JSChoi.org
J. S. Choi, MD is a
physician of internal medicine and
clinical informatics.
GitHub account
Education
Indiana University
Clinical Informatics Fellowship
2021–
Kettering Medical Center
Internal Medicine Residency
2018–2021
Saint Louis University College of Medicine
Doctor of Medicine
2012–2018
University of Arizona College of Medicine
Bachelor of Science
2008–2012
Major in Physiology • Double minor in Computer Science and Chemistry.
Summa cum laude. GPA 3.98.
Research and development
2018–
Proposal for new Hack-style pipe operator for the
JavaScript programming language, under consideration by
ECMA Technical Committee 39. The
Hack pipes proposal was published on GitHub.
Proposal for body-part emoji:
Neck, chest, abdomen, back, thyroid, stomach, intestine, liver,
pancreas, kidney, bladder, uterus, prostate, breast, muscle, skin,
and antibody
2020–
A formal proposal to encode seventeen new body organs and regions as
standardized emoji was written and is being submitted to the Unicode
Consortium. Each proposed emoji character was analyzed for factors of
inclusion, including web-search query frequency and prevalence of related
medical disorders. The
proposal and artwork for the proposed emoji characters
are publicly available under an open Creative Commons license. The
document will be under consideration by the Unicode Consortium Emoji
Subcommittee, and new emoji characters may be standardized as soon as
2021.
Text mining educational material for internal medicine
Kettering Medical Center
2020–
The competent physician of general medicine acquires and retains large
volumes of biomedical, clinical, epidemiological, and sociobehavioral
declarative knowledge. All levels of medical education—undergraduate,
graduate, postgraduate, and continuing—therefore involve the memorization,
review, and testing of this required knowledge. However, the body of
published literature on which much required knowledge is based is vast and
continues to grow, making difficult the creation of educational review
material with which to review and test the required knowledge. Text mining
open-access biomedical literature offers a potential solution.
A study is being performed in which (1) an open-source ML model is trained
to recognize textual mentions of declarative core knowledge for competent
internal-medicine physicians and (2) an open-source web application is
being developed that extracts educational review material, from
open-access biomedical literature, regarding that declarative core
knowledge.
Practice guidelines of BPSD pharmacologic management: a review
with George T. Grossberg, MD
• SLU SOM Department of Neurology & Psychiatry
2016–2017
A systematic literature review was performed over
recently published practice guidelines on the pharmacologic management of
behavioral and psychiatric symptoms of dementia.
Patient-data displays: visual presentation and design
with David Burroughs, PhD
• SLU Center for Outcomes Research
2015–2016
This project compiled existing information on the
visual design of patient-data displays. Information
included methods of presentation and visualization of patient data to
healthcare providers and how these methods affect
clinical decision making, diagnosis, medical errors, and medical
outcomes.
Meaningful-use standards for pediatric electronic health record systems
with Feliciano B. Yu, Jr., MD
• Washington University in St. Louis • St. Louis Children’s Hospital
2013
This project consolidated several standards of meaningful-use requirements
for electronic medical records’ user interfaces into a single database. It
focused on requirements from the
Children’s EHR Format, HL7, and NIST 7865 standards.
Patient electronic-health-record programmatic data analysis
with Allan Wachter, MD
2011–2012
Data analysis was performed on
longitudinal patient data on behalf of A. Wachter in an
retrospective study for predicting asthma with the United
States Navy. A computer program was developed to
clean, normalize, and combine heterogenous patient data
from several independent sources: the US Navy, the US Department of
Veterans Affairs, and several pharmaceutical companies.
FnParse
2008–2012
Clojure computer-programming library used by other developers to parse
text into data structures using grammars that the programmer creates.
FnParse was published on GitHub.
Skills and Interests
Technology skills
UNIX command-line administration, programming, and data processing.
Experience in multiple programming languages: JavaScript / ECMAScript
(ES), Clojure (i.e., Lisp), Java, HTML5, SVG, and other
XML, and CSS; capable of learning new languages quickly. Methods of data
visualization (cf. Edward Tufte) and methods of
interactively exploring data (cf. Bret Victor).
Hobbies and interests
- Technology standards
-
Independent contributor to discussions of several standards bodies
including the WHATWG (HTML5), ECMA Technical Committee 39 (JavaScript
/ ECMAScript), the W3C (CSS, Web Annotations), and the Unicode
Consortium. Also interested in health-informatics standards such as
HL7.
-
Student Member of the
Unicode Consortium, 2016–2017.
-
Contributor to
ECMA Technical Committee 39,
2018.
- Public health and education
- Evidence- and science-based medicine
-
Measuring and improving medical outcomes through educational aids and
data visualizations
- Clinical decision making, usability, design
-
Relationships between clinical decision making, cognitive psychology,
usability, and user-interface design
- Testing and improving medical computer interfaces
- Computer science and engineering
- Programming-language design
- Text parsing
- Relational data modeling
- Data-interchange formats
- Licensed amateur-radio operator
- Arts
- Digital typography
- Violin, piano, and music composition
- Interactive multimedia