Times are displayed in (UTC-07:00) Pacific Time (US & Canada) Change
11/9/2024 |
8:30 AM – 12:00 PM |
Continental Ballroom 6
W03: Designing machine learning algorithms and analytics to support interdisciplinary clinical subspecialties
Presentation Type: Workshop/Tutorial
Designing machine learning algorithms and analytics to support interdisciplinary clinical subspecialties
Presentation Time: 08:30 AM - 12:00 PM
Abstract Keywords: Machine Learning, Data Mining, Clinical Decision Support, Healthcare Quality, Large Language Models (LLMs)
Primary Track: Applications
Mainstream analytical products and machine learning (ML) algorithms were designed to support traditional medical specialties and clinical environments, i.e. oncology, cardiology, anesthesiology, and quality improvement. However, modern medicine evolved into complex multispecialty and multidisciplinary environments where culture shifted towards comprehensive care delivered by physicians specializing and subspecializing in various diseases and treatments, i.e. cardio-oncology, onco-anesthesiology, neuro-orthopedics. These latest medical subspecialties require analytical solutions to (1) identify eligible patients, (2) link patients and providers via referrals, (3) assess and predict risks, and (4) visualize data to support operational decision making. Development of analytical algorithms to support emerging medical subspecialties represents a paradigm shift in traditional ML approaches in medicine.
We identify targeted informatics training in analytical support for enhancing the delivery of multidisciplinary medicine as the current gap, and propose a workshop focused on cardio-oncology as an example of an emerging subspecialty. Cardio-oncology is a new subspecialty of cardiology that treats cardiotoxicity.
Our workshop will deliver value to participants based on 3 major strengths:
1. Avant-garde clinical use case in analytics for interdisciplinary medicine.
2. Development of hands-on data analysis and ML algorithm design skills, using tools widely utilized in academic and patient care settings.
3. Application of the skills to a synthetic data set derived from real patient data.
The workshop will include hands-on utilization of the cloud-based and/or open-sourced ML tools, variables selection for predictive algorithms, and mining OMOP data. We will partner with participants to design a working ML model via full development cycle from clinical definition to design and implementation.
Speaker(s):
Jacob Krive, PhD
University of Illinois at Chicago
Randi Foraker, PhD, MA, FAHA, FAMIA, FACMI
Institute for Informatics, Data Science, and Biostatistics at Washington University School of Medicine in St. Louis
Adam Wilcox, PhD
Washington University in St. Louis
Sam Al-Droubi, M.S.
Author(s):
Presentation Time: 08:30 AM - 12:00 PM
Abstract Keywords: Machine Learning, Data Mining, Clinical Decision Support, Healthcare Quality, Large Language Models (LLMs)
Primary Track: Applications
Mainstream analytical products and machine learning (ML) algorithms were designed to support traditional medical specialties and clinical environments, i.e. oncology, cardiology, anesthesiology, and quality improvement. However, modern medicine evolved into complex multispecialty and multidisciplinary environments where culture shifted towards comprehensive care delivered by physicians specializing and subspecializing in various diseases and treatments, i.e. cardio-oncology, onco-anesthesiology, neuro-orthopedics. These latest medical subspecialties require analytical solutions to (1) identify eligible patients, (2) link patients and providers via referrals, (3) assess and predict risks, and (4) visualize data to support operational decision making. Development of analytical algorithms to support emerging medical subspecialties represents a paradigm shift in traditional ML approaches in medicine.
We identify targeted informatics training in analytical support for enhancing the delivery of multidisciplinary medicine as the current gap, and propose a workshop focused on cardio-oncology as an example of an emerging subspecialty. Cardio-oncology is a new subspecialty of cardiology that treats cardiotoxicity.
Our workshop will deliver value to participants based on 3 major strengths:
1. Avant-garde clinical use case in analytics for interdisciplinary medicine.
2. Development of hands-on data analysis and ML algorithm design skills, using tools widely utilized in academic and patient care settings.
3. Application of the skills to a synthetic data set derived from real patient data.
The workshop will include hands-on utilization of the cloud-based and/or open-sourced ML tools, variables selection for predictive algorithms, and mining OMOP data. We will partner with participants to design a working ML model via full development cycle from clinical definition to design and implementation.
Speaker(s):
Jacob Krive, PhD
University of Illinois at Chicago
Randi Foraker, PhD, MA, FAHA, FAMIA, FACMI
Institute for Informatics, Data Science, and Biostatistics at Washington University School of Medicine in St. Louis
Adam Wilcox, PhD
Washington University in St. Louis
Sam Al-Droubi, M.S.
Author(s):
W03: Designing machine learning algorithms and analytics to support interdisciplinary clinical subspecialties
Description
Date: Saturday (11/09)
Time: 8:30 AM to 12:00 PM
Room: Continental Ballroom 6
Time: 8:30 AM to 12:00 PM
Room: Continental Ballroom 6
