Vital Signs: OpenSearch and Bioobservability
summary
A long‑stay EEG records a tonic‑clonic seizure and pins the misfiring to the left temporal lobe; seven out of ten epilepsy cases show no root cause. OpenSearch ingests high‑volume vitals, time series and sample audio, supports real‑time search, dashboards and anomaly detection that flagged the seizure signal, and stores logs, metrics, traces for longitudinal research. Gaps remain: no production demo or sample code was shown, EEG and other biosamples lack ready vector embeddings and pretrained models for semantic search, so a working pipeline is still unbuilt.
AI-written summary · about
chapters (11)
- 0:00Introduction
- 3:39What is observability
- 5:27Bioobservability on smartwatch
- 8:40Body as distributed system
- 12:20OpenSearch for bioobservability
- 17:28AI in bioobservability
- 18:29Mental health limits
- 20:15Medical samples in OpenSearch
- 22:30Beyond humans applications
- 27:27Call to action
- 29:58Q&A
related talks
20mSeeing Is Believing: OpenTelemetry and OpenSearch in Next-Gen Observability
06mSponsored Keynote: OpenSearch - See Everything: Open Observability for Agentic AI
33mPatients as Time-Series: High-Stakes Clinical Alerting With Ope...
42mObserve Smarter, Not Harder: Scaling AI-Powered Observability With O...
06mWhy OpenSearch?
22mFrom Telemetry to Insight: Building AI-Powered Observability Pipelines with OpenSearch
42mBuilding Self-Healing Infrastructure: Agentic AI & Vector Search in OpenSearch
34mHow to Do Microservice Observability with OpenSearch