J Kyle Armstrong, Phd Email & Phone Number

Philadelphia, Pa, Us

Philadelphia, Pennsylvania, Us

𝐑𝐞𝐬𝐩𝐨𝐧𝐬𝐢𝐛𝐢𝐥𝐢𝐭𝐢𝐞𝐬● Design and teach innovative Master's level courses including Data Wrangling, Deep Learning, and Advanced Data Analysis.● Integrate modern technologies including Python, R and machine learning into curriculum, and serve on faculty committees.HDS 500 - 𝘍𝘶𝘯𝘥𝘢𝘮𝘦𝘯𝘵𝘢𝘭𝘴 𝘰𝘧 𝘋𝘢𝘵𝘢 𝘞𝘳𝘢𝘯𝘨𝘭𝘪𝘯𝘨● Course Development & Content Development● Instructor of record: Spring 2022, Summer 2022HDS 502 - 𝘈𝘥𝘷𝘢𝘯𝘤𝘦𝘥 𝘋𝘢𝘵𝘢 𝘈𝘯𝘢𝘭𝘺𝘴𝘪𝘴 ● Assisted with course designPHI 518 - 𝘋𝘢𝘵𝘢 𝘚𝘤𝘪𝘦𝘯𝘤𝘦 𝘸𝘪𝘵𝘩 𝘙 ● Course Development & Content Development● Instructor of record: Spring 2020 , Summer 2020 , Spring 2021HDS 519 - 𝘋𝘢𝘵𝘢 𝘚𝘤𝘪𝘦𝘯𝘤𝘦 𝘐𝘐 - 𝘋𝘦𝘦𝘱 𝘓𝘦𝘢𝘳𝘯𝘪𝘯𝘨 𝘸𝘪𝘵𝘩 𝘱𝘺𝘵𝘩𝘰𝘯● Course Development & Content Development● Instructor of record: Spring 2022, Spring 2023𝐂𝐨𝐦𝐦𝐢𝐭𝐭𝐞𝐞𝐬● Dissertation - Karen Walsh (University of Indianapolis)● Masters - Guorong Li

Philadelphia, Pa, Us

𝐑𝐞𝐬𝐩𝐨𝐧𝐬𝐢𝐛𝐢𝐥𝐢𝐭𝐢𝐞𝐬• Actively research and utilize the latest developments in machine learning and artificial intelligence to work on projects involving both structured and unstructured data.• Lead data scientist and machine learning engineer on projects, working with stakeholders to gather requirements and make key decisions on methodology, process, and code.• Develop and implement machine learning pipelines to solve complex business challenges.• Evaluated medical management programs to determine their clinical and financial outcomes and provided recommendations to Executive Management.• Provide mentorship and promote best practices related to code standards and optimization.• Collaborate with business partners to develop statements of work and partnerships with research organizations.𝐏𝐫𝐨𝐣𝐞𝐜𝐭𝐬• Managed ETL processes and SQL using R, Python, TeraData, HIVE, and Impala for data wrangling to support 40+ machine learning models, reducing model creation time from months to weeks.• Developed Likelihood of Impact predictive model to identify members likely to have cost savings with Case Management, using nearest neighbor and random forest classifiers. • Built an Adverse Pregnancy Outcomes Predictive Model using OMOP CDM to identify pregnancies and multiclass classifier to predict normal pregnancies, various adverse events, or deaths.• Created Estimation of Cost Estimates for Commercial Population Spark Random Forest regression model, providing monthly estimates on total medical costs to inform cost-saving initiatives.

Philadelphia, Pa, Us

𝐑𝐞𝐬𝐩𝐨𝐧𝐬𝐢𝐛𝐢𝐥𝐢𝐭𝐢𝐞𝐬• Evaluated medical management programs to determine their clinical and financial outcomes and provided recommendations to Executive Management.• Developed mathematical and statistical models to analyze data and provide actionable insights.• Reported findings, interpreted results, and provided suggestions for further analysis.• Demonstrated proficiency in using various tools and platforms including Cloudera Data Science Work Bench, TeraData, & Jupyter.• Developed libraries of R, SparklyR, Python, PySpark, and SAS programs and macros to structure, organize, and analyze data.• Designed data and analysis pipelines for predictive modeling. 𝐏𝐫𝐨𝐣𝐞𝐜𝐭𝐬• Conducted a Case-Control evaluation of a value-based healthcare organization for End of Life Cost Evaluation using a High Dimensional Propensity Score Matching Algorithm.• Led the feasibility, quality assurance, and parallel development of a company-wide migration from TeraData and SAS to R, Python, and distributed computational platforms utilizing Hadoop.• Estimated risk scores and costs for new members within the Affordable Care Act Population and predicted multiple outcomes using SparklyR with Spark ML pipelines and cross-fold validation.• Conducted a Payment Integrity, Fraud, Waste, and Abuse (FWA) Predictive Model Pilot, utilizing PySpark feature derivation and random forest models to predict FWA with Modifier 25 using medical claims.• Predicted Adverse Pregnancy Outcomes in a cohort of over 20K females, independently programming over 1K analysis variables and various outcome conditions. Produced a predictive model with over .75 c-statistic and provided estimates for mean time until adverse pregnancy outcome and Kaplan-Meier curves.

Philadelphia, Pa, Us

𝐑𝐞𝐬𝐩𝐨𝐧𝐬𝐢𝐛𝐢𝐥𝐢𝐭𝐢𝐞𝐬• Serve as lead analyst on multiple projects, making key decisions on study methodology• Develop Statements of Work and serve as primary point of contact for partnerships with leading research organizations• Evaluate medical management programs to determine clinical and financial outcomes and make recommendations to Executive Management• Act as a mentor by providing methodology suggestions and training• Research and apply developments in methodology and analysis of Big Data• Produce mathematical and statistical models to analyze data and provide actionable insights• Independently develop predictive models• Report findings, interpret results, and provide suggestions for further analysis• Write efficient SQL and SAS code to interface with extensive relational Teradata database• Development of a library of SAS programs and macros to structure, organize, and analyze data• Use SAS Enterprise Miner to develop predictive models• Review, edit, and develop Python and PySpark code for data preparation, analysis, and predictive modeling𝐏𝐫𝐨𝐣𝐞𝐜𝐭𝐬• Case-Control evaluation utilizing a High Dimensional Propensity Score matching algorithm with over 7K exposed members and 30K unexposed members. • A/B comparison of diabetic predictive models. • Evaluation of optical character recognition (OCR) and natural language processing (NLP) compared to manual review on a set of scanned medical charts.• Analysis of Survival and Efficient Treatment Options in Subjects with Chronic Kidney Disease and End Stage Renal Failure.

Morrisville, Nc, Us

𝐑𝐞𝐬𝐩𝐨𝐧𝐬𝐢𝐛𝐢𝐥𝐢𝐭𝐢𝐞𝐬• Provide statistical consulting• Serve as client facing lead on multiple studies• Prepare statistical sections of protocols• Perform sample size and power calculations for studies• Create randomization schedule per study design• Review study protocols, case report forms, data management plans, and other related study documents • Independently develop statistical analysis plans in accordance with study protocols• Support integrated summaries of safety and efficacy for regulatory submissions• Independently prepare analysis dataset specifications• Develop SAS programs to create derived variables and analysis datasets• Independently develop SAS programs to generate tables, listings, and figures • Perform quality control of analysis datasets, tables, listings, and figures• Prepare statistical reports• Review, provide input and interpretation of analysis results to clinical study reports• Apply programming techniques to enhance programming efficiency and improve productivity• Provide input and facilitate conversations regarding CDISC, SDTM, and ADaM standards.• Provide a knowledge base and support to other Biostatisticians𝐓𝐡𝐞𝐫𝐚𝐩𝐞𝐮𝐭𝐢𝐜 𝐄𝐱𝐩𝐞𝐫𝐢𝐞𝐧𝐜𝐞• Genetics• Oncology• Pain

Tallahassee, Fl, Us

𝐑𝐞𝐬𝐩𝐨𝐧𝐬𝐢𝐛𝐢𝐥𝐢𝐭𝐢𝐞𝐬• Classroom instruction and evaluation of approximately 35-40 undergraduate students per section.MAC 2313 - 𝘊𝘢𝘭𝘤𝘶𝘭𝘶𝘴 𝘸𝘪𝘵𝘩 𝘈𝘯𝘢𝘭𝘺𝘵𝘪𝘤 𝘎𝘦𝘰𝘮𝘦𝘵𝘳𝘺 𝘐𝘐𝘐• Instructor of record: Fall 2011 & Summer 2010. MAC 2312 - 𝘊𝘢𝘭𝘤𝘶𝘭𝘶𝘴 𝘸𝘪𝘵𝘩 𝘈𝘯𝘢𝘭𝘺𝘵𝘪𝘤 𝘎𝘦𝘰𝘮𝘦𝘵𝘳𝘺 𝘐𝘐• Teaching Assistant: Spring 2009 & Fall 2008MAC 2311 - 𝘊𝘢𝘭𝘤𝘶𝘭𝘶𝘴 𝘸𝘪𝘵𝘩 𝘈𝘯𝘢𝘭𝘺𝘵𝘪𝘤 𝘎𝘦𝘰𝘮𝘦𝘵𝘳𝘺 𝘐• Instructor of record: Summer 2009 & Summer 2008• Teaching Assistant: Spring 2010MAC 1140 - 𝘗𝘳𝘦𝘤𝘢𝘭𝘤𝘶𝘭𝘶𝘴 𝘈𝘭𝘨𝘦𝘣𝘳𝘢• Instructor of record: Fall 2009, Spring 2007, & Fall 2007• Teaching Assistant: Fall 2009, Spring 2007, & Fall 2007

Doctor Of Philosophy (Phd), Mathematics

Master Of Science (Ms), Mathematics

Bachelor'S Of Science (Bs), Mathematics, Statistics (Double Major)