Architecting with Google Cloud — selected guide
How to architect an AI/ML powered Healthcare platform on Google Cloud
Ten case studies on scale, service boundaries, data platforms, machine learning, and security
Ten 2021–2023 case studies explain how teams handle global storage, concurrent requests, service boundaries, media ingestion, analytics migration, machine learning, healthcare data, and security response. Results describe the participating organizations' systems, not universal performance guarantees.
In this collection Browse 10 summaries 9 of 10
Vida Health brings virtual care for multiple conditions into one platform. Vice president of engineering Amol describes how its 2021 cloud consolidation connected patient data and clinical documents to tools that help providers find information and prioritize care.
Key Points Covered
- Consolidation reduced the number of environments engineers managed. Amol attributes an approximately 50% cost reduction and lower deployment and monitoring overhead to bringing the systems together; these are his reported outcomes, not an isolated comparison of cloud providers. 00:00–03:10(opens in a new tab)
- The recommendation system retrieves clinical protocols for providers. Vidapedia indexes protocol documents written in Google Workspace and matches them to patient–provider interactions, surfacing relevant guidance during or after a session. Amol also describes automatic note-taking and prioritizing patients by severity as assistance for clinicians, not autonomous treatment. 03:10–07:22(opens in a new tab)
- BigQuery brings different patient records together for analysis. Claims and eligibility files, wearable and glucose readings, consultation notes, and relational application data arrive through batch and real-time ingestion. Analysts query that data and data scientists build models over it; Vertex AI was still being explored for experiments. 07:22–10:31(opens in a new tab)
- Migration and interoperability were at different stages. Vida had completed the migration with SADA after planning data transfer, downtime, and infrastructure ownership. Using Cloud Healthcare API and FHIR, a healthcare-data exchange standard, remained a plan to make transfers to other providers easier instead of relying on Vida's proprietary formats. 16:54–21:07(opens in a new tab)
Full video: https://www.youtube.com/watch?v=QjMl9lgcOU0(opens in a new tab)