This can be a joint weblog with AWS and Philips.
Philips is a well being expertise firm centered on bettering individuals’s lives by significant innovation. Since 2014, the corporate has been providing prospects its Philips HealthSuite Platform, which orchestrates dozens of AWS providers that healthcare and life sciences firms use to enhance affected person care. It companions with healthcare suppliers, startups, universities, and different firms to develop expertise that helps medical doctors make extra exact diagnoses and ship extra personalised therapy for thousands and thousands of individuals worldwide.
One of many key drivers of Philips’ innovation technique is synthetic intelligence (AI), which permits the creation of sensible and personalised services and products that may enhance well being outcomes, improve buyer expertise, and optimize operational effectivity.
Amazon SageMaker supplies purpose-built instruments for machine studying operations (MLOps) to assist automate and standardize processes throughout the ML lifecycle. With SageMaker MLOps instruments, groups can simply practice, take a look at, troubleshoot, deploy, and govern ML fashions at scale to spice up productiveness of knowledge scientists and ML engineers whereas sustaining mannequin efficiency in manufacturing.
On this put up, we describe how Philips partnered with AWS to develop AI ToolSuite—a scalable, safe, and compliant ML platform on SageMaker. This platform supplies capabilities starting from experimentation, knowledge annotation, coaching, mannequin deployments, and reusable templates. All these capabilities are constructed to assist a number of strains of enterprise innovate with pace and agility whereas governing at scale with central controls. We define the important thing use instances that offered necessities for the primary iteration of the platform, the core elements, and the outcomes achieved. We conclude by figuring out the continuing efforts to allow the platform with generative AI workloads and quickly onboard new customers and groups to undertake the platform.
Buyer context
Philips makes use of AI in varied domains, corresponding to imaging, diagnostics, remedy, private well being, and linked care. Some examples of AI-enabled options that Philips has developed over the previous years are:
Philips SmartSpeed – An AI-based imaging expertise for MRI that makes use of a novel Compressed-SENSE based mostly deep studying AI algorithm to take pace and picture high quality to the subsequent stage for a big number of sufferers
Philips eCareManager – A telehealth answer that makes use of AI to assist the distant care and administration of critically ailing sufferers in intensive care models, by utilizing superior analytics and medical algorithms to course of the affected person knowledge from a number of sources, and offering actionable insights, alerts, and proposals for the care staff
Philips Sonicare – A sensible toothbrush that makes use of AI to research the brushing conduct and oral well being of customers, and supply real-time steering and personalised suggestions, corresponding to optimum brushing time, strain, and protection, to enhance their dental hygiene and stop cavities and gum ailments.
For a few years, Philips has been pioneering the event of data-driven algorithms to gas its modern options throughout the healthcare continuum. Within the diagnostic imaging area, Philips developed a large number of ML purposes for medical picture reconstruction and interpretation, workflow administration, and therapy optimization. Additionally in affected person monitoring, picture guided remedy, ultrasound and private well being groups have been creating ML algorithms and purposes. Nonetheless, innovation was hampered resulting from utilizing fragmented AI growth environments throughout groups. These environments ranged from particular person laptops and desktops to various on-premises computational clusters and cloud-based infrastructure. This heterogeneity initially enabled completely different groups to maneuver quick of their early AI growth efforts, however is now holding again alternatives to scale and enhance effectivity of our AI growth processes.
It was evident {that a} basic shift in the direction of a unified and standardized atmosphere was crucial to actually unleash the potential of data-driven endeavors at Philips.
Key AI/ML use instances and platform necessities
AI/ML-enabled propositions can remodel healthcare by automating administrative duties executed by clinicians. For instance:
AI can analyze medical photographs to assist radiologists diagnose ailments quicker and extra precisely
AI can predict future medical occasions by analyzing affected person knowledge and bettering proactive care
AI can suggest personalised therapy tailor-made to sufferers’ wants
AI can extract and construction data from medical notes to make record-taking extra environment friendly
AI interfaces can present affected person assist for queries, reminders, and symptom checkers
General, AI/ML guarantees diminished human error, time and value financial savings, optimized affected person experiences, and well timed, personalised interventions.
One of many key necessities for the ML growth and deployment platform was the power of the platform to assist the continual iterative growth and deployment course of, as proven within the following determine.
The AI asset growth begins in a lab atmosphere, the place the info is collected and curated, after which the fashions are educated and validated. When the mannequin is prepared and authorized to be used, it’s deployed into the real-world manufacturing programs. As soon as deployed, mannequin efficiency is constantly monitored. The true-world efficiency and suggestions are ultimately used for additional mannequin enhancements with full automation of the mannequin coaching and deployment.
The extra detailed AI ToolSuite necessities have been pushed by three instance use instances:
Develop a pc imaginative and prescient utility aimed toward object detection on the edge. The info science staff anticipated an AI-based automated picture annotation workflow to hurry up a time-consuming labeling course of.
Allow an information science staff to handle a household of basic ML fashions for benchmarking statistics throughout a number of medical models. The mission required automation of mannequin deployment, experiment monitoring, mannequin monitoring, and extra management over the complete course of finish to finish each for auditing and retraining sooner or later.
Enhance the standard and time to marketplace for deep studying fashions in diagnostic medical imaging. The present computing infrastructure didn’t permit for operating many experiments in parallel, which delayed mannequin growth. Additionally, for regulatory functions, it’s essential to allow full reproducibility of mannequin coaching for a number of years.
Non-functional necessities
Constructing a scalable and sturdy AI/ML platform requires cautious consideration of non-functional necessities. These necessities transcend the precise functionalities of the platform and concentrate on making certain the next:
Scalability – The AI ToolSuite platform should have the ability to scale Philips’s insights technology infrastructure extra successfully in order that the platform can deal with a rising quantity of knowledge, customers, and AI/ML workloads with out sacrificing efficiency. It must be designed to scale horizontally and vertically to fulfill rising calls for seamlessly whereas offering central useful resource administration.
Efficiency – The platform should ship high-performance computing capabilities to effectively course of advanced AI/ML algorithms. SageMaker gives a variety of occasion sorts, together with situations with highly effective GPUs, which might considerably speed up mannequin coaching and inference duties. It additionally ought to reduce latency and response instances to offer real-time or near-real-time outcomes.
Reliability – The platform should present a extremely dependable and sturdy AI infrastructure that spans throughout a number of Availability Zones. This multi-AZ structure ought to guarantee uninterrupted AI operations by distributing assets and workloads throughout distinct knowledge facilities.
Availability – The platform should be out there 24/7, with minimal downtime for upkeep and upgrades. AI ToolSuite’s excessive availability ought to embody load balancing, fault-tolerant architectures, and proactive monitoring.
Safety and Governance – The platform should make use of sturdy safety measures, encryption, entry controls, devoted roles, and authentication mechanisms with steady monitoring for uncommon actions and conducting safety audits.
Information Administration – Environment friendly knowledge administration is essential for AI/ML platforms. Rules within the healthcare trade name for particularly rigorous knowledge governance. It ought to embody options like knowledge versioning, knowledge lineage, knowledge governance, and knowledge high quality assurance to make sure correct and dependable outcomes.
Interoperability – The platform must be designed to combine simply with Philips’s inside knowledge repositories, permitting seamless knowledge change and collaboration with third-party purposes.
Maintainability – The platform’s structure and code base must be nicely organized, modular, and maintainable. This permits Philips ML engineers and builders to offer updates, bug fixes, and future enhancements with out disrupting the complete system.
Useful resource optimization – The platform ought to monitor utilization studies very intently to verify computing assets are used effectively and allocate assets dynamically based mostly on demand. As well as, Philips ought to use AWS Billing and Price Administration instruments to verify groups obtain notifications when utilization passes the allotted threshold quantity.
Monitoring and logging – The platform ought to use Amazon CloudWatch alerts for complete monitoring and logging capabilities, that are essential to trace system efficiency, establish bottlenecks, and troubleshoot points successfully.
Compliance – The platform may assist enhance regulatory compliance of AI-enabled propositions. Reproducibility and traceability should be enabled mechanically by the end-to-end knowledge processing pipelines, the place many necessary documentation artifacts, corresponding to knowledge lineage studies and mannequin playing cards, might be ready mechanically.
Testing and validation – Rigorous testing and validation procedures should be in place to make sure the accuracy and reliability of AI/ML fashions and stop unintended biases.
Resolution overview
AI ToolSuite is an end-to-end, scalable, fast begin AI growth atmosphere providing native SageMaker and related AI/ML providers with Philips HealthSuite safety and privateness guardrails and Philips ecosystem integrations. There are three personas with devoted units of entry permissions:
Information scientist – Put together knowledge, and develop and practice fashions in a collaborative workspace
ML engineer – Productionize ML purposes with mannequin deployment, monitoring, and upkeep
Information science admin – Create a mission per staff request to offer devoted remoted environments with use case-specific templates
The platform growth spanned a number of launch cycles in an iterative cycle of uncover, design, construct, take a look at, and deploy. As a result of uniqueness of some purposes, the extension of the platform required embedding current customized elements like knowledge shops or proprietary instruments for annotation.The next determine illustrates the three-layer structure of AI ToolSuite, together with the bottom infrastructure as the primary layer, widespread ML elements because the second layer, and project-specific templates because the third layer.
Layer 1 incorporates the bottom infrastructure:
A networking layer with parametrized entry to the web with excessive availability
Self-service provisioning with infrastructure as code (IaC)
An built-in growth atmosphere (IDE) utilizing an Amazon SageMaker Studio area
Platform roles (knowledge science admin, knowledge scientist)
Artifacts storage
Logging and monitoring for observability
Layer 2 incorporates widespread ML elements:
Automated experiment monitoring for each job and pipeline
A mannequin construct pipeline to launch a brand new mannequin construct replace
A mannequin coaching pipeline comprised of mannequin coaching, analysis, registration
A mannequin deploy pipeline to deploy the mannequin for closing testing and approval
A mannequin registry to simply handle mannequin variations
A mission function created particularly for a given use case, to be assigned to SageMaker Studio customers
A picture repository for storing processing, coaching, and inference container photographs constructed for the mission
A code repository to retailer code artifacts
A mission Amazon Easy Storage Service (Amazon S3) bucket to retailer all mission knowledge and artifacts
Layer 3 incorporates project-specific templates that may be created with customized elements as required by new tasks. For instance:
Template 1 – Features a element for knowledge querying and historical past monitoring
Template 2 – Features a element for knowledge annotations with a customized annotation workflow to make use of proprietary annotation tooling
Template 3 – Consists of elements for customized container photographs to customise each their growth atmosphere and coaching routines, devoted HPC file system, and entry from a neighborhood IDE for customers
The next diagram highlights the important thing AWS providers spanning a number of AWS accounts for growth, staging, and manufacturing.
Within the following sections, we focus on the important thing capabilities of the platform enabled by AWS providers, together with SageMaker, AWS Service Catalog, CloudWatch, AWS Lambda, Amazon Elastic Container Registry (Amazon ECR), Amazon S3, AWS Identification and Entry Administration (IAM), and others.
Infrastructure as code
The platform makes use of IaC, which permits Philips to automate the provisioning and administration of infrastructure assets. This method may even assist reproducibility, scalability, model management, consistency, safety, and portability for growth, testing, or manufacturing.
Entry to AWS environments
SageMaker and related AI/ML providers are accessed with safety guardrails for knowledge preparation, mannequin growth, coaching, annotation, and deployment.
Isolation and collaboration
The platform ensures knowledge isolation by storing and processing individually, decreasing the chance of unauthorized entry or knowledge breaches.
The platform facilitates staff collaboration, which is important in AI tasks that sometimes contain cross-functional groups, together with knowledge scientists, knowledge science admins, and MLOps engineers.
Function-based entry management
Function-based entry management (RBAC) is important in managing permissions and simplifying entry administration by defining roles and permissions in a structured method. It makes it simple to handle permissions as groups and tasks develop and entry management for various personas concerned in AWS AI/ML tasks, corresponding to the info science admin, knowledge scientist, annotation admin, annotator, and MLOps engineer.
Entry to knowledge shops
The platform permits SageMaker entry to knowledge shops, which ensures that knowledge might be effectively utilized for mannequin coaching and inference with out the necessity to duplicate or transfer knowledge throughout completely different storage areas, thereby optimizing useful resource utilization and decreasing prices.
Annotation utilizing Philips-specific annotation instruments
AWS gives a set of AI and ML providers, corresponding to SageMaker, Amazon SageMaker Floor Reality, and Amazon Cognito, that are totally built-in with Philips-specific in-house annotation instruments. This integration permits builders to coach and deploy ML fashions utilizing the annotated knowledge throughout the AWS atmosphere.
ML templates
The AI ToolSuite platform gives templates in AWS for varied ML workflows. These templates are preconfigured infrastructure setups tailor-made to particular ML use instances and are accessible by providers like SageMaker mission templates, AWS CloudFormation, and Service Catalog.
Integration with Philips GitHub
Integration with GitHub enhances effectivity by offering a centralized platform for model management, code opinions, and automatic CI/CD (steady integration and steady deployment) pipelines, decreasing handbook duties and boosting productiveness.
Visible Studio Code integration
Integration with Visible Studio Code supplies a unified atmosphere for coding, debugging, and managing ML tasks. This streamlines the complete ML workflow, decreasing context switching and saving time. The mixing additionally enhances collaboration amongst staff members by enabling them to work on SageMaker tasks collectively inside a well-recognized growth atmosphere, using model management programs, and sharing code and notebooks seamlessly.
Mannequin and knowledge lineage and traceability for reproducibility and compliance
The platform supplies versioning, which helps maintain monitor of adjustments to the info scientist’s coaching and inference knowledge over time, making it simpler to breed outcomes and perceive the evolution of the datasets.
The platform additionally permits SageMaker experiment monitoring, which permits end-users to log and monitor all of the metadata related to their ML experiments, together with hyperparameters, enter knowledge, code, and mannequin artifacts. These capabilities are important for demonstrating compliance with regulatory requirements and making certain transparency and accountability in AI/ML workflows.
AI/ML specification report technology for regulatory compliance
AWS maintains compliance certifications for varied trade requirements and rules. AI/ML specification studies function important compliance documentation, showcasing adherence to regulatory necessities. These studies doc the versioning of datasets, fashions, and code. Model management is important for sustaining knowledge lineage, traceability, and reproducibility, all of that are vital for regulatory compliance and auditing.
Mission-level finances administration
Mission-level finances administration permits the group to set limits on spending, serving to to keep away from surprising prices and making certain that the ML tasks keep inside finances. With finances administration, the group can allocate particular budgets to particular person tasks or groups, which helps groups establish useful resource inefficiencies or surprising value spikes early on. Along with finances administration, with the characteristic to mechanically shut down idle notebooks, staff members keep away from paying for unused assets, additionally releasing priceless assets when they don’t seem to be actively in use, making them out there for different duties or customers.
Outcomes
AI ToolSuite was designed and applied as an enterprise-wide platform for ML growth and deployment for knowledge scientists throughout Philips. Various necessities from all enterprise models have been collected and regarded through the design and growth. Early within the mission, Philips recognized champions from the enterprise groups who offered suggestions and helped consider the worth of the platform.
The next outcomes have been achieved:
Person adoption is likely one of the key main indicators for Philips. Customers from a number of enterprise models have been educated and onboarded to the platform, and that quantity is anticipated to develop in 2024.
One other necessary metric is the effectivity for knowledge science customers. With AI ToolSuite, new ML growth environments are deployed in lower than an hour as an alternative of a number of days.
Information science groups can entry a scalable, safe, cost-efficient, cloud-based compute infrastructure.
Groups can run a number of mannequin coaching experiments in parallel, which considerably diminished the common coaching time from weeks to 1–3 days.
As a result of the atmosphere deployment is totally automated, it requires nearly no involvement of the cloud infrastructure engineers, which diminished operational prices.
The usage of AI ToolSuite considerably enhanced the general maturity of knowledge and AI deliverables by selling the usage of good ML practices, standardized workflows, and end-to-end reproducibility, which is vital for regulatory compliance within the healthcare trade.
Trying ahead with generative AI
As organizations race to undertake the subsequent state-of-the-art in AI, it’s crucial to undertake new expertise within the context of the group’s safety and governance coverage. The structure of AI ToolSuite supplies a superb blueprint for enabling entry to generative AI capabilities in AWS for various groups at Philips. Groups can use basis fashions made out there with Amazon SageMaker JumpStart, which supplies an unlimited variety of open supply fashions from Hugging Face and different suppliers. With the mandatory guardrails already in place by way of entry management, mission provisioning, and value controls, will probably be seamless for groups to begin utilizing the generative AI capabilities inside SageMaker.
Moreover, entry to Amazon Bedrock, a totally managed API-driven service for generative AI, might be provisioned for particular person accounts based mostly on mission necessities, and the customers can entry Amazon Bedrock APIs both by way of the SageMaker pocket book interface or by their most well-liked IDE.
There are extra concerns regarding the adoption of generative AI in a regulated setting, corresponding to healthcare. Cautious consideration must be given to the worth created by generative AI purposes towards the related dangers and prices. There’s additionally a must create a danger and authorized framework that governs the group’s use of generative AI applied sciences. Components corresponding to knowledge safety, bias and equity, and regulatory compliance have to be thought-about as a part of such mechanisms.
Conclusion
Philips launched into a journey of harnessing the facility of data-driven algorithms to revolutionize healthcare options. Through the years, innovation in diagnostic imaging has yielded a number of ML purposes, from picture reconstruction to workflow administration and therapy optimization. Nonetheless, the varied vary of setups, from particular person laptops to on-premises clusters and cloud infrastructure, posed formidable challenges. Separate system administration, safety measures, assist mechanisms, and knowledge protocol inhibited a complete view of TCO and complex transitions between groups. The transition from analysis and growth to manufacturing was burdened by the shortage of lineage and reproducibility, making steady mannequin retraining troublesome.
As a part of the strategic collaboration between Philips and AWS, the AI ToolSuite platform was created to develop a scalable, safe, and compliant ML platform with SageMaker. This platform supplies capabilities starting from experimentation, knowledge annotation, coaching, mannequin deployments, and reusable templates. All these capabilities have been constructed iteratively over a number of cycles of uncover, design, construct, take a look at, and deploy. This helped a number of enterprise models innovate with pace and agility whereas governing at scale with central controls.
This journey serves as an inspiration for organizations trying to harness the facility of AI and ML to drive innovation and effectivity in healthcare, finally benefiting sufferers and care suppliers worldwide. As they proceed to construct upon this success, Philips stands poised to make even larger strides in bettering well being outcomes by modern AI-enabled options.
To be taught extra about Philips innovation on AWS, go to Philips on AWS.
In regards to the authors
Frank Wartena is a program supervisor at Philips Innovation & Technique. He coordinates knowledge & AI associated platform belongings in assist of our Philips knowledge & AI enabled propositions. He has broad expertise in synthetic intelligence, knowledge science and interoperability. In his spare time, Frank enjoys operating, studying and rowing, and spending time along with his household.
Irina Fedulova is a Principal Information & AI Lead at Philips Innovation & Technique. She is driving strategic actions centered on the instruments, platforms, and greatest practices that pace up and scale the event and productization of (Generative) AI-enabled options at Philips. Irina has a robust technical background in machine studying, cloud computing, and software program engineering. Outdoors work, she enjoys spending time along with her household, touring and studying.
Selvakumar Palaniyappan is a Product Proprietor at Philips Innovation & Technique, in command of product administration for Philips HealthSuite AI & ML platform. He’s extremely skilled in technical product administration and software program engineering. He’s presently engaged on constructing a scalable and compliant AI and ML growth and deployment platform. Moreover, he’s spearheading its adoption by Philips’ knowledge science groups with the intention to develop AI-driven well being programs and options.
Adnan Elci is a Senior Cloud Infrastructure Architect at AWS Skilled Companies. He operates within the capability of a Tech Lead, overseeing varied operations for purchasers in Healthcare and Life Sciences, Finance, Aviation, and Manufacturing. His enthusiasm for automation is clear in his intensive involvement in designing, constructing and implementing enterprise stage buyer options throughout the AWS atmosphere. Past his skilled commitments, Adnan actively dedicates himself to volunteer work, striving to create a significant and constructive influence throughout the neighborhood.
Hasan Poonawala is a Senior AI/ML Specialist Options Architect at AWS, Hasan helps prospects design and deploy machine studying purposes in manufacturing on AWS. He has over 12 years of labor expertise as an information scientist, machine studying practitioner, and software program developer. In his spare time, Hasan likes to discover nature and spend time with family and friends.
Sreoshi Roy is a Senior International Engagement Supervisor with AWS. Because the enterprise accomplice to the Healthcare & Life Sciences Prospects, she comes with an unparalleled expertise in defining and delivering options for advanced enterprise issues. She helps her prospects make strategic targets, outline and design cloud/ knowledge methods and implement the scaled and sturdy answer to fulfill their technical and enterprise targets. Past her skilled endeavors, her dedication lies in making a significant influence on individuals’s lives by fostering empathy and selling inclusivity.
Wajahat Aziz is a pacesetter for AI/ML & HPC in AWS Healthcare and Life Sciences staff. Having served as a expertise chief in several roles with life science organizations, Wajahat leverages his expertise to assist healthcare and life sciences prospects leverage AWS applied sciences for growing state-of-the-art ML and HPC options. His present areas of focus are early analysis, medical trials and privateness preserving machine studying.
Wioletta Stobieniecka is a Information Scientist at AWS Skilled Companies. All through her skilled profession, she has delivered a number of analytics-driven tasks for various industries corresponding to banking, insurance coverage, telco, and the general public sector. Her information of superior statistical strategies and machine studying is nicely mixed with a enterprise acumen. She brings current AI developments to create worth for purchasers.