Applied Machine Learning for Healthcare and Life Sciences Using AWS: Transformational AI implementations for biotech, clinical, and healthcare organizations

Description

Build real-world artificial intelligence apps on AWS to overcome challenges faced by healthcare providers and payers, as well as pharmaceutical, life sciences research, and commercial organizations Key FeaturesLearn about healthcare industry challenges and how machine learning can solve themExplore AWS machine learning services and their applications in healthcare and life sciencesDiscover practical coding instructions to implement machine learning for healthcare and life sciencesBook Description While machine learning is not new, it's only now that we are beginning to uncover its true potential in the healthcare and life sciences industry. The availability of real-world datasets and access to better compute resources have helped researchers invent applications that utilize known AI techniques in every segment of this industry, such as providers, payers, drug discovery, and genomics. This book starts by summarizing the introductory concepts of machine learning and AWS machine learning services. You'll then go through chapters dedicated to each segment of the healthcare and life sciences industry. Each of these chapters has three key purposes -- First, to introduce each segment of the industry, its challenges, and the applications of machine learning relevant to that segment. Second, to help you get to grips with the features of the services available in the AWS machine learning stack like Amazon SageMaker and Amazon Comprehend Medical. Third, to enable you to apply your new skills to create an ML-driven solution to solve problems particular to that segment. The concluding chapters outline future industry trends and applications. By the end of this book, you'll be aware of key challenges faced in applying AI to healthcare and life sciences industry and learn how to address those challenges with confidence. What you will learnExplore the healthcare and life sciences industryFind out about the key applications of AI in different industry segmentsApply AI to medical images, clinical notes, and patient dataDiscover security, privacy, fairness, and explainability best practicesExplore the AWS ML stack and key AI services for the industryDevelop practical ML skills using code and AWS servicesDiscover all about industry regulatory requirementsWho this book is for This book is specifically tailored toward technology decision-makers, data scientists, machine learning engineers, and anyone who works in the data engineering role in healthcare and life sciences organizations. Whether you want to apply machine learning to overcome common challenges in the healthcare and life science industry or are looking to understand the broader industry AI trends and landscape, this book is for you. This book is filled with hands-on examples for you to try as you learn about new AWS AI concepts. Table of ContentsIntroducing Machine Learning and the AWS Machine Learning StackExploring Key AWS Machine Learning Services for Healthcare and Life SciencesMachine Learning for Patient Risk StratificationUsing Machine Learning to Improve Operational Efficiency for Healthcare ProvidersImplementing Machine Learning for Healthcare PayorsImplementing Machine Learning for Medical Devices and Radiology ImagesApplying Machine Learning to GenomicsApplying Machine Learning to Molecular DataApplying Machine Learning to Clinical Trials and PharmacovigilanceUtilizing Machine Learning in the Pharmaceutical Supply ChainUnderstanding Common Industry Challenges and SolutionsUnderstanding Current Industry Trends and Future Applications Cover Title Page Copyright and Credits Contributors About the reviewers Table of Contents Preface Part 1: Introduction to Machine Learning on AWS Chapter 1: Introducing Machine Learning and the AWS Machine Learning Stack What is ML? Supervised versus unsupervised learning ML terminology Exploring the ML life cycle Problem definition Data processing and feature engineering Model training and deployment Introducing ML on AWS Introducing the AWS ML stack Summary Chapter 2: Exploring Key AWS ML Services for Healthcare and Life Sciences Applying ML in healthcare and life sciences Healthcare providers Healthcare payors Medical devices and imaging Pharmaceutical organizations Genomics Introducing AWS AI/ML services for healthcare and life sciences Introducing Amazon Comprehend Medical Introducing Amazon Transcribe Medical Introducing Amazon HealthLake Other AI services for healthcare and life sciences Summary Further reading Part 2: Machine Learning Applications in the Healthcare Industry Chapter 3: Machine Learning for Patient Risk Stratification Technical requirements Understanding risk stratification Identifying at-risk patients Implementing ML for patient risk stratification Introducing SageMaker Canvas Implementing ML for breast cancer risk prediction Importing your dataset Building the model Analyzing the model Predicting from the model Summary Chapter 4: Using Machine Learning to Improve Operational Efficiency for Healthcare Providers Technical requirements Introducing operational efficiency in healthcare Automating clinical document processing in healthcare Working with voice-based applications in healthcare Building a smart medical transcription application on AWS Creating an S3 bucket Downloading the audio file and Python script Running the application Summary Further reading Chapter 5: Implementing Machine Learning for Healthcare Payors Technical requirements Introducing healthcare claims processing Implementing ML in healthcare claims processing workflows Introducing Amazon SageMaker Studio SageMaker Data Wrangler SageMaker Studio notebooks Building an ML model to predict Medicare claim amounts Acquiring the data Feature engineering Building, training, and evaluating the model Summary Chapter 6: Implementing Machine Learning for Medical Devices and Radiology Images Technical requirements Introducing medical devices Classes of medical devices Understanding SaMD Introducing radiology imaging system components Applying ML to medical devices and radiology imaging Introducing Amazon SageMaker training Understanding the SageMaker training architecture Understanding training options with SageMaker training Building a medical image classification model using SageMaker Acquiring the dataset and code Summary Part 3: Machine Learning Applications in the Life Sciences Industry Chapter 7: Applying Machine Learning to Genomics Technical requirements Introducing genomic sequencing Categorizing genomic sequencing stages Looking into the evolution of genomic sequencing Challenges with processing genomic data Storage volume Sharing, access control, and privacy Compute Interpretation and analysis Applying ML to genomic workflows Introducing Amazon SageMaker Inference Understanding real-time endpoint options on SageMaker Understanding Serverless Inference on SageMaker Understanding Asynchronous Inference on SageMaker Understanding batch transform on SageMaker Building a genomic and clinical NER application Acquiring the genomic test report Understanding the pre-trained genetic entity detection model Summary Further reading Chapter 8: Applying Machine Learning to Molecular Data Technical requirements Understanding molecular data Small molecules Large molecules Introducing drug discovery and design Understanding structure-based drug design Applying ML to molecular data Molecular reaction prediction Molecular property prediction Molecular structure prediction Language models for proteins and compounds Introducing custom containers in SageMaker Adapting your container for SageMaker training Adapting your container for SageMaker inference Building a molecular property prediction model on SageMaker Running the Jupyter Notebook Summary Further reading Chapter 9: Applying Machine Learning to Clinical Trials and Pharmacovigilance Technical requirements Understanding the clinical trial workflow Understanding RWE studies Introducing PV Signal generation Hypothesis testing Applying ML to clinical trials and PV Literature search and protocol design Trial participant recruitment Adverse event detection and reporting Real-world data analysis Introducing SageMaker Pipelines and Model Registry Defining pipeline and steps Caching pipelines Introducing the SageMaker Model Registry Building an adverse event clustering model pipeline on SageMaker Running the Jupyter notebooks Reviewing the pipeline and model Deploying model and running inference Summary Chapter 10: Utilizing Machine Learning in the Pharmaceutical Supply Chain Technical requirements Understanding the pharmaceutical supply chain landscape Drug manufacturers Distributors or wholesalers Consumers Summarizing key challenges faced by the pharmaceutical supply chain industry Introducing pharmaceutical sales Applying ML to the pharmaceutical supply chain and sales Targeting providers and patients Forecasting drug demand and sales outlook Implementing predictive maintenance Understanding market sentiment and completive analysis Introducing Amazon Forecast Amazon Forecast algorithms Importing a dataset Training forecasting models Generating forecasts Building a pharmaceutical sales forecasting model using Amazon Forecast Acquiring the dataset Running the Jupyter notebook Summary Part 4: Challenges and the Future of AI in Healthcare and Life Sciences Chapter 11: Understanding Common Industry Challenges and Solutions Technical requirements Understanding challenges with implementing ML in healthcare and life sciences Healthcare and life sciences regulations Security and privacy Bias and transparency in ML Reproducibility and generalizability of ML models Understanding options for solving these challenges Encrypting and anonymizing your data Explainability and bias detection Building reproducible ML pipelines Auditability and review Introducing SageMaker Clarify and Model Monitor Detecting bias using SageMaker Clarify Pre-training bias metrics Post-training bias metrics Explaining model predictions with SageMaker Clarify Monitoring models with SageMaker Model Monitor Detecting bias and explaining model predictions for healthcare coverage amounts Acquiring the dataset Running the Jupyter notebooks Viewing bias and explainability reports in SageMaker Studio Summary Chapter 12: Understanding Current Industry Trends and Future Applications Key factors influencing advancements of AI in healthcare and life sciences Availability of multimodal data Active learning with human-in-the-loop pipelines Democratization with no-code AI tools Better-performing infrastructure and models Better focus on responsible AI Understanding current industry trends in the application of AI for healthcare and life sciences Curing incurable diseases such as cancer Telehealth and remote care Using the Internet of Things (IoT) and robotics Simulations using digital twins Surveying the future of AI in healthcare Reinforcement learning Federated learning Virtual reality Blockchain Quantum computing Concluding thoughts Summary Index Other Books You May Enjoy