Syllabus 2017
Syllabus 2017
INSTRUCTOR
Oleg Pianykh, PhD
‐ Director of Medical Analytics Group, MGH Radiology
‐ Assistant Professor of Radiology, Harvard Medical School
Email: opianyATgmail.com (replace AT by @)
CLASS SCHEDULE
Fall 2017, as posted
COURSE OBJECTIVES
The course applies the tools of big data analytics to datasets and processes found in a modern hospital. While the
data has been captured for decades, we are only starting to mine it for information, discovering the invaluable
knowledge it was hiding. It has become imperative to use this data to drive health care improvements. You will
learn to build models of complex health care practices, and apply bid data techniques in clinical environments.
The course is designed for people in a wide range of backgrounds, but will require your time and attention.
LEARNING OUTCOMES
Learning how to:
Design, run and optimize complex healthcare processes
Perform healthcare big‐data research
Develop and implement data‐driven healthcare solutions
COURSE TOPICS
Introduction: Data‐driven healthcare
Data sources:
Digital healthcare standards – how they work, and why they make sense
HIS, RIS, PACS. IHE and workflow integration
Healthcare operations and management:
Working with operational data: benefits and pitfalls
Patient flow and scheduling optimization
Converting big data patterns to rational management
Clinical modeling and performance optimization.
Healthcare business intelligence:
Big data platforms and tools
Big data or big garbage?
Healthcare analytics: why and how?
Clinical decision support:
Clinical decision support
Strategies and gaming in healthcare. Forcing healthcare improvements.
Medical imaging and data visualization:
Working with digital imaging data: acquisition, diagnostic display, enhancement and analysis
Imaging the invisible: Computed tomography
From billions of pixels to a solid diagnosis: CAD and advanced diagnostic image processing
Complex data visualization
Medical data exchange
Networking and telemedicine
Security and confidentiality in medicine
Clinical software development
Clinical software development. Medical startups
Bringing data applications to a hospital: disaster or success?
Fault‐tolerance, scalability, and robustness of data‐driven apps
Unusual problems: pharaohs, criminals, and pure art
COURSE PREREQUISITES
Basic programming skills, such as CSCI e‐10A or the equivalent, and STATS e‐100 or the equivalent.
We will be using Matlab for 70% of our homework – if you do not know Matlab yet, check online tutorials
on www.mathworks.com or YouTube.
Good understanding of information technology (hardware, software, networking)
Course grade is based on homework, class participation, and final exam.
TEXTBOOKS
Required
None
Recommended
Trevor L. Strome , Healthcare Analytics for Quality and Performance Improvement
Oleg S. Pianykh, Digital Imaging and Communications in Medicine (DICOM): A
Practical Introduction and Survival Guide, Springer. Second edition
Oleg S. Pianykh, Digital Image Quality in Medicine (Understanding Medical
Informatics)
CLASS RULES
Homework – must be submitted on time, or it won’t be graded. All work must be your own, so start early.
Individual work only.
Cell phones, beepers, gadgets, food ‐ outside.