Deep Learning Exhibits ❯ RSNA 2019

A Radiologist’s Guide to Deep Learning and Artificial Intelligence: What You Need to Know for the Road Ahead

 

 

A Radiologist’s Guide to Deep Learning and Artificial Intelligence: What You Need to Know for the Road Ahead

Sara Raminpour
Lilly Kauffman
Hannah Ahn
Elliot K. Fishman, MD

Russell H. Morgan Department of Radiology and Radiological Science, Johns Hopkins Medicine

 

 

Deep Learning for Radiologists: A Beginner's Guide is a free website initially developed in March 2018 as a part of CTisus.com. It started growing widely and in February 2019 was established as a separate website with 8 different categories.
This interactive website, designed to optimize the user experience, is dedicated to explaining deep learning for practicing radiologists and meeting their unique needs .
Deep Learning for Radiologists: A Beginner's Guide presents rich resources to enhance radiologists’ knowledge on various aspects of deep learning and AI in medicine and beyond.

 

 

Deep Learning for Radiologists: A Beginner's Guide provides all deep learning materials in one place including pearls, journal clubs, lectures, NVIDIA resources, and more.
This mobile friendly website – a work in progress as it is continually updated – offers almost daily changes in some sections and monthly updates in other sections, like pearls and journal clubs, using a wide range of resources.
Deep Learning for Radiologists: A Beginner's Guide has had over 1,000 users in the last 3 months and currently averages 60 page views per day.
This presentation includes a review of each section of the website.

 

 

A Radiologist’s Guide to Deep Learning and Artificial Intelligence: What You Need to Know for the Road Ahead

 

Introduction to Deep Learning

  • Deep learning is used in the research community and in industry to help solve many big data problems such as computer vision, speech recognition, and natural language processing.
  • Deep learning is a subset of artificial intelligence (AI) that uses multiple layers of algorithms and data analysis methods to develop a mathematical model.
  • Deep learning helps medical researchers to better explore medical data to find more effective disease treatments.
  • Deep learning may enhance radiologists’ skills to analyze medical images and achieve earlier detection of some deadly diseases such as pancreatic cancers.

 

Introduction to Deep Learning for Radiologists: a Beginner’s Guide

This guideline:
  1. Provides radiologists with the key concepts of AI through a series of learning modules that include lectures, feature articles, pearls, and literary reviews.
  2. Provides radiologists with basic knowledge on deep learning and AI.
  3. Helps radiologists understand the current state of the art and how it may impact clinical practice.
  4. Helps radiologists keep up to date with applications pending FDA approval and nearing use in clinical practice.

 

Presentation Contents

  1. Pearls
  2. Journal Club
  3. Lectures
  4. Glossary
  5. NVIDIA Resources for Radiology
  6. AI in the News
  7. Exhibits
  8. The Felix Project: A Lustgarten Initiative

 

 

A Radiologist's Guide to Deep Learning and Artificial Intelligence: What You Need to Know for the Road Ahead

 

Pearls

Pearls for radiologists include insights or facts that can help them make the correct diagnosis or better understand an image. Just like case studies, pearls represent a classic part of radiology education.
In this section we present 24 items on various deep learning topics such as AI, machine learning, human interface, FDA regulations, radiomics, deep learning roles in anatomical regions such as the GI tract, liver, pancreas, brain, etc. These entries are filled with pearls from the literature or our imaging experiences, which can enhance users’ understanding of deep learning and AI.

 

Pearls

Pearls

 

Pearls

Pearls

 

Journal Club

The Journal Club page includes notable articles in radiology and non-radiology journals on topics related to deep learning and AI. We include summaries of these feature articles and highlight key points.
In this section we also present 24 items on various topics such as AI, machine learning, human interface, FDA regulations, radiomics, deep learning roles in anatomical regions such as the GI tract, liver, pancreas, brain, etc. from the medical and scientific literature.

 

Journal Club

Journal Club

 

Journal Club

Journal Club

 

Lectures

As in other branches of medicine, the lecture is one of the classic modes of education in radiology. The evolution of podcasting, and then vodcasting, has had a major impact on education.
The Lectures page offers discussions on current topics in AI and deep learning in radiology such as The Early Detection of Pancreatic Cancer Using Deep Learning: Preliminary Observations.
Each video consists of explanations of multiple education slides, images, CT scans, and graphs.

 

Lectures

Lectures

 

Lectures

Lectures

 

Glossary

The Glossary features definitions for the numerous terms related to deep learning that have been provided by NVIDIA®.
This section provides medical and non-medical terms, explaining abbreviations such as MRI, PET, and CT, as well as concepts including angiography, framework, nuclear medicine, and even a scientific definition for deep learning.
  • Deep learning: is a subfield of machine learning concerned with algorithms inspired by the structure and function of the brain called artificial neural networks. A deep neural network (DNN) is an artificial neural network (ANN) with multiple hidden layers of units between the input and output layers. Similar to shallow ANNs, DNNs can model complex non-linear relationships.

 

Glossary

Glossary

 

NVIDIA® Resources for Radiology

NVIDIA® Resources for Radiology presents a series of over 100 lectures on topics of interest to radiologists.
This page includes articles that mainly focus on the role of AI and deep learning in healthcare and radiology such as:
  • How will AI Change Healthcare?
  • How AI is Advancing the Fight Against Breast Cancer 
  • How an AI App Can Translate a Photo into a Skin Cancer Diagnosis 
  • Artificial intelligence in radiology: Hype or hope?
  •  Heart Smart: How HeartFlow Uses AI to Detect Heart Disease 
And more…

 

NVIDIA® Resources for Radiology

NVIDIA Resources for Radiology

 

AI in the News

The AI in the News section provides articles on a range of topics related to deep learning within and beyond medicine.
This page includes two sections:
  • Latest Posts: presents all articles that are usually posted within the previous 30 days, beginning with the most recent.
  • Posts Archive: presents all previously posted articles in 15 different categories based on the role of AI in different topics of medicine as well as beyond medicine, including the following subjects:
    • AI and Cancer Detection
    • AI and Computer Science/Vision
    • AI and Drug Discovery
    • AI and FDA Approvals
    • AI and Medical Care
    • AI and Radiology
    • AI and Social Good/Ethics
    • And more…

 

AI in the News

AI in the News

 

AI in the News

AI in the News

 

AI in the News

AI in the News

 

Exhibits

The Exhibits page includes PowerPoint presentations on deep learning provided for self-learning.
This section features the best CT exhibits on AI that have been presented at the RSNA and ARRS annual meetings. Visitors to this section of the website will be able to navigate through each presentation and view each slide individually.
Currently, this page provides two exhibits (see below) from RSNA 2018 and will soon offer presentations from RSNA 2019:

 

Exhibits

Exhibits

 

Exhibits

Exhibits

 

Exhibits

Exhibits

 

Exhibits

Exhibits

 

Exhibits

Exhibits

 

Exhibits

Exhibits

 

The Felix Project: a Lustgarten Initiative

  • The Felix Project, funded by the Lustgarten Foundation, is a multidisciplinary research collaboration developed to use deep learning to automatically find the pancreas and pancreatic ductal adenocarcinoma (PDAC) – the third most common cause of cancer death in the US – on CT scans.
  • The purpose of the Felix Project is to design software that can detect and identify early-stage pancreatic cancers, also including pancreatic neuroendocrine tumors (PNET), which may save lives from this silent killer disease.
  • The Felix team comprises experts in medical imaging, pathology, oncology, cancer research, and computer science.
This page includes Scientific Publications & Presentations, which features articles generated by this ambitious project, and In the News, which gives access to interviews conducted with the Felix team on various news stations .

 

The Felix Project: a Lustgarten Initiative

The Felix Project

 

Disclosure

  • The Lustgarten Foundation funds the Felix Project. Elliot K. Fishman, MD is the Principal Investigator of the Felix Project and receives grant support from the Lustgarten Foundation for this project.
  • The NVIDIA® lectures are free to all users on the web and on the NVIDIA® company site, https://www.nvidia.com/en-us/, as well.

 

Conclusion

Deep Learning for Radiologists: A Beginner's Guide has been developed to enhance users’ knowledge of deep learning and AI and its important role in medicine, radiology, and beyond. This exhibit presents deep learning basics for radiologists and a guide to how to use this rapidly growing website.
To access Deep Learning for Radiologists: A Beginner's Guide visit: https://www.ctisus.com/responsive/deep-learning/default.asp.
© 2020 Elliot K. Fishman, MD, FACR
All Rights Reserved.
www.CTISUS.com
CTisus CT Scanning CTisus CT Scanning CTisus CT Scanning CTisus CT Scanning CTisus CT Scanning CTisus CT Scanning CTisus CT Scanning CTisus CT Scanning CTisus CT Scanning