Publications

Unified Digital Biomarker Platform for Wearable Devices Assuring Integrity, Heterogeneity, and Scalability

Scientific Reports, 2026.

Yeonho Yoo, Junseok Lee, Seungwoo Jung, Yunkyung Kang, Dosun Lim, Youngpil Kim, Gyeongsik Yang, Chuck Yoo

Download PDF

Abstract

This study presents a cloud-based platform for managing personal digital biomarker data collected from wearable and mobile devices. The platform aims to address four essential challenges for personalized healthcare: ensuring data integrity, supporting various biomarker data types, maintaining measurement reliability, and enabling system scalability. To protect data integrity and privacy, the platform uses private blockchain technology. It also supports heterogeneous data formats, such as time-series ECG signals and dietary images, through standardized interfaces. Measurement reliability is ensured by Gaussian mixture model that successfully detect faulty data with 100% recall. In real-world clinical tests with patients, the platform detects all attempts of data tampering and processes diverse biomarker data with an average latency of less than 1.76 s. The platform's scalability maintains stable performance under high data loads, and practical use is demonstrated through integration with our mobile application and biomarker web server. We validate the platform by measuring actual biomarker data from real-world atrial fibrillation patients and demonstrate its capability to securely and efficiently integrate wearable biomarker data for personalized healthcare.

Paper summary

Keep wearable measurements usable and verifiable

Wearable data needs a consistent path from collection to clinical review, with checks on both stored integrity and measurement quality.

ECG time series and dietary images arrive in different formats, and a correctly stored record may still contain a poor measurement. Xalute separates these concerns in a cloud platform connecting wearable devices, a mobile application and a clinician-facing web service.

Combine a common interface with independent checks

A FHIR-based API accepts the different biomarker formats. The raw data is encrypted in private storage, while a private blockchain records hashes used to detect subsequent changes. A Gaussian-mixture-model component separately screens measurement quality; resource management scales the service as load increases.

Hash verification protects record integrity, while quality screening assesses the measurement.

The platform connects wearable collection, quality screening, protected storage and authorized review.View full-size figure
The platform connects wearable collection, quality screening, protected storage and authorized review.

Evaluate the data pipeline in a clinical setting

The study enrolls 100 patients with atrial fibrillation at Korea University Anam Hospital and deploys the backend on Google Cloud. In the integrity experiment, the platform detects all 500 deliberately modified samples by comparing their hashes with the stored records.

A separate transmission experiment uses 500 ECG signals and 500 dietary images. Average end-to-end latency is 1.76 seconds, measured through storage in both the private database and blockchain ledger.

Data-handling outcomes for the tested ECG and dietary-image workflow.
MetricResult
Deliberately modified records detected500 of 500
Average end-to-end storage latency1.76 seconds

Key takeaways

  1. A common workflow can span unlike biomedical records. Numeric ECG data and dietary images enter the same collection and authorized-review platform.
  2. Record integrity and measurement quality are separate questions. Hash verification checks stored content, while screening addresses the quality of the original measurement.