Building a Scalable Health Data Platform

“We implemented a robust data pipeline on AWS that ensures reliable data flow from devices through the core platform to customers.”

Gilles Lunzenfichter, CEO

Customer

Medisanté Group provides innovative digital health solutions that connect medical devices with healthcare providers, insurance companies, and medical practices. Their platform enables seamless collection, processing, and distribution of vital health data from various medical devices, supporting better patient care and health monitoring.

Background

Medisanté operates in a complex healthcare ecosystem where medical device data needs to flow efficiently between patients, healthcare providers, and insurance companies. Their platform processes data from diverse medical devices including body scales, blood glucose monitors, and blood pressure measurement devices. Each device type has its own data format and communication protocol, while each customer segment (hospitals, insurance companies, doctor’s offices) has unique integration requirements and data consumption patterns.

The challenge was to build a flexible, scalable architecture that could handle this heterogeneous landscape while maintaining high availability, security, and compliance with healthcare data regulations.

Project Goal

The primary objective was to build a modern, cloud-native platform on AWS that could:

• Support a core application architecture with pluggable device adapters
• Handle data ingestion from multiple medical device types with different protocols
• Provide flexible integration patterns for diverse customer types
• Scale automatically to handle varying data volumes
• Ensure secure handling of sensitive health data
• Enable efficient management and monitoring of the entire platform
• Provide a user-friendly interface for platform administration

bespinian’s Role

bespinian played a central role in designing and implementing the complete platform infrastructure and application stack:

Cloud-Native Architecture Design

We designed a scalable, event-driven architecture on AWS that separates core functionality from device-specific and customer-specific integrations. This modular approach allows Medisanté to add new device types and customer integrations without affecting the core platform. The architecture leverages AWS serverless services to ensure automatic scaling and cost efficiency.

Device Adapter Implementation

We implemented specialized adapters for different medical device types as AWS Lambda functions written in Go. Each adapter handles the specific communication protocols and data formats of its device category (body scales, blood glucose monitors, blood pressure devices, etc.). These Lambda functions process incoming device data, normalize it to a common format, and feed it into the core data pipeline. The serverless approach ensures that each adapter scales independently based on data volume.

Customer Integration Layer

Similar to the device adapters, we built customer-specific integration adapters as Lambda functions in Go. These adapters transform the normalized health data into the formats and protocols required by different customer types, whether hospitals, insurance companies, or doctor’s offices. This flexibility enables Medisanté to onboard new customers quickly without modifying the core platform.

Infrastructure as Code

We automated the entire infrastructure provisioning using Terraform. This includes the setup of Lambda functions, API Gateway endpoints, data storage solutions, networking components, and monitoring infrastructure. The infrastructure-as-code approach ensures reproducibility, version control, and enables rapid deployment across multiple environments (development, staging, production).

Management Application

We developed a comprehensive IoT management application using React that provides Medisanté’s team with full visibility and control over the platform. The frontend communicates with a custom Go API backend that manages platform configuration, monitors data flows, and provides operational insights. The application enables efficient management of device adapters, customer integrations, and overall platform health.

Data Pipeline Architecture

We implemented a robust data pipeline on AWS that ensures reliable data flow from devices through the core platform to customers. The pipeline handles data validation, transformation, storage, and routing while maintaining data integrity and audit trails for compliance purposes.

Technologies Used

• Infrastructure: Amazon Web Services (AWS)
• Infrastructure as Code: Terraform
• Serverless Computing: AWS Lambda
• Backend Development: Go (Golang)
• Frontend Application: React
• API Management: AWS API Gateway
• Data Storage: AWS S3, DynamoDB
• Monitoring: Amazon CloudWatch