Nanodegree key: nd320
Version: 1.0.0
Locale: en-us
The course is focused on using AI technologies for practical applications in health care.
Content
Part 01 : Welcome to the AI for Healthcare Nanodegree Program
-
Module 01: AI for Healthcare Nanodegree Program Orientation
Part 02 : Applying AI to 2D Medical Imaging Data
-
Module 01: 2D Imaging
-
Lesson 01: Introduction to AI for 2D Medical Imaging
In this lesson, you will be given an introduction to this course about AI for 2D medical imaging, why AI is important and where AI fits in the space.
- Concept 01: Meet Your Instructor
- Concept 02: Course Outline
- Concept 03: Lesson Outline
- Concept 04: Introduction to AI for 2D Medical Imaging
- Concept 05: History
- Concept 06: Importance of AI
- Concept 07: When to Use AI
- Concept 08: Business Stakeholders
- Concept 09: Project: Pneumonia Detection from X-Ray Images
- Concept 10: Recap
- Concept 11: Good Luck
-
Lesson 02: Clinical Foundations of 2D Medical Imaging
In this lesson, we will cover clinical foundations such as clinical workflows, applications of 2D imaging in clinical settings and how machine learning impacts clinics.
- Concept 01: Clinical Foundations
- Concept 02: Lesson Outline
- Concept 03: Intuition About 2D Medical Images
- Concept 04: Clinical Applications
- Concept 05: Clinical Applications Quizzes
- Concept 06: Clinical Applications Exercise
- Concept 07: Clinical Applications Exercise Solution
- Concept 08: Apply Machine Learning
- Concept 09: Apply Machine Learning Quizzes
- Concept 10: Apply Machine Learning Exercise
- Concept 11: Apply Machine Learning Exercise Solution
- Concept 12: Performance of ML
- Concept 13: Performance of ML Quizzes
- Concept 14: Performance of ML Exercise
- Concept 15: Performance of ML Exercise Solution
- Concept 16: Regulatory Landscape
- Concept 17: Regulatory Landscape Quizzes
- Concept 18: Regulatory Landscape Exercise
- Concept 19: Regulatory Landscape Exercise Solution
- Concept 20: Final Review
- Concept 21: Lesson Conclusion
-
Lesson 03: 2D Medical Imaging Exploratory Data Analysis
In this lesson, we will learn the DICOM standard in medical imaging, and how to explore medical imaging data and prepare it for machine learning applications.
- Concept 01: DICOM Standard and EDA
- Concept 02: Lesson Outline
- Concept 03: DICOM Standard
- Concept 04: DICOM Standard Quizzes
- Concept 05: DICOM Standard Exercise
- Concept 06: DICOM Standard Exercise Solution
- Concept 07: Explore 2D Imaging Properties
- Concept 08: Explore 2D Imaging Properties Quizzes
- Concept 09: Explore 2D Imaging Properties Exercise
- Concept 10: Explore 2D Imaging Properties Exercise Solution
- Concept 11: Prepare DICOM Images for ML
- Concept 12: Prepare DICOM Images for ML Quizzes
- Concept 13: Prepare DICOM Images for ML Exercise
- Concept 14: Prepare DICOM Images for ML Exercise Solution
- Concept 15: Exploring Population Metadata
- Concept 16: Exploring Population Metadata Quizzes
- Concept 17: Exploring Population Metadata Exercise
- Concept 18: Exploring Population Metadata Exercise Solution
- Concept 19: Edge Case
- Concept 20: Final Review
- Concept 21: Lesson Conclusion
-
Lesson 04: Classification Models of 2D Medical Images
In this lesson, we'll dive deep into classification tasks for 2D medical imaging using different machine learning models. and we will talk about pre-process data, train, test, and validate models.
- Concept 01: Models for Classification of 2D Medical Images
- Concept 02: Lesson Outline
- Concept 03: Big Picture
- Concept 04: Intuition about Classification Model Development
- Concept 05: Differentiate Between Models
- Concept 06: Differentiate Between Models Quizzes
- Concept 07: Differentiate Between Models Exercise
- Concept 08: Differentiate Between Models Exercise Solution
- Concept 09: Split Dataset for Model Development
- Concept 10: Split Dataset for Model Development Quizzes
- Concept 11: Split Dataset for Model Development Exercise
- Concept 12: Split Dataset for Model Development Exercise Solution
- Concept 13: Obtaining a Gold Standard
- Concept 14: Obtaining a Gold Standard Quizzes
- Concept 15: Obtaining a Gold Standard Exercise
- Concept 16: Obtaining a Gold Standard Exercise Solution
- Concept 17: Image Pre-processing for Model Training
- Concept 18: Image Pre-processing for Model Training Quizzes
- Concept 19: Image Pre-processing for Model Training Exercise
- Concept 20: Image Pre-processing for Model Training Exercise Solution
- Concept 21: Fine-tuning CNNs for Classification
- Concept 22: Fine-tuning CNNs for Classification Quizzes
- Concept 23: Fine-tuning CNNs for Classification Exercise
- Concept 24: Fine-tuning CNNs for Classification Exercise Solution
- Concept 25: Evaluating Your Model
- Concept 26: Evaluating Your Model Quizzes
- Concept 27: Evaluating Your Model Exercise
- Concept 28: Evaluating Your Model Exercise Solution
- Concept 29: Final Review
- Concept 30: Lesson Conclusion
-
Lesson 05: Translating AI Algorithms for Clinical Settings with the FDA
In this lesson, you will learn about how your work fits into the bigger picture, and how it’s regulated by the FDA, which is an often-overlooked, but incredibly important.
- Concept 01: Translating Algorithm into a Clinical Setting with the FDA
- Concept 02: Lesson Outline
- Concept 03: Big Picture
- Concept 04: Intended Use and Indications for Use
- Concept 05: Intended Use and Clinical Impact Quizzes
- Concept 06: Intended Use and Clinical Impact Exercise
- Concept 07: Intended Use and Clinical Impact Exercise Solution
- Concept 08: Algorithmic Limitations
- Concept 09: Algorithmic Limitations Quizzes
- Concept 10: Algorithmic Limitations Exercise
- Concept 11: Algorithmic Limitations Exercise Solution
- Concept 12: Translate Performance into Clinical Utility
- Concept 13: Translate Performance into Clinical Utility Quizzes
- Concept 14: Translate Performance into Clinical Utility Exercise
- Concept 15: Translate Performance into Clinical Utility Solution
- Concept 16: Designing an FDA Validation Plan
- Concept 17: Designing an FDA Validation Plan Quizzes
- Concept 18: Designing an FDA Validation Plan Exercise
- Concept 19: Designing an FDA Validation Plan Exercise Solution
- Concept 20: Final Review
- Concept 21: Lesson Conclusion
-
Part 03 : Applying AI to 3D Medical Imaging Data
-
Module 01: 3D Imaging
-
Lesson 01: Introduction to AI for 3D Medical Imaging
In this lesson, we will introduce the course and instructors. We will give you an overview of the context for AI in 3D medical imaging space, and cover the objectives of the course.
-
Lesson 02: 3D Medical Imaging - Clinical Fundamentals
In this lesson, we cover the basic terminology and concepts related to 3D medical imaging. We will look at the problem space from a clinical standpoint and learn how CT and MR scanners produce images.
- Concept 01: Introduction
- Concept 02: What are 3D Medical Images?
- Concept 03: Why use 3D Medical Images?
- Concept 04: Who uses 3D Medical Images?
- Concept 05: Example Clinical Scenario
- Concept 06: Exercise 1: Choosing a Clinical Problem
- Concept 07: Exercise 1: Solution
- Concept 08: Clinical Section Outro
- Concept 09: Physical Principles of CT Scanners
- Concept 10: Exercise 2: Computing a Sinogram
- Concept 11: Exercise 2: Solution
- Concept 12: CT Scanners: Summary
- Concept 13: Physical Principles of MR Scanners
- Concept 14: MRI: Gradients & RF Pulses
- Concept 15: MRI: K-space, Reconstruction, T1 and T2
- Concept 16: MR Scanners: Summary
- Concept 17: Common 3D imaging data tasks
- Concept 18: 3D Imaging Tasks: Windowing
- Concept 19: 3D Imaging Tasks: MPR
- Concept 20: 3D Imaging Tasks: 3D Reconstruction
- Concept 21: 3D Imaging Tasks: Registration
- Concept 22: Exercise 3: Volumetric Rendering
- Concept 23: Exercise 3: Solution
- Concept 24: 3D Imaging Tasks: Summary
- Concept 25: Lesson Summary
-
Lesson 03: 3D Imaging Exploratory Data Analysis
In this lesson, we will dive deeper into medical imaging formats NIFTI and DICOM, how scanner data is represented, and how to read medical volumes stored in these files and analyze them.
- Concept 01: Introduction
- Concept 02: The DICOM Standard
- Concept 03: Anatomy of DICOM Datasets
- Concept 04: NIFTI File Format
- Concept 05: Viewing 3D Medical Images
- Concept 06: Exercise 1: Load Image
- Concept 07: Important Parameters of 3D Medical Images
- Concept 08: Basic DICOM Volume EDA
- Concept 09: Exercise 2: 3D Volume MPR
- Concept 10: Basic 3D DICOM Dataset EDA
- Concept 11: Exercise 3: 3D Dataset EDA and Curation
- Concept 12: Lesson Summary
-
Lesson 04: 3D Medical Imaging - End-to-End Deep Learning Applications
In this lesson, we cover the basics of building deep neural networks for 3D medical imaging (mostly segmentation & classification) and performance evaluation from a software and clinical perspective.
- Concept 01: Introduction
- Concept 02: Classification
- Concept 03: Methods for Feature Extraction
- Concept 04: Exercise 1: Fun with Convolutions
- Concept 05: Classification: Summary
- Concept 06: Segmentation
- Concept 07: Segmentation Methods
- Concept 08: Exercise 2: Segmentation Hands On
- Concept 09: Creating Ground Truth For Segmentation
- Concept 10: Evaluating Performance as Data Scientist
- Concept 11: Evaluating Performance as a Clinician
- Concept 12: Exercise 3: Measuring Performance
- Concept 13: Lesson Summary and Looking Beyond
-
Lesson 05: Deploying AI Algorithms in Real World Scenarios
In this lesson, we'll talk about clinical networks, architecture, and AI deployment, tools and their use by data scientists and clinicians, as well as medical device regulation and data privacy.
- Concept 01: Introduction
- Concept 02: DICOM Networking: Introduction
- Concept 03: DICOM Networking: Services
- Concept 04: Clinical Networks
- Concept 05: Requirements for Integration
- Concept 06: Clinical Network Architecture Summary
- Concept 07: Tools of the Trade - Intro
- Concept 08: Tools of the Trade - Scripting
- Concept 09: Exercise 1: Scripting for DICOM Networks
- Concept 10: Tools of the Trade - A Radiologists in Action
- Concept 11: Tools of the Trade - Viewers - OHIF
- Concept 12: Tools of the Trade - Viewers - 3D Slicer
- Concept 13: Exercise 2: Creating Ground Truth with Slicer
- Concept 14: Tools of the Trade Summary
- Concept 15: Regulatory Landscape: Medical Devices
- Concept 16: Regulatory Landscape: FDA Process
- Concept 17: Regulatory landscape: HIPAA, Anonymization
- Concept 18: Exercise 3: Anonymization
- Concept 19: Lesson Summary
- Concept 20: The End and Final Project
-
Lesson 06: Hippocampal Volume Quantification in Alzheimer's Progression
In this project, you will curate a dataset of brain MRIs, train a segmentation on a CNN, and integrate this into a clinical network to quantify hippocampal volume for Alzheimer's progression.
Project Description - Hippocampal Volume Quantification in Alzheimer's Progression
Project Rubric - Hippocampal Volume Quantification in Alzheimer's Progression
- Concept 01: Project Overview
- Concept 02: Section 1: Curating a Dataset of Brain MRIs
- Concept 03: Workspace for Section 1
- Concept 04: Section 2: Training a Segmentation CNN
- Concept 05: Workspace for Section 2
- Concept 06: Section 3: Integrating into a Clinical Network
- Concept 07: Workspace for Section 3
- Concept 08: Closing Remarks
-
Part 04 : Applying AI to EHR Data
-
Module 01: Applying AI to EHR Data
-
Lesson 01: Applying AI to EHR Data Introduction
Introduction to the EHR Data course and the instructor.
-
Lesson 02: EHR Data Security and Analysis
Learn about the importance of data security and the different standards that apply to EHR. Also, learn about analyzing EHR data.
- Concept 01: EHR Data Security and Analysis Overview
- Concept 02: Importance of Data Privacy and Security
- Concept 03: Key Healthcare Data Security and Privacy Standards
- Concept 04: Storing and Accessing PHI Data
- Concept 05: PII Data Solution
- Concept 06: Importance of EDA
- Concept 07: Dataset Schema Analysis
- Concept 08: Value Distributions
- Concept 09: Missing Values and Outliers
- Concept 10: Analyzing Dataset for High Cardinality
- Concept 11: Analyze Dataset Exercise
- Concept 12: EDA Exercise Solution
- Concept 13: Demographic Analysis
- Concept 14: Compare Demographics Exercise
- Concept 15: Exercise Compare Demographics Solution
- Concept 16: EHR Data Security and Analysis Lesson Recap
-
Lesson 03: EHR Code Sets
In this lesson you will learn how to work with different EHR codes and how to map them properly to records.
- Concept 01: EHR Code Sets Overview
- Concept 02: Codes Sets Background
- Concept 03: Diagnosis Codes Part 1
- Concept 04: Diagnosis Codes Part 2
- Concept 05: Grouping codes Walkthrough
- Concept 06: Diagnosis Code Grouping Exercise
- Concept 07: Diagnosis Code Grouping Solution
- Concept 08: Procedure Codes
- Concept 09: Procedure Code Grouping Exercise
- Concept 10: Procedure Code Grouping Solution
- Concept 11: Medication Codes
- Concept 12: Grouping/Categorizing Systems
- Concept 13: Group or Categorize Procedures Exercise
- Concept 14: Group or Categorize Procedures Solution
- Concept 15: EHR Code Sets Recap
-
Lesson 04: EHR Transformations & Feature Engineering
In this lesson, you'll gain skills in feature engineering and transformation of EHR.
- Concept 01: EHR Transformations & Feature Engineering Overview
- Concept 02: EHR Dataset Levels
- Concept 03: Encounter Representation
- Concept 04: Longitudinal Representation
- Concept 05: How to Convert to Longitudinal Level
- Concept 06: Select Last Encounter Exercise
- Concept 07: Select Last Encounter Solution
- Concept 08: Dataset Splitting Without Data Leakage
- Concept 09: How to Split Dataset at Patient Level
- Concept 10: Data Splitting Exercise
- Concept 11: Data Splitting Solution
- Concept 12: ETL with TensorFlow Dataset API
- Concept 13: Numerical Features & Feature Column API
- Concept 14: Numerical Feature Column API Exercise
- Concept 15: Numerical Features Solution
- Concept 16: Categorical Features & Feature Column API
- Concept 17: Categorical Features Exercise
- Concept 18: Categorical Features with Feature Column API Solution
- Concept 19: Transformations and Feature Engineering Recap
-
Lesson 05: Building, Evaluating and Interpreting Models
In this final lesson, you'll be putting all of your skills together to build, evaluate and interpret ML models for Bias and Uncertainty.
- Concept 01: Building, Evaluating & Interpreting Models Overview
- Concept 02: Tensorflow Regression Model with DenseFeatures
- Concept 03: TF DenseFeatures Walkthrough
- Concept 04: Common EHR Model Evaluation Metrics
- Concept 05: Common EHR Model Evaluation Metrics Exercise
- Concept 06: Common EHR Model Evaluation Metrics Solution
- Concept 07: Demographic Bias Analysis
- Concept 08: Demographic Bias Analysis Walkthrough
- Concept 09: Demographic Bias Analysis Exercise
- Concept 10: Demographic Bias Analysis Solution
- Concept 11: Uncertainty Estimation
- Concept 12: Train Uncertainty Estimation Model Walkthrough
- Concept 13: Uncertainty Estimation Exercise
- Concept 14: Uncertainty Estimation Solution
- Concept 15: Model Interpretability with Shapley Values
- Concept 16: Building Models Recap
- Concept 17: AI and EHR Recap
-
Lesson 06: Project: Patient Selection for Diabetes Drug Testing
In this project students will use what they learn in the classroom to apply AI in healthcare for patient data.
Project Description - Patient Selection for Diabetes Drug Testing
Project Rubric - Patient Selection for Diabetes Drug Testing
-
Part 05 : Applying AI to Wearable Device Data
-
Module 01: Wearable Data
-
Lesson 01: Introduction to Wearable Data
We’ll cover what wearables are and the scope of the class. Learn who your instructor is and his thoughts on the promise and caveats of wearables in medical research and decision making.
-
Lesson 02: Intro to Digital Sampling & Signal Processing
A brief tour through sampling theory, signal processing, the Fourier transform, and other related topics. We’ll briefly cover some plotting and visualization techniques here as well.
- Concept 01: Introduction
- Concept 02: Refresher on Signals
- Concept 03: Sampling Analog Signals
- Concept 04: Plotting Signals In Time-domain
- Concept 05: Time-domain Plotting Continued
- Concept 06: Exercise 1: Plotting
- Concept 07: Exercise 1: Solution
- Concept 08: Interpolation
- Concept 09: Interpolation in Review
- Concept 10: Exercise 2: Interpolation
- Concept 11: Exercise 2: Solution
- Concept 12: Fourier Transform
- Concept 13: Fourier Transform In Practice
- Concept 14: Fourier Transform in Review
- Concept 15: Exercise 3: The Fourier Transform
- Concept 16: Exercise 3: Solution
- Concept 17: Plotting Signals in Frequency Domain
- Concept 18: Exercise 4: Spectrograms
- Concept 19: Exercise 4: Solution
- Concept 20: Harmonics
- Concept 21: Recap: Intro to Digital Sampling & Signal Processing
-
Lesson 03: Introduction to Sensors
We cover the basics of the accelerometer, the PPG sensor, and the ECG sensor, as well as what these signals look like in typical environments and the types of noise that we will encounter.
- Concept 01: Introduction to Sensors
- Concept 02: Inertial Measurement Unit
- Concept 03: Accelerometer Deep Dive
- Concept 04: Exercise 1: Step Cadence
- Concept 05: Exercise 1: Solution
- Concept 06: PPG Sensor
- Concept 07: Exercise 2: PPG Peaks
- Concept 08: Exercise 2: Solution
- Concept 09: PPG SNR
- Concept 10: Exercise 3: PPG SNR
- Concept 11: Exercise 3: Solution
- Concept 12: ECG Sensor
- Concept 13: Apple Heart Study - Revisited
- Concept 14: Recap: Introduction to Sensors
-
Lesson 04: Activity Classification
Build an activity classifier using a wrist-worn accelerometer!
- Concept 01: Intro to Activity Classifiers
- Concept 02: Data Exploration
- Concept 03: Exercise 1: Data Exploration
- Concept 04: Exercise 1: Solution
- Concept 05: Feature Extraction I
- Concept 06: Exercise 2: Feature Extraction
- Concept 07: Feature Extraction Continued...
- Concept 08: Activity Classification
- Concept 09: Hyperparameter Tuning & Regularization
- Concept 10: Cross-Validation and Feature Importance
- Concept 11: Hyperparameter Tuning in Review
- Concept 12: Exercise 3: A Quirk in the Dataset
- Concept 13: Exercise 3: Solution
- Concept 14: Recap: Activity Classification
-
Lesson 05: ECG Signal Processing
We deep dive into a fundamental algorithm for ECG processing and use that as the basis for an arrhythmia detection algorithm.
- Concept 01: Intro to ECG Signal Processing
- Concept 02: Heart Physiology
- Concept 03: Heart Physiology in Review
- Concept 04: Pan-Tompkins QRS Detection
- Concept 05: Pan-Tompkins In Code
- Concept 06: Exercise 1: Extend the Pan-Tompkins Algorithm
- Concept 07: Exercise 1: Solution
- Concept 08: Atrial Fibrillation
- Concept 09: Arrhythmia Detection: Dataset
- Concept 10: Arrhythmia Detection: Features
- Concept 11: Exercise 2: Arrhythmia Features
- Concept 12: Exercise 2: Solution
- Concept 13: Exercise 3: Atrial Fibrillation
- Concept 14: Exercise 3: Solution
- Concept 15: Recap: ECG Signal Processing
- Concept 16: Course Recap
-
Lesson 06: Motion Compensated Pulse Rate Estimation
In this project, you will create a pulse rate algorithm that takes into account activity and apply this algorithm to a new data set to determine clinically significant features.
Project Description - Motion Compensated Pulse Rate Estimation
-
-
Module 02: Career Services
-
Lesson 01: Take 30 Min to Improve your LinkedIn
Find your next job or connect with industry peers on LinkedIn. Ensure your profile attracts relevant leads that will grow your professional network.
- Concept 01: Get Opportunities with LinkedIn
- Concept 02: Use Your Story to Stand Out
- Concept 03: Why Use an Elevator Pitch
- Concept 04: Create Your Elevator Pitch
- Concept 05: Use Your Elevator Pitch on LinkedIn
- Concept 06: Create Your Profile With SEO In Mind
- Concept 07: Profile Essentials
- Concept 08: Work Experiences & Accomplishments
- Concept 09: Build and Strengthen Your Network
- Concept 10: Reaching Out on LinkedIn
- Concept 11: Boost Your Visibility
- Concept 12: Up Next
-