Tutorial Information on 17 MAY, ICMHI 2024
Time: 14:00-17:00
Registration: 50USD fee for each participant
Deadline: April 15, 2024
The Multimodal Learning for Electronic Medical Records Cookbook
This tutorial is intended to cover the needs and interests of
researchers and analysts who want to build prediction models with
multiple types of clinical data, such as tabular data, time-series
data, and others. The main targeted participants of this tutorial
are clinical data analysts and other personnel who have a chance to
access EHR or clinical data.
With the advent of EHRs, leveraging big data analytics has become
indispensable for advancing clinical research. Enriched EHRs harbor
vital insights into disease progression, offering a wealth of
information for treatment selection and disease diagnosis. Despite
this, most studies predominantly rely on a single type of clinical
data, limiting the full utilization of EHRs' rich information.
This instructional guide aims to bridge this gap by introducing
multimodal learning mechanisms, enabling the construction of
predictive models using Python and Google Colab. The tutorial
unfolds in three parts:
EHR Overview and Standards:
In this initial section, we delve into Electronic Health Records
(EHRs) and explore relevant standards, providing a foundation drawn
from state-of-the-art research papers.
Building Models with Multimodal Learning:
The second segment focuses on fundamental concepts and strategies
for constructing models using multiple types of clinical data.
Practical Implementation:
In the third and final section, participants will engage in a
hands-on experience, walking through the step-by-step process of
building models with multimodal learning.
By the end of this tutorial, you will be equipped with the knowledge
and skills to leverage a diverse array of clinical data types,
revolutionizing the way predictive models are developed and utilized
in healthcare research. Let's embark on this transformative journey
together!
#About the lecturers:
Yi-Ju Tseng, PhD
Yi-Ju Tseng is an associate professor at National Yang Ming
Chiao Tung with extensive experience in claims data and electronic
medical records analysis and machine learning. Tseng's work focuses
on improving infection surveillance by using informatics techniques
and applying machine learning techniques to infectious disease
diagnosis and prognosis prediction. Tseng received the MOST Young
Scholar Fellowship and Special Outstanding Talent Award from the
Ministry of Science and Technology, Taiwan. Her research interests
include medical informatics, public health informatics, clinical
decision support, data science, and infection surveillance.