Skip to content

Latest commit

 

History

History
158 lines (95 loc) · 13.7 KB

README.md

File metadata and controls

158 lines (95 loc) · 13.7 KB
Raw

Educational Informatics Resource For Pathologists and Other Healthcare Professionals.

1. Pathology Informatics Essentials for Residents, also known as (PIER.) Available during pathology residency.

2. Clinical Informatics Fellowship Programs. Clinical Informatics became a medical subspecialty on Sept. 22, 2011.

3. Pathology Informatics Training Programs.

  • Henry Ford Hospital and Health System
  • University of Pittsburgh
  • Partners HealthCare- Massachusetts General, Brigham and Women's, Northshore Medical Center
  • Michigan Medicine- University of Michigan

4. Beginner's Learning Resources for Programming, Data Science, Bioinformatics, and Machine Learning.

The web is full of resources but it can be a daunting place to start for someone with little or no computer science background such as medical students, residents, pathologists, and other physicians. This is meant to be a gentle and informative guide to get started.

1. Pathology Informatics Essentials for Residents (PIER)

Usually taken during the pathology residency/fellowship. PIER is a research-based instructional resource developed by the Association of Pathology Chairs, Association for Pathology Informatics, and College of American Pathologists that presents training topics, implementation strategies and resource options for PRODS and faculty to effectively provide informatics training to their residents and meet ACGME informatics milestone requirements. PIER is also an effective resource for aspiring specialists to develop prerequisite pathology informatics knowledge and skills prior to advanced training or fellowships.

Image Source

2. Clinical Informatics Fellowship Programs Are Available to Pathologists and Other Physicians in All 24 Specialties.

There are over 30 ACGME-accredited Clinical Informatics Fellowship Programs (8 based in departments of pathology currently).

Rationale for the Clinical Informatics Subspecialty (source)

Successful health information system implementation depends in large measure on the knowledge and skills of the individuals who design, integrate, and implement these systems. While clinical informatics is multi-disciplinary in nature, there is a particular need for physicians who understand the care process, informatics concepts, and information technology. Creation of the formal subspecialty will help to standardize key elements of clinical informatics training programs and will likely increase the number of training opportunities. Further, it will provide an immediately recognized credential for organizations seeking to hire clinical informaticians. Finally, this initiative is consistent with the current national emphasis on strengthening the health information technology workforce.

Clinical Informatics fits within the spectrum of biomedical informatics as depicted by the American Medical Informatics Association (https://www.amia.org/)

3. Pathology Informatics Fellowships are also available which long predate the Clinical Informatics Fellowships (since mid 1990's).

These may or may not be accredited by the ACGME as clinical informatics fellowships, but through 2022, completing two years of training in a non-accredited PI program will qualify fellows to sit for the CI boards.

  • Henry Ford Hospital and Health System
  • University of Pittsburgh
  • Partners HealthCare- Massachusetts General, Brigham and Women's, Northshore Medical Center
  • Michigan Medicine- University of Michigan  

4. Beginner's Learning Resources for Programming, Data Science, Bioinformatics, and Machine Learning for Pathologists and Other Healthcare Professionals without Computer Science Background.

There are enormous amount of resources online to choose from. One can read anything and everything online and take classes but not everyone, especially healthcare professionals, will have the necessary time to do so. Therefore, to learn effectively, a concise outline and map to navigate through these materials would be of importance.

Why do healthcare professionals need to get involved in deep learning and artificial intelligence (i.e. informatics)?

In his 2018 JAMA article, Geoffrey Hinton gives health care professionals an intuitive understanding of the technology underlying deep learning with potential to transform health care. In accompanying JAMA article, David Naylor, a physician, outlines some of the factors propelling adoption of deep learning and artificial intelligence technology in medicine and health care.

From the perspective of the field of pathology and laboratory medicine, computational pathology is the natural discipline that are being advocated due to advances in high-throughput laboratory and health information technologies. Louis DN et al. states in their article "Pathologists, who are at the nexus of diagnostic data, models of disease pathogenesis, and clinical correlation, are ideally positioned to provide leadership in the emerging “big data” era of medical care." Joseph Sirintrapun, MD, opines in his 2018 ASCP article that "Computational pathology extends on pathology informatics in leveraging innovative tools of deep learning and software development, the ramification of which requires informatics leadership to build and guide computational teams experienced in leveraging such modern tools". Before computational pathology (or even training in computational pathology) can succeed, pathologists and especially residents must decide to engage in, and build careers around, informatics and/or computation. The full article titled "Computational Pathology: A Path Ahead" can be found here.

 

Learning path resources will be divided into the following sections (A-G):

For starters, attending "data carpentry" workshops available near your university is a great way to start! Workshops are usually a 1-2 day hands-on instructor-taught tutorial. Their initial target audience is learners who have little to no prior computational experience. These are great for learning basic essentials before starting the learnng path. There are two main ones now known as "The Carpentries". These classes are archived and freely accessible online here.

Data Carpentry develops and teaches workshops on the fundamental data skills needed to conduct research.

Software Carpentry teach three core topics: the Unix shell, version control with Git, and a programming language (Python or R)

Computer Science 101 (Stanford Online) is a short course that teaches the essential ideas of computer science for a zero prior-experience audience.

A. Knowledge of Linux and Command Line Operations

B. Python and R for Data Science

C. Basics of Mathematics & Statistics

The real prerequisite for informatics and machine learning is not about mastering mathematics, it's about data engineering and analysis.

An efficient way is to learn topic by topic, essential concepts first.

  • Linear Algebra and Matrix Multiplication
  • Probability and Stastistics
  • Multivariate calculus
  • Optimization Theory

Most of the common informatics and machine learning libraries and tools take care of the advanced mathematics for you. However, knowing the basic underlying mathematics and statistics will allow the student to maximize his or her informatics and machine learning capacity.

  • Khan Academy is a great resource for a targeted learning in areas you feel weakest.

  • 3Blue1Brown is one of the most popular resource on YouTube Channel for teaching hard-to-understand mathematical concepts behind machine learning such as backpropagation, gradient descent, neural network.

  • Basic Statistics (coursera)

  • Essential Math for Machine Learning: Python Edition (edX)

  • Essential Math for Machine Learning: R Edition (edX)

  • Statistical Learning (Stanford Online)

  • Mathematics for Machine Learning Specialization (coursera)

D. Machine Learning and Deep Learning

E. Advanced Deep Learning Topics

F. Introduction to Bioinformatics

G. Bioinformatics Repositories for DNA, RNA, and Microbiome Analysis

For Whole Genome, Whole Exome, and Targeted Sequencing Pipeline:

For RNAseq Pipeline:

For Microbiome Analysis Pipeline: