digital biomarker discovery pipeline

What digital biomarker discovery pipeline does

Digital Biomarker Discovery Pipeline (DBDP) is an open-source software platform for building digital biomarkers from wearable and mHealth data. Founded in 2020 and affiliated with the BIG IDEAs Lab at Duke University, the platform is maintained by a coalition of academic researchers at Duke, UC San Diego, UCSF, and Arizona State University, led by founder and PI Jessilyn Dunn, PhD, alongside co-founders Brinnae Bent, Ryan Runge, and Ida Sim. DBDP serves academic researchers, data scientists, clinicians, and students who need transparent, reproducible tooling to extract validated biomarkers from time-series sensor data produced by consumer wearables and clinical devices.

The platform is organized as a modular ecosystem of over 29 active repositories on GitHub covering the complete biomarker discovery workflow: data (Digital Health Data Repository, Awesome-CGM), preprocessing (Pre-process, devicely), feature generation (Heart-Rate-Variability, Sleep, Human-Activity-Recognition, wearablecompute with 50+ domain-driven features, Resting-Heart-Rate, wearablevar, nutritionSearchTool), specialized analysis (cgmquantify/iglu for CGM glycemic variability, CovIdentify for infectious disease prediction, DBDP-Autonomic for multimodal autonomic nervous system analysis), deep learning (DeepPostures for posture recognition, WatchSleepNet for sleep staging), and supporting tools (Data-Compression-Toolbox, Exploratory-Data-Analysis, SGCTools, latentcor, flirt). Implementations span both Python and R environments, with many modules delivered as Jupyter Notebooks for reproducibility. Algorithms in core modules are validated against clinical gold standards such as Kubios for HRV. The platform's scientific credibility is anchored by 16 peer-reviewed publications in outlets including Nature Medicine, npj Digital Medicine, Journal of Clinical and Translational Science, and Frontiers in Digital Health, plus presentations at IEEE-EMBS BHI, BIRS, and JupyterCon.

DBDP operates without a commercial revenue model. All code is released under the MIT License and distributed freely via GitHub. The project is sustained through institutional affiliation with Duke University and collaborator relationships with the Chan Zuckerberg Initiative, Open mHealth, and other academic partners. Distribution relies entirely on community-led, organic adoption: researchers and developers discover the tools through GitHub, peer-reviewed publications, workshops, and the dedicated DBDP Learn tutorial platform. There are no pricing tiers, no paying customers, no funding rounds, and no parent company or external institutional investment; the entity functions as a non-commercial, community-driven research infrastructure project.

Name

digital biomarker discovery pipeline

Legal name

Digital Biomarker Discovery Pipeline

Website

https://dbdp.org

Company type

Private

Founded year

2020

Operating status

Operating

Headcount range

1–10 employees

Short description

Digital Biomarker Discovery Pipeline (DBDP) is an open-source platform, affiliated with Duke University's BIG IDEAs Lab, that provides modular Python and R tools for building digital biomarkers from wearable and mHealth sensor data, serving academic researchers, data scientists, clinicians, and students worldwide under the MIT License.

Ownership category

akta.pro rank

Product category

Digital Health Research Software

NAICS

Research and Development in the Physical, Engineering, and Life Sciences (54171)

SIC

Services-Computer Programming, Data Processing, Etc. (7370), Services-Commercial Physical & Biological Research (8731)

akta.pro primary industry

SaMD – Digital Biomarkers & Algorithmic Endpoints (HLACANAF)

akta.pro secondary industry

Real-World Data (RWD) Platforms & Clinical Data Curation (HLACAIAJ)

Keywords

• Digital biomarker discovery
• Wearable data analytics
• Mhealth signal processing
• Open-source health tools
• Biomedical informatics software

GTM type

B2B

Offering type

Software

Cost components

Personnel, Technology or R&D, Operations, Marketing or Sales, Infrastructure

Revenue model

1. Open-source/free access: The DBDP is an open-source platform released under the MIT License with no commercial revenue model. The project is developed and maintained by academic researchers at Duke University's BIG IDEAs Lab with support from collaborators and the open-source community.

Go-to-market motion1 record

Distribution channels1 record

Marketing channels5 records

Core offering

DBDP is an open-source software platform for developing digital biomarkers from wearable and mHealth sensor data. It provides modular Python and R packages covering the full biomarker discovery pipeline: data ingestion, preprocessing, feature generation, domain-specific analysis (glucose, heart rate variability, sleep, activity, autonomic function), and deep learning, supported by a curated Digital Health Data Repository of sample datasets. All tools are released under the MIT License and made available to the global research community via GitHub.

Product overview

DBDP is an open-source software platform for building digital biomarkers from wearable data, maintained by the BIG IDEAs Lab at Duke University. The platform operates as a modular ecosystem of over 29 active repositories covering the complete biomarker discovery pipeline: data repositories (Digital Health Data Repository), preprocessing (Pre-process), feature generation (Heart-Rate-Variability, Sleep, Human-Activity-Recognition, wearablecompute, wearablevar), specialized analysis (cgmquantify/iglu for glucose, CovIdentify for infectious disease, DBDP-Autonomic for autonomic nervous system), deep learning (DeepPostures), and supporting tools (Data-Compression-Toolbox, Exploratory-Data-Analysis, devicely). All tools are open-source and designed for transparent, reproducible research with wearable and mHealth sensor data.

Differentiator

Problem solved

Functional benefit

Products and services

• Digital Biomarker Discovery Pipeline (DBDP) Platform An open-source, MIT-licensed software platform providing modular Python and R packages for the complete digital biomarker discovery pipeline: preprocessing, feature generation, domain-specific analysis (glucose, heart rate variability, sleep, activity, autonomic function), deep learning, and supporting tooling. Built on Jupyter Notebook environments and released through 29 active GitHub repositories.
• Digital Health Data Repository (DHDR) A repository containing curated sample datasets for use with DBDP tools, enabling reproducible research workflows and method validation against shared benchmark data.
• DBDP Learn Platform An educational platform offering tutorials, workshops, and learning materials for digital biomarker methods using DBDP tools, including sessions delivered at IEEE-EMBS BHI (2023, 2025), BIRS (2020, 2025), NCSSM High School Workshop, and JupyterCon 2025.

Quantifiable outcome

• DBDP-Autonomic platform integrates multimodal data for autonomic nervous system analysis with published methodology in Frontiers in Digital Health
• +1 more outcomes

Named customers1 record

Segments3 records

Ideal customer profiles3 records

Technology focussed Yes

API detail

Has API

No

API docs

API detail

Core technology

AI maturity

App detail

Integration2 records

AI capability6 records

Feature9 records

Partnerships

Four partnerships are on record, tiered core.

• Chan Zuckerberg InitiativecoreStrategic or Co-development PartnerFeatured collaborator shown in the DBDP partners/collaborators section on the website, indicating active collaboration or support for the open-source digital biomarker platform.
• Open mHealthcoreStrategic or Co-development PartnerFeatured collaborator in the DBDP partners/collaborators section, representing alignment with open mHealth standards and interoperability for digital health data.
• BIG IDEAs Lab at Duke UniversitycoreStrategic or Co-development PartnerThe core academic home of DBDP, led by Jessilyn Dunn, PhD. Hosts the primary research team and develops/maintains all platform tools and educational resources.
• Duke UniversitycoreStrategic or Co-development PartnerAcademic institution affiliated with BIG IDEAs Lab, providing institutional support and research infrastructure for the DBDP platform.

Scale indicators4 records

Recent moves6 records

Expansion highlights6 records

Direct peers

• VivoSense: VivoSense is a commercial digital biomarker analytics platform that processes wearable and sensor data for clinical research and regulated trials. It is directly comparable as a peer DBDP contributor (Peter Cho) also works there, highlighting overlap in wearable data analytics for clinical-grade biomarker endpoints.
• ActiGraph: ActiGraph provides wearable actigraphy hardware and analytic software widely used in academic and clinical research. It is comparable as a leading provider of research-grade wearable data capture and analysis — overlapping DBDP's activity recognition and HRV/sleep feature generation.
• Biofourmis: Biofourmis is a digital biomarker and remote patient monitoring platform that analyzes continuous biosensor data to support clinical decision-making. It is comparable as a commercial digital biomarker platform overlapping DBDP's work in autonomic, cardiac, and activity-based digital endpoints.
• Evidation: Evidation operates a digital health measurement platform that aggregates wearable and survey data from large populations to derive health insights and digital measures. It is comparable in using consumer wearable data to develop health-related digital measures at population scale.
• Empatica: Empatica develops medical-grade wearable sensors and digital biomarker algorithms cleared for clinical use, including seizure detection. It is comparable as a regulated wearable platform producing validated digital biomarkers from physiological signals — directly overlapping DBDP's HRV and autonomic work.
• PhysIQ: PhysIQ applies AI to continuous biosensor data to compute personalized physiology-based digital biomarkers for clinical and research applications. It is comparable in deriving clinically validated biomarkers from wearable streams — overlapping DBDP's autonomic and HRV analyses.

Broad incumbents

• Medidata (Dassault Systèmes): Medidata provides a broad clinical trial technology platform that increasingly incorporates wearable-derived digital endpoints. It is comparable as an incumbent platform incorporating digital biomarkers into regulated clinical research workflows.
• Fitbit (Google Health): Fitbit (now part of Google) operates consumer wearables with a research API and Fitbit Health Solutions for population health. It is comparable as a major wearable data source whose API DBDP integrates with for tutorials and biomarker workflows.
• Verily (Alphabet): Verily develops health-focused wearable sensors and digital health platforms including the Verily Study Watch used in clinical research. It is comparable as a broad health-tech incumbent producing wearable hardware and digital biomarker analyses for research and clinical programs.

Emerging players

• Open mHealth: Open mHealth is a non-profit that develops open standards and schemas for mobile health data. It is comparable as a listed DBDP collaborator focused on interoperability standards for mHealth sensor data — adjacent to DBDP's wearable data pipelines.

Market position

Strengths4 records

Weaknesses4 records

Competitive moat5 records

Key risks5 records

Key highlights6 records

Customer concentration

• GitHubgithub.com/DigitalBiomarkerDiscoveryPipeline
• Xx.com/TheDBDP
• LinkedInlinkedin.com/company/digital-biomarker-discovery-pipeline

Revenue estimate

Valuation estimate

Number of profiles

Profiles15 records

Funding overview

Funding rounds

Investors

Funding detail is available on the Subscription and Enterprise plan.Contact sales →

M&A

Investments

M&A and investment is available on the Subscription and Enterprise plan.Contact sales →

Frequently asked questions about digital biomarker discovery pipeline