I lead advanced analytics initiatives to discover insights from clinical trials data using statistical, machine learning and AI approaches to make informed decisions.

In the Development and Discovery team, my role commences at the frontier where conventional analytical efforts reach their limits. Leveraging data from both early and late-phase clinical trials, I address the critical, often million-dollar questions posed by our stakeholders. My foundation in statistics and research empowers me to thoughtfully apply machine learning techniques and explainable AI, uncovering insights that ignite fresh perspectives and inspire innovation. These are what we term "true insights"—the kind that provide definitive answers to those pivotal, high-stakes inquiries. Python is my go to language, supplemented by R in specific scenarios utilizing Databricks computing environment, ensuring a comprehensive and nuanced approach to data analysis.

I oversee clinical and biomedical research activities through strategic leadership, mentorship, and hands-on participation in clinical studies. The collaboration has led to over 20 peer-reviewed publications in high-impact journals including The Lancet, The Orphanet Journal of Rare Diseases, Autism, PLOS One, among others.- Serve as Co‑PIs in clinical studies- Lead protocol development, IRB document preparation, writing and reviewing Statistical Analysis Plans (SAPs)- Supervise mid level Biostatisticians and Data Scientists and coordinate their workflows to ensure accuracy and timely completion of final deliverable- End‑to‑end project management, taking research initiatives, from gathering requirements, and refining the hypothesis to reporting and publications- Mentor and train early‑career clinicians and statisticians about biomedical and public health research methodology

My role is twofold: statistical SME, and data product engineering. The end products support pharmaceutical companies and HEOR researchers utilizing real-world data (RWD) for real-world evidence (RWE) generation in oncology.I own several disease-specific real-world analytic data products. I define/update business rules, write SQL scripts to build the pipeline, and maintain them in collaboration with data engineers.I write analytic data product planning documents and define product roadmaps based on internal stakeholder meetings.Take a multimodal approach, including machine learning for missing-value imputation and evaluating what-if scenarios to increase fill rates.Collaboratively maintain the data product manuals providing sample codes and example use cases

I supported AstraZeneca in their Oncology Business Unit (OBU) for real-world evidence (RWE) generation using real-world data (RWD). **Tools/platform:** Amazon Redshift, RStudio, RMarkdown, Dataiku (for both R, and Python), Jupyter notebook, PostgreSQL SQLWithin the OBU, I supported HEOR researchers in their feasibility studies, protocol-driven studies, and ongoing research works with real-world data/evidence. This involves requirements gathering through rounds of discussions, providing minimal viable product for buying in from the stakeholders, and delivering solutions per protocolUtilized **Flatiron** (clinical, EMR, Genomic), **Tempus**, and **CancerLinQ** for evidence generation in breast cancer and non-small cell lung cancer (NSCLC) studies. Used Instant Health Data (IHD) Analytics platform and cBioPortal for cancer genomics analysisUtilized RStudio to connect to Amazon Redshift database to access these datasets, join multiple tables, wrangle to create the analysis data sets, fit survival models (statistical model for time-to-event data) and generate reports using Rmarkdown for sharing with the line-manager and stakeholdersUsed Dataiku Cloud Data Science Platform for handling larger data sets and to create automated analytic pipelines as per the project's needs

**Tools/platform:** Python, PySpark, R, Jupyterhub, AWS, Domino, Hadoop, Hive/HiveQL, LookerI partnered with the stakeholders to understand, refine, and translate business questions into analytically solvable problems. I provided strategic guidance and subject-matter expertise on choosing statistical methods and development, validation, and use of algorithms for data-based decision-making. Performed product evaluations using **matched case-control studies** and provided actionable insights drawn from claims data to impact pricing decisions toward affordability and cost-savingUsed statistical and machine learning tools for **causal inference** using observational dataHandled and analyzed a massive amount of medical and claims data efficiently using Python tools. These data sets are gigabyte to terabyte size. The miniature data set had over half a million observations.Developed and maintained an analytic pipeline that was **6x more efficient** (from data aggregation to analytical data set creation) that fed data to multiple data science teams involved with different product evaluationsCreated and maintained a Python library for running a matched case-control study that was **10x faster than the existing tool.** For example, the process took 10 minutes instead of 2+ hours for a use case involving ~2 million observationsPerformed QC checks for functions written in Python and provided guidance and recommendations for improved efficiency

As an embedded data scientist for the Department of Surgery, I design clinical studies, contribute to grant writing, and perform analytics using advanced statistical and machine learning algorithms to improve patient outcomes. The goal is to improve quality of research and increase funding potentials for resident physicians and clinical investigators.

- Work with clinicians to design and analyze clinical studies, calculate sample size, write manuscripts- Coordinate with program directors to translate business problems into analytic projects- Provide end to end solutions from problem formulation to delivery of products to leadership - Extract data from disparate sources to create analysis data sets- Create analysis plan, write code, validate and test results- Build and validate models for prediction

-Taught graduate level programming with R, and SAS (basic and advanced programming, R functions, SAS macro, SAS/SQL, ODS)- Taught graduate level Design of Experiments course using SAS and R- Taught graduate level multivariate analysis (Logistic regression, Segmentation, Cluster Analysis, Discriminant analysis, dimensionality reduction with PCA, SVD, and Factor Analysis).

Assistant Professor of Environmental StatisticsDepartment of Natural and Applied SciencesTaught undergraduate courses in Mathematical Statistics, Intro to StatisticsMember of admissions committee in the Environmental Science and Policy graduate programServed on masters-level thesis and project committees and provided statistical consultation and design of experiments and analysis of data.

Data Analyst works with the epidemiologist to assist with management of large health and geographic datasets particularly those related to chronic diseases and disease risk factors.This position supports the data quality framework for epidemiology, research and surveillance programs and oversees the implementation and maintenance of a number of public health information systems.Created technical report while working with epidemiologistsManaged and analyzed large administrative health datasets and created data summaries for stakeholdersPrepared internal reports by analyzing complex survey data (Canadian Community Health Survey, CCHS) Organized and participated meetings with stakeholders and regional partnersSupervised and monitored performance of a research assistant

Taught undergraduate statistics courses in medium and large classes with diverse learning needs; developed course materials, prepared and conducted tests and evaluation materials.

Completed coursework on Bayesian statistics, Experimental Design, Statistical InferenceWrote an R package on shrinkage estimation in linear models. See codes here: https://github.com/raheems/r-packages As a consultant at the Center for Statistical Consulting, University of Windsor, provided consultations on the following projects:Tilbury District Family Health Team, Ontario, Canada Designed code-plan and prepared data entry form Cleaned and analyzed data using SAS Produced summary results through cross tables and graphicsCanadian National Survey on Child Restraint Use Designed efficient data entry forms, Suggested analysis plan Hospital Patients’ Falls data, Windsor, ON, Canada Cleaned and prepared raw data for analysis, Suggested analysis plans.Conducted methodological research and published peer reviewed articles in computational journals (see below for a list of publications)

Taught undergraduate and graduate courses in Applied Statistics, service to the department and community, student supervision, collaborative research.