
#23 Google's Update Speed Is INSANE! What's announced and what’s next! (Fitbit Air & Pixel Watch)
Shownotes
Rob and Stephan break down Google's latest health-tech announcements, analyzing the Pixel Watch 5's 3D-modeled GPS tracking, cutting-edge foundation models for predicting insulin resistance and hypertension, and the future of smartphone-based body composition analysis.
📝Summary
In this episode, we examine the massive wave of health-tech research and firmware updates unveiled around the Google Pixel Watch 5 and Fitbit Air. Rob recounts his hands-on testing trip to Berlin with Google engineers to benchmark the Pixel Watch 5's new server-side GPS algorithms, which leverage 3D urban mapping and physics-based multipath correction against surveying-grade backpacks. The hosts then dive into Google's latest biomedical AI models, beginning with a Nature study using wearable sensor embeddings to predict insulin resistance, effectively utilizing passive biometric signals to replace missing fasting insulin labs. They dissect the WavesFM foundation model for continuous hypertension risk assessment against 24-hour ambulatory monitoring, discuss sleep apnea and emergency breathing alerts, and evaluate PhotoScan, an MRI-trained computer vision model estimating DEXA-level body fat percentage from smartphone photos. Finally, they review GlucoseFM for continuous glucose monitoring imputation and reflect on Google's cohesive strategy of bridging consumer hardware, foundation models, and clinician-evaluated personal health agents for preventative medicine.
⏳Chapters
00:00:00 Testing in Berlin: Rob's behind-the-scenes look at Google's GPS event
00:04:01 Smart Algorithms: Server-side 3D mapping and urban multipath correction
00:06:48 Benchmark Methodology: Survey backpacks, ground stations, and run metrics
00:13:38 Tech Review Dynamics: Feedback loops, NDAs, and golden sample testing
00:18:38 Made by Google: Introducing the Health Guardian suite
00:20:31 Insulin Resistance Trends: Understanding HOMA-IR and metabolic dysfunction
00:22:39 Sensor Embeddings: Replacing fasting insulin labs with wearable data
00:29:50 Hypertension Detection: Wrist-based classification vs. 24-hour cuffs
00:37:22 WavesFM: Multi-scale waveform learning and understanding preprints
00:42:13 Critical Alerts: Sleep apnea trends and breathing emergency detection
00:45:56 Personal Health Agents: Clinician-evaluated AI in Google Health
00:48:19 PhotoScan: Estimating DEXA body fat percentage from smartphone photos
00:53:40 GlucoseFM: Imputation and multi-scale dynamics for continuous glucose monitoring
00:58:00 Preventative Healthcare: Tech giants, foundation models, and systemic lifestyle nudges
📚Resources
How we built the new GPS on Pixel Watch 5
Pixel Watch 5: Proactive assistance and advanced health tracking on your wrist
Track subtle changes in your body with Health Guardian features on Pixel and Fitbit
Multipath propagation in GPS signals
Insulin signal transduction pathway
…There is more: complete show notes here
🎙️About
Fit For Science is a deep-dive podcast hosted by two biological data scientists, Rob and Stephan, exploring the intersection of research, health tech, and data-driven lifestyle design. The hosts provide evidence-based systems, layered with practical "N=2" personal experimentation, to cut through the noise and enable everyone to become their best N-of-1.
Please leave us a review if you enjoyed this episode. We always want to hear from our audience whether that's in the form of feedback, new ideas or a topic recommendation!
Learn more and subscribe on your favorite platforms:
⚠️Disclaimer: This podcast represents our own opinions and is for informational purposes only. It does not constitute medical or financial advice or a professional relationship.