AI for healthcare with Keras and Tensorflow 2.0 : design, develop, and deploy machine learning models using healthcare data / Anshik.

Format:

Book

Author/Creator:

Anshik, author.

Language:

English

Subjects (All):

Artificial intelligence--Medical applications.

Artificial intelligence.

Physical Description:

1 online resource (391 pages)

Place of Publication:

[Place of publication not identified] : Apress, [2021]

Summary:

Learn how AI impacts the healthcare ecosystem through real-life case studies with TensorFlow 2.0 and other machine learning (ML) libraries. This book begins by explaining the dynamics of the healthcare market, including the role of stakeholders such as healthcare professionals, patients, and payers. Then it moves into the case studies. The case studies start with EHR data and how you can account for sub-populations using a multi-task setup when you are working on any downstream task. You also will try to predict ICD-9 codes using the same data. You will study transformer models. And you will be exposed to the challenges of applying modern ML techniques to highly sensitive data in healthcare using federated learning. You will look at semi-supervised approaches that are used in a low training data setting, a case very often observed in specialized domains such as healthcare. You will be introduced to applications of advanced topics such as the graph convolutional network and how you can develop and optimize image analysis pipelines when using 2D and 3D medical images. The concluding section shows you how to build and design a closed-domain Q&A system with paraphrasing, re-ranking, and strong QnA setup. And, lastly, after discussing how web and server technologies have come to make scaling and deploying easy, an ML app is deployed for the world to see with Docker using Flask. By the end of this book, you will have a clear understanding of how the healthcare system works and how to apply ML and deep learning tools and techniques to the healthcare industry. What You Will Learn * Get complete, clear, and comprehensive coverage of algorithms and techniques related to case studies * Look at different problem areas within the healthcare industry and solve them in a code-first approach * Explore and understand advanced topics such as multi-task learning, transformers, and graph convolutional networks * Understand the industry and learn ML Who This Book Is For Data scientists and software developers interested in machine learning and its application in the healthcareindustry

Contents:

Intro

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

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

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?

Chapter 5: Extracting Structured Data from Receipt Images Using a Graph Convolutional Network

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

Train-Test Split and Target Encoding

Creating Flow for Training in StellarGraph

Training and Model Performance Plots

Chapter 6: Handling Availability of Low-Training Data in Healthcare

Semi-Supervised Learning

GANs

Autoencoders

Transfer Learning

Weak Supervised Learning

Exploring Snorkel

Data Exploration

Labeling Functions

Regex

Syntactic

Distance Supervision

Pipeline

Writing Your LFs

Working with Decorators

Preprocessor in Snorkel

Evaluation

Generating the Final Labels

Chapter 7: Federated Learning and Healthcare

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

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

Image Segmentation for 3-D Images

Bias Field Correction

Removing Unwanted Slices

Performance Evaluation

Transfer Learning for Medical Images

References

Chapter 9: Machines Have All the Answers, Except What's the Purpose of Life

Getting Data

Designing Your Q&amp

A

Retriever Module

Query Paraphrasing

Retrieval Mechanics

Term/Phrase-Based

Semantic-Based

Reranking

Comprehension

BERT for Q&amp

Fine-Tuning a Q&amp

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

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

Index.

Notes:

Description based on print version record.

ISBN:

9781484270868

148427086X

OCLC:

1260347533