Today, tens of thousands of customers, including leading healthcare and life sciences (HCLS) organizations such as GE Healthcare, Cerner, Bristol Myers Squibb, Roche, and more, use Amazon SageMaker for machine learning (ML). The HCLS industry faces mounting pressure to deliver more personalized treatments, streamline processes, modernize every aspect of the pharma value chain, and keep patient information private and secure. ML addresses these challenges by automatically identifying anomalies in medical images such as X-rays, building personalized healthcare treatment plans based on historical data and documents, and identifying suspicious healthcare claims so HCLS organizations can offer higher quality, more holistic treatment at lower costs. SageMaker enables patients, providers, payers, and researchers to prepare, build, train, and deploy high quality ML models and offers built in solutions to get started with ML faster.
Top use cases for Amazon SageMaker
Extract and analyze data from documents
To make decisions faster, healthcare and life sciences organizations need to understand text in medical documents, such as patient forms. With Amazon SageMaker, you can build ML models to automatically extract, process, and analyze data from handwritten and electronic documents so you can process documents faster and more accurately. SageMaker provides built-in ML algorithms that are optimized for text classification, natural language processing (NLP), and optical character recognition (OCR), that you can readily use to train and deploy models, or you can use Amazon SageMaker Autopilot to automatically generate text processing models.
To keep patient data secure, it is important for healthcare and life sciences organizations to use fraud detection models to spot suspicious healthcare claims before they impact customers. With Amazon SageMaker, you can build ML models to detect suspicious transactions before they occur and alert your customers in a timely fashion. SageMaker provides built-in ML algorithms, such as Random Cut Forrest and XGBoost, that you can use to train and deploy fraud detection models. In addition, SageMaker provides a set of solutions for fraud detection that can be deployed with just a few clicks.
Healthcare and life sciences organizations continue to look for ways to automatically identify anomalies and accelerate patient diagnosis. With Amazon SageMaker, you can build computer vision models to spot anomalies in medical images and automatically flag for deep analysis and diagnosis. SageMaker provides a broad set of capabilities purpose-built for machine learning including built-in algorithms that are optimized for computer vision, such as image classification, that can improve the diagnosis of patients, reduce the subjectivity in diagnosis, and help save time for pathologists.
Disease understanding and drug development can be tedious and time-consuming, and life sciences companies constantly look for ways to accelerate the drug development process. With Amazon SageMaker, you can easily label training data for a variety of use cases so you can speed the time to train and deploy highly accurate ML models. By automating this tedious work using SageMaker, life sciences companies can accelerate the drug discovery process, introduce medicines to market more rapidly, and deliver life-changing medicines that contribute value to patients and society.
With Amazon SageMaker and Amazon EC2 P3 Instances, Celgene (now part of Bristol Myers Squibb) accelerated time-to-train models and productivity, allowing them to focus on groundbreaking research and discovery.
GE Healthcare is transforming healthcare by delivering better outcomes for providers and patients. Amazon SageMaker allows GE Healthcare to access powerful artificial intelligence tools and services to advance improved patient care.
AWS is helping Novartis transform its manufacturing process by unifying access to all information and enabling Novartis to make quick and informed critical decisions. They are using Amazon SageMaker to build a computer vision-based model that will determine line clearance.
Propeller Health applies ML with solutions such as Amazon SageMaker and Amazon Redshift, along with its infrastructure built on AWS, to give patients a forecast of their health based on local weather conditions, recent medication use, and other factors.
Product defect detection in images
Differential privacy for sentiment classification
Document summarization, entity, and relationship extraction
Handwriting recognition with Amazon SageMaker
Filling in missing values in tabular records
Flagging suspicious healthcare claims with Amazon SageMaker
Creating a model for predicting orthopedic pathology using Amazon SageMaker
Building predictive disease models using Amazon SageMaker with Amazon HealthLake normalized data