AI for Healthcare with Keras and Tensorflow 2.0: Design, Develop, and Deploy Machine Learning Models Using Healthcare Data

Table of Contents
About the Author
About the Technical Reviewers
Introduction
Chapter 1: Healthcare Market: A Primer
	Different Stakeholders of the Healthcare Marketplace
		Regulators
			Food and Drug Administration (FDA)
			Center for Medicare and Medicaid Services (CMS)
			Center for Medicare and Medicaid Innovation (CMMI)
		Payers
		Providers
	Regulation of Healthcare Information
	AI Applications in Healthcare
		Screening
		Diagnosis
		Prognosis
		Response to Treatment
	What Is the Industry Landscape?
	Conclusion
Chapter 2: Introduction and Setup
	Introduction to TensorFlow 2
		TensorFlow Core
		TensorFlow JS
		TensorFlow Lite
		TensorFlow Extended
	TensorFlow 1.x vs 2.x
		What Is TF 1.x?
		Embracing TF 2.x
			Eager Execution
			AutoGraph
			TensorFlow Datasets
			tf.keras
			Estimators
		Recommendations for Best Use
	Installation and Setup
		Python Installation
		Using the Virtual Environment
		Library and Versions
			TensorFlow and GPU
			Others
	Conclusion
Chapter 3: Predicting Hospital Readmission by Analyzing Patient EHR Records
	What Is EHR Data?
	MIMIC 3 Data: Setup and Introduction
		Access
		Introduction and Setup
	Data
		Social and Demographic
		Admissions Related
		Patient’s Clinical Data
		Lab Events
		Comorbidity Score
	Modeling for Patient Representation
		A Brief Introduction to Autoencoders
		Feature Columns in TensorFlow
		Creating an Input Pipeline Using tf.data
		Creating Feature Columns
		Building a Stacked Autoencoder
	Cohort Discovery
		What Is an Ideal Cohort Set?
		Optimizing K-Means Performance
		Deciding the Number of Clusters by Inertia and Silhouette Score Analysis
		Checking Cluster Health
	Multitask Learning Model
		What Is Multitask Learning ?
		Different Ways to Train a MTL Model
		Training Your MTL Model
	Conclusion
Chapter 4: Predicting Medical Billing Codes from Clinical Notes
	Introduction
	Data
		NOTEEVENTS
		DIAGNOSES_ICD
	Understanding How Language Modeling Works
		Paying Attention
		Transforming the NLP Space: Transformer Architecture
			Positional Encoding
			Multi-Head Attention
		BERT: Bidirectional Encoder Representations from Transformers
			Input
				Token Embeddings
				Segment Embeddings
			Training
				Masked Language Modeling
				Next-Sentence Prediction
	Modeling
		BERT Deep-Dive
			What Does the Vocabulary Actually Contain?
		Training
	Conclusion
Chapter 5: Extracting Structured Data from Receipt Images Using a Graph Convolutional Network
	Data
		Mapping Node Labels to OCR Output
		Node Features
		Hierarchical Layout
			Line Formation
			Graph Modeling Algorithm
		Input Data Pipeline
	What Are Graphs and Why Do We Need Them?
	Graph Convolutional Networks
		Convolutions over Graph
		Understanding GCNs
		Layer Stacking in GCNs
		Training
	Modeling
		Train-Test Split and Target Encoding
		Creating Flow for Training in StellarGraph
		Training and Model Performance Plots
	Conclusion
Chapter 6: Handling Availability of Low-Training Data in Healthcare
	Introduction
		Semi-Supervised Learning
			GANs
			Autoencoders
		Transfer Learning
		Weak Supervised Learning
	Exploring Snorkel
	Data Exploration
		Introduction
		Labeling Functions
			Regex
			Syntactic
			Distance Supervision
	Pipeline
		Writing Your LFs
			Working with Decorators
			Preprocessor in Snorkel
		Training
		Evaluation
		Generating the Final Labels
	Conclusion
Chapter 7: Federated Learning and Healthcare
	Introduction
		How Does Federation Learning Work?
		Types of Federated Learning
			Horizontal Federated Learning
			Vertical Federated Learning
			Federated Transfer Learning
	Privacy Mechanism
		Secure Aggregation
		Differential Privacy
	TensorFlow Federated
		Input Data
		Custom Data Load Pipeline
		Preprocessing Input Data
		Creating Federated Data
		Federated Communications
		Evaluation
	Conclusion
Chapter 8: Medical Imaging
	What Is Medical Imaging?
		Image Modalities
		Data Storage
	Dealing with 2-D and 3-D Images
		Handling 2-D Images
			DICOM in Python
			EDA on DICOM Metadata
				View Position
				Age
				Sex
				Pixel Spacing
				Mean Intensity
		Handling 3-D Images
			NIFTI Format
			Introduction to MRI Image Processing
			Non-Even Pixel Distribution
			Correlation Test
				Cropping and Padding
	Image Classification on 2-D Images
		Image Preprocessing
			Histogram Equalization
			Isotropic Equalization of Pixels
		Model Creation
		Preparing Input Data
		Training
	Image Segmentation for 3-D Images
		Image Preprocessing
			Bias Field Correction
			Removing Unwanted Slices
		Model Creation
		Preparing Input Data
		Training
		Performance Evaluation
	Transfer Learning for Medical Images
	Conclusion
	References
Chapter 9: Machines Have All the Answers, Except What’s the Purpose of Life
	Introduction
	Getting Data
	Designing Your Q&A
		Retriever Module
			Query Paraphrasing
			Retrieval Mechanics
				Term/Phrase-Based
				Semantic-Based
			Reranking
		Comprehension
			BERT for Q&A
			Fine-Tuning a Q&A Dataset
	Final Design and Code
		Step 0: Preparing the Document Data
		Step 1: BERT-QE Expansion
			Step 1.1: Extract the Top k Documents for a Query Using BM-25
			Step 1.2: Relevance Score on the Top 200 Documents
		Step 2: Semantic Passage Retrieval
		Step 3: Passage Reranking Using a Fine-Tuned Covid BERT Model on the Med-Marco Dataset
		Step 4: Comprehension
	Conclusion
Chapter 10: You Need an Audience Now
	Demystifying the Web
		How Does an Application Communicate?
		Cloud Technology
	Docker and Kubernetes
		Why Docker?
		OS Virtualization
		Kubernetes
	Deploying the QnA System
		Building a Flask Structure
			Deep Dive into app.py
			Understanding index.html
		Dockerizing Your Application
			Creating a Docker Image
				Base Image and FROM Command
				COPY and EXPOSE
				WORKDIR, RUN, and CMD
				Dockerfile
			Building Docker Image
		Making It Live Using Heroku
	Conclusion
Index