BrainKey.ai is a health tech company that specializes in tracking changes in the brain over time and providing personalized recommendations for nutrition, exercise, and lifestyle changes (focus: brain health monitoring and lifestyle recommendations).
The company uses MRI scans and genetic data to analyze changes, with a particular emphasis on retrieving and processing imaging data (such as x-rays) from hospitals.
Challenges
– Data Management: managing sensitive patient data, including x-rays and medical history, in compliance with HIPAA regulations was crucial.
– Integration: secure integration with hospital networks was required to facilitate data exchange while ensuring privacy and security.
– Scalability: the infrastructure needed to quickly scale to handle peak loads and growing demands, especially when new scans were received.
– Reliable Data Transmission: ensuring reliable and continuous data transmission through a VPN with unified standards across different hospitals was essential.
– Dynamic Scaling: implementing a system that could dynamically scale based on real-time data analysis needs.
Solutions
- HIPAA Compliance
Gart Solutions implemented strict data management practices and utilized HashiCorp Vault to securely store and manage sensitive information, ensuring compliance with HIPAA regulations. (review our other case study of Electronic Medical Records Software for the government-based E-Health platform).
- Secure Integration
A secure network architecture was designed using Kubernetes for container orchestration, enabling safe integration with hospital networks (review another project with Kubernetes solution).
- Scalable Infrastructure
Terraform was leveraged to automate infrastructure provisioning, and Jenkins was used for continuous integration and deployment, ensuring quick scalability to handle peak loads. (review another project where the Terraform cloud tool was used to manage dynamic environments).
- Reliable Data Transmission
A unified VPN standard was established to enable continuous and secure data transmission between BrainKey.ai and connected hospital networks.
- Dynamic Scaling depends on the workload
A monitoring system tracked the number of records in the processing queue. When the queue exceeded a set threshold, the system automatically scaled up resources. Likewise, when volume dropped below a threshold, resources were scaled down—ensuring efficiency without sacrificing performance. RabbitMQ was used for queue management.
