Skip to content
View makotoy56's full-sized avatar

Achievements

Achievement: Pull Sharkx2

Achievements

Achievement: Pull Sharkx2

Block or report makotoy56

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
makotoy56/README.md

Makoto Yoshida, PhD

Publication-backed biomedical researcher applying evidence-generation discipline to clinical and healthcare analytics projects.

My professional foundation is biomedical and translational research. This GitHub profile is a public evidence hub for how I extend that evidence-generation discipline into clinical analytics: EHR cohort construction, exposure and outcome definition, statistical modeling, reproducible reporting, and cautious interpretation of observational data.

Best-Fit Review Paths

This portfolio is most relevant for:

  • Clinical Data Analytics
  • Healthcare Data Analytics
  • Research Data Analytics
  • Public Health Analytics
  • RWE-adjacent observational clinical analytics

Featured Portfolio Projects

Start here Best reviewer use What it demonstrates Links
Non-ICU RAAS Mortality Analysis Broad clinical and healthcare analytics review Large-scale MIMIC-IV hospital EHR cohort, early medication exposure definition, logistic regression, adjusted predicted risks, average marginal effects, selected SAS-Python validation Repo / Report
COPD ICU RAAS Survival Analysis Deeper clinical analytics and survival-analysis review ICU COPD cohort definition, pre-ICU exposure timing, time-to-event mortality, Kaplan-Meier curves, Cox proportional hazards modeling, diagnostics, sensitivity analysis, selected SAS-Python validation Repo / Report
Public Health Statistics Workflow Supporting analytics, reporting, and visualization review Descriptive public-health workflow, age-adjusted regression, forest plots, exploratory interaction checks, Quarto reporting, GitHub Pages deployment Repo / Report

Recommended Review Order

  1. Start with the Non-ICU RAAS Mortality Analysis for the broadest clinical EHR analytics example.
  2. Review the COPD ICU RAAS Survival Analysis for time-to-event modeling, ICU clinical reasoning, diagnostics, and sensitivity-analysis depth.
  3. Review the Public Health Statistics Workflow for reproducible reporting, age adjustment, forest plots, and public-health analytics communication.

What This Portfolio Demonstrates

  • Clinical and healthcare analytics workflows built as portfolio evidence.
  • MIMIC-IV EHR cohort, exposure, covariate, and outcome construction.
  • SQL and BigQuery workflows for analysis-ready clinical datasets.
  • Python and Jupyter statistical analysis workflows.
  • Logistic regression, adjusted predicted risk, and average marginal effects.
  • Kaplan-Meier and Cox proportional hazards survival analysis.
  • Sensitivity analysis, model diagnostics, and limitation-aware interpretation.
  • Selected SAS-Python validation checks.
  • Quarto reporting, GitHub Pages publication, and reproducible documentation.
  • Evidence-boundary discipline grounded in publication-backed biomedical research.

Evidence Scope

These repositories are portfolio projects, not employment history. The MIMIC-IV findings are observational and hypothesis-generating, so they should not be interpreted as causal treatment-effect evidence.

The profile emphasizes documented evidence: publication-backed biomedical research, EHR analytics projects, reproducible reporting, and selected validation workflows.

Research Foundation

Before building this analytics portfolio, my primary foundation was publication-backed biomedical and translational research. That work supports the same habits this portfolio emphasizes: question framing, endpoint definition, data interpretation, scientific writing, collaborator communication, and careful limitation handling.

For publication records and research identifiers, see ORCID.

Additional Evidence

I also have resume-backed experience evaluating AI-generated biomedical, technical, analytical, and Japanese/English content for factuality, reasoning quality, intent alignment, and language quality using rubric-based review workflows.

Data Governance and Reproducibility

The MIMIC-IV clinical projects use de-identified data under PhysioNet access requirements. No patient-level source data, PHI, credentials, or restricted datasets are included in the repositories.

The repositories are designed for public review of code, documentation, aggregate outputs, figures, reports, and reproducibility structure. Reproduction of MIMIC-IV analyses requires independent PhysioNet approval, appropriate data access, and local environment configuration.

Technical Stack

  • SQL / BigQuery
  • Python
  • pandas / NumPy
  • statsmodels / lifelines
  • Jupyter
  • Selected SAS-Python validation workflows
  • Quarto
  • Git / GitHub
  • GitHub Pages

Contact

Pinned Loading

  1. mimic-iv-copd-raas-analysis mimic-iv-copd-raas-analysis Public

    EHR-based observational survival analysis of ICU patients with COPD using MIMIC-IV. Evaluates the association between pre-ICU RAAS inhibitor exposure and in-hospital mortality using time-to-event m…

    Jupyter Notebook

  2. mimic-iv-nonicu-medication mimic-iv-nonicu-medication Public

    EHR-based observational analysis of adult non-ICU hospital admissions using MIMIC-IV. Evaluates early RAAS inhibitor exposure and in-hospital mortality with multivariable logistic regression, absol…

    Jupyter Notebook

  3. public-health-statistics-workflow public-health-statistics-workflow Public

    Reproducible public health statistics workflow with descriptive epidemiology, age-adjusted regression, forest plots, and Quarto reporting.

    Jupyter Notebook