How has Ideas2IT leveraged AWS to build a HIPAA-compliant Healthcare Solution?

Ideas2IT serves 100+ clients globally across sectors. What was interesting is that COVID left the company unfazed, while it was a pretty difficult years for most companies across sectors. Since its inception in 2009, it has been crafting and co-creating products with its customers with innovation at its core. The company follows a unique 4I approach - Intensity, Intelligence, Imagination, Integrity. A handpicked team of a 300 people sits out of its India headquarters in Chennai. With their support, the company has been nimble in its approach towards growth. It had a 50%+ growth rate in the last three years. Now, it expects to maintain this growth rate for the next couple of years.

In an interaction with CiOL, Gayathri Vivekanandan, CEO, Ideas2IT, outlines how the company built a personalised platform for their healthcare customer.

The Customer

Cancer is a leading cause of death worldwide, accounting for nearly 10 million deaths in 2020 alone. Historically, cancer was treated as a single disease, but we now know that there are more than 200 different types; each with different behaviours, rates of growth and stages of progression. This has triggered a shift away from a ‘one-size-fits-all’ treatment approach (e.g. chemotherapy) towards tailored therapy and personalised healthcare. The customer is a $60B Global Pharma major with a focused approach to develop innovative diagnostics and medicines to treat cancer patients.

The Challenge

The customer wanted to build data-driven and value-based clinical decision support solutions to provide personalized healthcare to cancer patients. The customer roped us in to develop a HIPAA-compliant data platform to handle diversified patient records (clinical and diagnostic information). We also had to provide a unified view to doctors and clinicians.

This entailed deploying best-in-class technology for data extraction, cleansing, transformation, and building a decision support system. The system would navigate the large volumes of hospital and patient data records and transforms them into actionable insights. Ideas2IT would also deliver a solution that adhered to US’ Regulatory and Compliance Reporting requirements.

The Solution

We built a data platform on AWS S3 to structurally store and manage the large pool of patient data from hospitals. The team created the data pipeline and data lake to handle the vast patient information to store the data in FHIR - HL7 format. We further developed APIs for different applications of the healthcare ecosystem to consume the FHIR resources including patient, hospital, meeting and other data.

We leveraged AWS, Java 11 and Angular 9 to develop a high-performing SaaS healthcare solution and built an Analytics module using D3 to provide customized insights. Lastly, we provisioned Analytics DB to support the integration with various third-party BI tools helping the accreditation needs at organization-level

The Team

We deployed a team of Tech Project Managers, Technical Architects and Business Analysts. We also had top Developers with experience in delivering HIPAA compliant solutions. Our AWS-certified architects and engineers leveraged their vast experience in developing Cloud Native scalable applications. They also built a sophisticated solution to deliver an intuitive dashboard leveraging the extensive healthcare information.

The Implementation

Our implementation included the following components. It all starts with the Data Ingestion Pipeline to receive the heterogeneous patient data bundles, which includes details of medications, diagnostic history, etc. Post which Ideas2IT processed the data into FHIR HL7-R4 format and placed it in the landing zone (AWS S3).

Next we had - Data Ingestion Service. The Data ingestion service helps pick the New data record notification by AWS Lambda function that captures the hospital metadata and passes it through AWS Kinesis stream for real-time processing.  AWS DynamoDB is used to track processed data to avoid duplicate processing.

We then moved the data to a Data Lake where the FHIR message bundle/Documents containing the Patient Data from the landing zone to the Data lake (pool of AWS S3 per hospital). This led us to Router services where Microservices mod listens to AWS SQS for any new data being added to the Data Lake. It also routes it to applications that have subscribed to (AWS SNS).

Finally, we arrive at Operational Analytics. Here, the nightly job refreshes the Analytics DB with patients’ info and their diagnostic info from the FHIR message bundle; and an Analytics dashboard (D3) offers different views - Patient, Cancers, Meetings with Doctors etc. Different views are presented to Management, Clinicians and Patients.

Results

Our team started producing incremental insights right from the fourth month into the engagement and the client could reap ROI in a matter of 8 to 10 months. The data-driven application also helped our clients to onboard new providers faster and enabled them to achieve their vision.