehensive Artificial Intelligence Framework For Dental Diagnosis And Charting, Tanjida Kabir, Chun-Teh Lee, Luyao Chen, Xiaoqian Jiang, Shayan Shams

Faculty, Staff and Student Publications

BACKGROUND: The aim of this study was to develop artificial intelligence (AI) guided framework to recognize tooth numbers in panoramic and intraoral radiographs (periapical and bitewing) without prior domain knowledge and arrange the intraoral radiographs into a full mouth series (FMS) arrangement template. This model can be integrated with different diseases diagnosis models, such as periodontitis or caries, to facilitate clinical examinations and diagnoses.

METHODS: The framework utilized image segmentation models to generate the masks of bone area, tooth, and cementoenamel junction (CEJ) lines from intraoral radiographs. These masks were used to detect and extract teeth bounding boxes utilizing several …

Atomistic Measurement And Modeling Of Intrinsic Fracture Toughness Of Two-Dimensional Materials, Xu Zhang, Hoang Nguyen, Xiang Zhang, Pulickel M Ajayan, Jianguo Wen, Horacio D Espinosa

Faculty, Staff and Student Publications

Quantifying the intrinsic mechanical properties of two-dimensional (2D) materials is essential to predict the long-term reliability of materials and systems in emerging applications ranging from energy to health to next-generation sensors and electronics. Currently, measurements of fracture toughness and identification of associated atomistic mechanisms remain challenging. Herein, we report an integrated experimental-computational framework in which in-situ high-resolution transmission electron microscopy (HRTEM) measurements of the intrinsic fracture energy of monolayer MoS

The Impact Of Covid-19 On Opioid-Related Overdose Deaths In Texas, Karima Lalani, Christine Bakos-Block, Marylou Cardenas-Turanzas, Sarah Cohen, Bhanumathi Gopal, Tiffany Champagne-Langabeer

Faculty, Staff and Student Publications

Prior to the COVID-19 pandemic, the United States was facing an epidemic of opioid overdose deaths, clouding accurate inferences about the impact of the pandemic at the population level. We sought to determine the existence of increases in the trends of opioid-related overdose (ORO) deaths in the Greater Houston metropolitan area from January 2015 through December 2021, and to describe the social vulnerability present in the geographic location of these deaths. We merged records from the county medical examiner's office with social vulnerability indexes (SVIs) for the region and present geospatial locations of the aggregated ORO deaths. Time series analyses …

Scgwas: Landscape Of Trait-Cell Type Associations By Integrating Single-Cell Transcriptomics-Wide And Genome-Wide Association Studies, Peilin Jia, Ruifeng Hu, Fangfang Yan, Yulin Dai, Zhongming Zhao

Faculty, Staff and Student Publications

BACKGROUND: The rapid accumulation of single-cell RNA sequencing (scRNA-seq) data presents unique opportunities to decode the genetically mediated cell-type specificity in complex diseases. Here, we develop a new method, scGWAS, which effectively leverages scRNA-seq data to achieve two goals: (1) to infer the cell types in which the disease-associated genes manifest and (2) to construct cellular modules which imply disease-specific activation of different processes.

RESULTS: scGWAS only utilizes the average gene expression for each cell type followed by virtual search processes to construct the null distributions of module scores, making it scalable to large scRNA-seq datasets. We demonstrated scGWAS in …

Federated Learning Algorithms For Generalized Mixed-Effects Model (Glmm) On Horizontally Partitioned Data From Distributed Sources, Wentao Li, Jiayi Tong, Md Monowar Anjum, Noman Mohammed, Yong Chen, Xiaoqian Jiang

Faculty, Staff and Student Publications

OBJECTIVES: This paper developed federated solutions based on two approximation algorithms to achieve federated generalized linear mixed effect models (GLMM). The paper also proposed a solution for numerical errors and singularity issues. And showed the two proposed methods can perform well in revealing the significance of parameter in distributed datasets, comparing to a centralized GLMM algorithm from R package ('lme4') as the baseline model.

METHODS: The log-likelihood function of GLMM is approximated by two numerical methods (Laplace approximation and Gaussian Hermite approximation, abbreviated as LA and GH), which supports federated decomposition of GLMM to bring computation to data. To solve …

Aging After Stroke: How To Define Post-Stroke Sarcopenia And What Are Its Risk Factors?, Sheng Li, Javier Gonzalez-Buonomo, Jaskiran Ghuman, Xinran Huang, Aila Malik, Nuray Yozbatiran, Elaine Magat, Gerard E Francisco, Hulin Wu, Walter R Frontera

Faculty, Staff and Student Publications

BACKGROUND: Sarcopenia, generally described as "aging-related loss of skeletal muscle mass and function", can occur secondary to a systemic disease.

AIM: This project aimed to study the prevalence of sarcopenia in chronic ambulatory stroke survivors and its associated risk factors using the two most recent diagnostic criteria.

DESIGN: A cross-sectional observational study.

SETTING: A scientific laboratory.

POPULATION: Chronic stroke.

METHODS: Twenty-eight ambulatory chronic stroke survivors (12 females; mean age=57.8±11.8 years; time after stroke=76±45 months), hand-grip strength, gait speed, and appendicular skeletal muscle mass (ASM) were measured to define sarcopenia. Risk factors, including motor impairment and spasticity, were identified using regression …

Svat: Secure Outsourcing Of Variant Annotation And Genotype Aggregation, Miran Kim, Su Wang, Xiaoqian Jiang, Arif Harmanci

Faculty, Staff and Student Publications

BACKGROUND: Sequencing of thousands of samples provides genetic variants with allele frequencies spanning a very large spectrum and gives invaluable insight into genetic determinants of diseases. Protecting the genetic privacy of participants is challenging as only a few rare variants can easily re-identify an individual among millions. In certain cases, there are policy barriers against sharing genetic data from indigenous populations and stigmatizing conditions.

RESULTS: We present SVAT, a method for secure outsourcing of variant annotation and aggregation, which are two basic steps in variant interpretation and detection of causal variants. SVAT uses homomorphic encryption to encrypt the data at …

Development Of The Invasive Candidiasis Discharge [I Can Discharge] Model: A Mixed Methods Analysis, Jinhee Jo, Truc T Tran, Nicholas D Beyda, Debora Simmons, Joshua A Hendrickson, Masaad Saeed Almutairi, Faris S Alnezary, Anne J Gonzales-Luna, Edward J Septimus, Kevin W Garey

Faculty, Staff and Student Publications

Patients with invasive candidiasis (IC) have complex medical and infectious disease problems that often require continued care after discharge. This study aimed to assess echinocandin use at hospital discharge and develop a transition of care (TOC) model to facilitate discharge for patients with IC. This was a mixed method study design that used epidemiologic assessment to better understand echinocandin use at hospital discharge TOC. Using grounded theory methodology focused on patients given echinocandins during their last day of hospitalization, a TOC model for patients with IC, the invasive candidiasis [I Can] discharge model was developed to better understand discharge barriers. …

Video-Urodynamics Efficacy Of Sacral Neuromodulation For Neurogenic Bladder Guided By Three-Dimensional Imaging Ct And C-Arm Fluoroscopy: A Single-Center Prospective Study, Shuaishuai Shan, Wen Zhu, Guoxian Zhang, Qinyong Zhang, Yingyu Che, Jianguo Wen, Qingwei Wang

Faculty, Staff and Student Publications

To assess the efficacy of sacral neuromodulation (SNM) for neurogenic bladder (NB), guided by intraoperative three-dimensional imaging of sacral computed tomography (CT) and mobile C-arm fluoroscopy through video-urodynamics examination. We enrolled 52 patients with NB who underwent conservative treatment with poor results between September 2019 and June 2021 and prospectively underwent SNM guided by intraoperative three-dimensional imaging of sacral CT and mobile C-arm fluoroscopy. Video-urodynamics examination, voiding diary, quality of life questionnaire, overactive bladder symptom scale (OABSS) scoring, and bowel dysfunction exam were completed and recorded at baseline, at SNM testing, and at 6-month follow-up phases. Finally, we calculated the …

Identifying Candidate Genes And Drug Targets For Alzheimer’S Disease By An Integrative Network Approach Using Genetic And Brain Region-Specific Proteomic Data, Andi Liu, Astrid M Manuel, Yulin Dai, Brisa S Fernandes, Nitesh Enduru, Peilin Jia, Zhongming Zhao

Faculty, Staff and Student Publications

Genome-wide association studies (GWAS) have identified more than 75 genetic variants associated with Alzheimer's disease (ad). However, how these variants function and impact protein expression in brain regions remain elusive. Large-scale proteomic datasets of ad postmortem brain tissues have become available recently. In this study, we used these datasets to investigate brain region-specific molecular pathways underlying ad pathogenesis and explore their potential drug targets. We applied our new network-based tool, Edge-Weighted Dense Module Search of GWAS (EW_dmGWAS), to integrate ad GWAS statistics of 472 868 individuals with proteomic profiles from two brain regions from two large-scale ad cohorts [parahippocampal gyrus …

Molecular Pathways Enhance Drug Response Prediction Using Transfer Learning From Cell Lines To Tumors And Patient-Derived Xenografts, Yi-Ching Tang, Reid T Powell, Assaf Gottlieb

Faculty, Staff and Student Publications

Computational models have been successful in predicting drug sensitivity in cancer cell line data, creating an opportunity to guide precision medicine. However, translating these models to tumors remains challenging. We propose a new transfer learning workflow that transfers drug sensitivity predicting models from large-scale cancer cell lines to both tumors and patient derived xenografts based on molecular pathways derived from genomic features. We further compute feature importance to identify pathways most important to drug response prediction. We obtained good performance on tumors (AUROC = 0.77) and patient derived xenografts from triple negative breast cancers (RMSE = 0.11). Using feature importance, …

Pim1 Promotes Hepatic Conversion By Suppressing Reprogramming-Induced Ferroptosis And Cell Cycle Arrest, Yangyang Yuan, Chenwei Wang, Xuran Zhuang, Shaofeng Lin, Miaomiao Luo, Wankun Deng, Jiaqi Zhou, Lihui Liu, Lina Mao, Wenbo Peng, Jian Chen, Qiangsong Wang, Yilai Shu, Yu Xue, Pengyu Huang

Faculty, Staff and Student Publications

Protein kinase-mediated phosphorylation plays a critical role in many biological processes. However, the identification of key regulatory kinases is still a great challenge. Here, we develop a trans-omics-based method, central kinase inference, to predict potentially key kinases by integrating quantitative transcriptomic and phosphoproteomic data. Using known kinases associated with anti-cancer drug resistance, the accuracy of our method denoted by the area under the curve is 5.2% to 29.5% higher than Kinase-Substrate Enrichment Analysis. We further use this method to analyze trans-omic data in hepatocyte maturation and hepatic reprogramming of human dermal fibroblasts, uncovering 5 kinases as regulators in the two …

Biomolecular Condensation: A New Phase In Cancer Research, Anupam K Chakravarty, Daniel J Mcgrail, Thomas M Lozanoski, Brandon S Dunn, David J H Shih, Kara M Cirillo, Sueda H Cetinkaya, Wenjin Jim Zheng, Gordon B Mills, S Stephen Yi, Daniel F Jarosz, Nidhi Sahni

Faculty, Staff and Student Publications

Multicellularity was a watershed development in evolution. However, it also meant that individual cells could escape regulatory mechanisms that restrict proliferation at a severe cost to the organism: cancer. From the standpoint of cellular organization, evolutionary complexity scales to organize different molecules within the intracellular milieu. The recent realization that many biomolecules can "phase-separate" into membraneless organelles, reorganizing cellular biochemistry in space and time, has led to an explosion of research activity in this area. In this review, we explore mechanistic connections between phase separation and cancer-associated processes and emerging examples of how these become deranged in malignancy.

SIGNIFICANCE: One …

Assessing Age-Specific Vaccination Strategies And Post-Vaccination Reopening Policies For Covid-19 Control Using Seir Modeling Approach, Xia Wang, Hulin Wu, Sanyi Tang

Faculty, Staff and Student Publications

As the availability of COVID-19 vaccines, it is badly needed to develop vaccination guidelines to prioritize the vaccination delivery in order to effectively stop COVID-19 epidemic and minimize the loss. We evaluated the effect of age-specific vaccination strategies on the number of infections and deaths using an SEIR model, considering the age structure and social contact patterns for different age groups for each of different countries. In general, the vaccination priority should be given to those younger people who are active in social contacts to minimize the number of infections, while the vaccination priority should be given to the elderly …

Controlling Multiple Covid-19 Epidemic Waves: An Insight From A Multi-Scale Model Linking The Behaviour Change Dynamics To The Disease Transmission Dynamics, Biao Tang, Weike Zhou, Xia Wang, Hulin Wu, Yanni Xiao

Faculty, Staff and Student Publications

COVID-19 epidemics exhibited multiple waves regionally and globally since 2020. It is important to understand the insight and underlying mechanisms of the multiple waves of COVID-19 epidemics in order to design more efficient non-pharmaceutical interventions (NPIs) and vaccination strategies to prevent future waves. We propose a multi-scale model by linking the behaviour change dynamics to the disease transmission dynamics to investigate the effect of behaviour dynamics on COVID-19 epidemics using game theory. The proposed multi-scale models are calibrated and key parameters related to disease transmission dynamics and behavioural dynamics with/without vaccination are estimated based on COVID-19 epidemic data (daily reported …

Immuno-Genomic Profiling Of Biopsy Specimens Predicts Neoadjuvant Chemotherapy Response In Esophageal Squamous Cell Carcinoma, Shota Sasagawa, Hiroaki Kato, Koji Nagaoka, Changbo Sun, Motohiro Imano, Takao Sato, Todd A Johnson, Masashi Fujita, Kazuhiro Maejima, Yuki Okawa, Kazuhiro Kakimi, Takushi Yasuda, Hidewaki Nakagawa

Faculty, Staff and Student Publications

Esophageal squamous cell carcinoma (ESCC) is one of the most aggressive cancers and is primarily treated with platinum-based neoadjuvant chemotherapy (NAC). Some ESCCs respond well to NAC. However, biomarkers to predict NAC sensitivity and their response mechanism in ESCC remain unclear. We perform whole-genome sequencing and RNA sequencing analysis of 141 ESCC biopsy specimens before NAC treatment to generate a machine-learning-based diagnostic model to predict NAC reactivity in ESCC and analyzed the association between immunogenomic features and NAC response. Neutrophil infiltration may play an important role in ESCC response to NAC. We also demonstrate that specific copy-number alterations and copy-number …

Measuring And Controlling Medical Record Abstraction (Mra) Error Rates In An Observational Study, Maryam Y Garza, Tremaine Williams, Sahiti Myneni, Susan H Fenton, Songthip Ounpraseuth, Zhuopei Hu, Jeannette Lee, Jessica Snowden, Meredith N Zozus, Anita C Walden, Alan E Simon, Barbara Mcclaskey, Sarah G Sanders, Sandra S Beauman, Sara R Ford, Lacy Malloch, Amy Wilson, Lori A Devlin, Leslie W Young

Faculty, Staff and Student Publications

BACKGROUND: Studies have shown that data collection by medical record abstraction (MRA) is a significant source of error in clinical research studies relying on secondary use data. Yet, the quality of data collected using MRA is seldom assessed. We employed a novel, theory-based framework for data quality assurance and quality control of MRA. The objective of this work is to determine the potential impact of formalized MRA training and continuous quality control (QC) processes on data quality over time.

METHODS: We conducted a retrospective analysis of QC data collected during a cross-sectional medical record review of mother-infant dyads with Neonatal …

Secure Human Action Recognition By Encrypted Neural Network Inference, Miran Kim, Xiaoqian Jiang, Kristin Lauter, Elkhan Ismayilzada, Shayan Shams

Faculty, Staff and Student Publications

Advanced computer vision technology can provide near real-time home monitoring to support "aging in place" by detecting falls and symptoms related to seizures and stroke. Affordable webcams, together with cloud computing services (to run machine learning algorithms), can potentially bring significant social benefits. However, it has not been deployed in practice because of privacy concerns. In this paper, we propose a strategy that uses homomorphic encryption to resolve this dilemma, which guarantees information confidentiality while retaining action detection. Our protocol for secure inference can distinguish falls from activities of daily living with 86.21% sensitivity and 99.14% specificity, with an average …

Building Consensus For A Shared Definition Of Adverse Events: A Case Study In The Profession Of Dentistry, Amy Franklin, Elsbeth Kalenderian, Nutan Hebballi, Veronique Delattre, Jini Etoule, Joel White, Ram Vaderhobli, Denice Stewart, Karla Kent, Alfa Yansane, Muhammad Walji

Faculty, Staff and Student Publications

BACKGROUND: To achieve high-quality health care, adverse events (AEs) must be proactively recognized and mitigated. However, there is often ambiguity in applying guidelines and definitions. We describe the iterative calibration process needed to achieve a shared definition of AEs in dentistry. Our alignment process includes both independent and consensus building approaches.

OBJECTIVE: We explore the process of defining dental AEs and the steps necessary to achieve alignment across different care providers.

METHODS: Teams from 4 dental institutions across the United States iteratively reviewed patient records after identification of charts using an automated trigger tool. Calibration across teams was supported through …

Charting The Proteome Landscape In Major Psychiatric Disorders: From Biomarkers To Biological Pathways Towards Drug Discovery, Brisa S Fernandes, Yulin Dai, Peilin Jia, Zhongming Zhao

Faculty, Staff and Student Publications

Schizophrenia (SZ), bipolar disorder (BD), and major depressive disorder (MDD) are major mental disorders that affect a significant proportion of the global population. Advancing our knowledge of the pathophysiology of these disorders and identifying biomarkers are urgent needs for developing objective diagnostic tests and new therapeutics. In this study, we performed a systematic review and then extracted, curated, and analyzed proteomics data from published studies, aiming to assess the proteome in peripheral blood of individuals with SZ, BD, or MDD. Then, we performed pathway and network analyses to illuminate the biological themes concatenated by the differentially expressed proteins by systematically …

Development Of A Quality Improvement Dental Chart Review Training Program, Elsbeth Kalenderian, Nutan B Hebballi, Amy Franklin, Alfa Yansane, Ana M Ibarra Noriega, Joel White, Muhammad F Walji

Faculty, Staff and Student Publications

INTRODUCTION: Chart review is central to understanding adverse events (AEs) in medicine. In this article, we describe the process and results of educating chart reviewers assigned to evaluate dental AEs.

METHODS: We developed a Web-based training program, "Dental Patient Safety Training," which uses both independent and consensus-based curricula, for identifying AEs recorded in electronic health records in the dental setting. Training included (1) didactic education, (2) skills training using videos and guided walkthroughs, (3) quizzes with feedback, and (4) hands-on learning exercises. In addition, novice reviewers were coached weekly during consensus review discussions. TeamExpert was composed of 2 experienced reviewers, …

Real-World Matching Performance Of Deidentified Record-Linking Tokens, Elmer V Bernstam, Reuben Joseph Applegate, Alvin Yu, Deepa Chaudhari, Tian Liu, Alex Coda, Jonah Leshin

Faculty, Staff and Student Publications

OBJECTIVE: Our objective was to evaluate tokens commonly used by clinical research consortia to aggregate clinical data across institutions.

METHODS: This study compares tokens alone and token-based matching algorithms against manual annotation for 20,002 record pairs extracted from the University of Texas Houston's clinical data warehouse (CDW) in terms of entity resolution.

RESULTS: The highest precision achieved was 99.9% with a token derived from the first name, last name, gender, and date-of-birth. The highest recall achieved was 95.5% with an algorithm involving tokens that reflected combinations of first name, last name, gender, date-of-birth, and social security number.

DISCUSSION: To protect …