MSBA, MSW, Data Science Manager
BD
Biography
Data Science Manager with over 7 years of work experience in healthcare and life sciences industries. Advanced machine learning modeling, statistical and visualization skills with expertise in life sciences.
Interests
- Applied Machine Learning
- Generative AI
- API Integration
- Natural Language Processing (NLP)
- Data visualization
- Electronic Health Records (EHRs)
- Real World Evidence (RWE)
- Recommendation Systems
- Bioinformatics
Education
Master of Business Analytics (MSBA), 2020
UCSD Rady School of Management, San Diego
Master of Social Welfare (MSW), 2018
UCLA Luskin School of Public Affairs, Los Angeles
Bachelor of Science (BSc) in Biology, 2013
Cal Poly San Luis Obispo
Skills
Analytics tools / Languages
Python
R
Generative AI
SQL
Cloud Computing (AWS, GCP)
DashBoarding (R-Shiny, Streamlit)
Domain
Electronic Healthcare Records
Graph Theory
Recommendation Systems
Device Data
Claims Data
Social Media Data
Experience
Manager, Data Science, Real World Evidence
Becton Dickinson
Data Scientist
Becton Dickinson
Automated Analytics Solutions:
- Created scalable models/analyses for HEOR, Global Medical Affairs, Operations, and Business Units.
- Developed network models to rank and recommend Key Opinion Leaders (KOL) using a variety of datasets.
- Developed machine learning algorithm in to stratify patients into a daily infection risk ranking tool.
- Built tools to identify competitors and optimize clinical studies.
- Utilized AI tools to streamline code development and enhance quality.
EHRs/Real World Data:
- Leverage RWD insights using admissions, microbiology, laboratory, prescription, procedures, clinical notes, financial data from multiple EHRs sources for MDR submissions, to support clinical and HEOR scientific outputs, and improve operations execution.
- Developed analytical processes and data pipelines using cloud computing to extract and distill EHRs information.
- Applied statistical methods and deep learning (GANs, VAEs) to synthesize patient data and created Python interfaces.
Leadership:
- Managed and mentored associates and student teams on advanced modeling tools.
- Led KOL project as technical product owner, project manager, and lead data scientist.
- Presented project progress to VP-level executives and provided technical mentorship.
- Provide technical mentorship to junior data scientist at BD.
- Collaborated with data architects, engineers, clinical specialists, and product owners.
- Help establish BD Data Science Community of Practice by giving presentations and connect colleagues together.
Awards:
- Recipient of two Exceptional Impact awards (12/2023 & 12/2022).
- Recipient of MedA award in Innovation (2022). The annual MedA Awards recognize individuals and teams that have made substantial contribution to the growth and stature of BD over the past year.
Business Intelligence Developer/Data Analyst 3
Becton Dickinson
- Lead development of interactive and automated analytical solutions using Infusion data in R and SQL for usage by R&D and marketing stakeholders.
- Spearhead development of an anomaly detection model to be used in identifying data sets of high integrity and quality.
- Pioneered an R&D model in Python and R to predict when a medical device will malfunction.
- Collaborated with data architects, clinical specialists and reliability engineers on analytical projects.
- Provided code reviews using version control tools (e.g., git repositories, azure devops).
- Utilized CRISP-DM to document and communicate project development with technical, leadership, and business stakeholders.
Data Science Associate
Repurpose.AI
- Conducted cheminformatics research to repurpose drugs for COVID-19.
- Utilized KNIME, keras, tensorflow, and GCP for data mining and analytics.
- Wrote SQL queries and processed training sets from public sources (e.g., BigQuery, ChEMBL, PubChem, DrugBank).
- Performed feature engineering, feature reduction, and model building to identify drug candidates.
- Conducted biological, chemical, and patent literature searches and prepared documentation.
Post-masters Fellow
VA San Diego Healthcare System/UCSD
- Provided psychotherapy and rehabilitation services to Veterans with severe mental illnesses.
- Administered clinical measures and interpreted data for treatment plans.
- Led research on nicotine and learning in Veterans with SMI; presented findings at a symposium.
Graduate Research Assistant
UCLA
- Reviewed evaluation and approval documents by the FDA of two psychotropic medications to study how experimental drug activity established in preclinical animal model trials warrants human clinical trials.
- Updated the CriticalThinkRx web-based curriculum on psychotropic medications.
- Participated in evaluation of the Psychiatric Medication History Protocol.
Projects
*
Leveraged RWD insights using EHR data to support HEOR scientific output.
Exploring the Relationship Between Nicotine and Learning Among Veterans with Severe Mental Illness
During a post-masters fellowship at the VA San Diego Healthcare System/UCSD, I led an independent research project exploring relationship between nicotine and learning among Veterans with SMI. Presented findings at 2019 Judd Young Investigators Symposium.
Conducted clinical research by participating in the evaluation of the Mindful Self-Compassion and Mindfulness Based Stress Reduction interventions with Veterans.
User-Defined Real Time Text Mining Automation Tool
Pilot project of a relational database and build an informative interface that incorporates historical prescription of Medicare transaction, psychotropic drug, corresponding diagnoses and patient’s reviews of psychotropic drugs.
Healthcare Investments and Length of Hospital Stay Executive Summary Purpose
The purpose of this report is to visualize Healthcare Investment and Length of Hospital Stay per Country using interactive data tables and plotly.
Create an account on https://www.ncbi.nlm.nih.gov/myncbi/. Once registered, click on your email to navigate to account settings. Scroll down to API Key Management. Create an API key to increase your e-utils limit to 10 requests/second.