Machine Learning for Handheld Vascular Studies (DopplerZAM)

Development and Validation of a Novel Machine-learning Algorithm to Assist in Handheld Vascular Diagnostics

The use of handheld arterial 'stethoscopes' (continuous wave Doppler devices) are ubiquitous in clinical practice. However, most users have received no formal training in their use or the interpretation of the returned data. This leads to delays in diagnosis and errors in diagnosis.

The investigators intend to create a novel machine-learning algorithm to assist clinicians in the use of this data. This study will allow the investigators to collect sound files from the use of the devices and compare the algorithms output to established, existing vascular testing. There will be no invasive procedures, and use of these stethoscopes is part of routine clinical care.

If successful, this data and algorithm will be later deployed via smartphone app for point of case testing in a separate study

Study Overview

• Status: Recruiting
• Conditions: Wounds and Injuries; Atherosclerosis
• Intervention / Treatment: Device: Non-invasive vascular testing; Device: machine-learning algorithm

Detailed Description

There are three main research tasks for this project: 1) the identification of discriminant features of Doppler audio for patient classification, 2) the selection and training of classification algorithms, and 3) CWD audio data enrichment using physics-based models. The investigators will determine which discriminant features are optimal for patient classification from ultrasound Doppler audio.

To this end, the investigators will employ signal features in the frequency domain such as bandwidth, peak frequency, mean power, mean frequency, and time harmonic distortion, among others.

Furthermore, the investigators will investigate whether time domain features are necessary for accurate sound classification. Other studies have shown that specific features of audio waveforms can classify the data. The investigators will employ some of the most effective machine-learning algorithms for classification such as SVM, logistic regression, and Naïve Bayes, among others. The investigators will start with a binary classification problem in which individuals will be classified as healthy or unhealthy. Then, the investigators will move in complexity to multi-class classification problems in which individuals will be categorized into different groups according to defined abnormal arterial conditions. Data enrichment using physics-based models employing physiologically accurate finite element models of fluid flow in arteries to generate synthetic sound signals corresponding to various arterial conditions. Physics-based simulations would allow the investigators to produce a wealth of training data that can span many known arterial conditions. This capability can augment the classification accuracy and generalization of our algorithms, as clinical data may not be exhaustive enough to incorporate all the known arterial conditions. The investigators will study the performance of the trained algorithms on patient data. To this end, the investigators will partition the data into training and testing samples. The training samples will be used for training of the algorithms, while the testing set will be used to assess generalization capability. The investigators will compute misclassification rates for each algorithm as a metric for performance.

• Study Type: Observational
• Enrollment (Estimated): 180

Contacts and Locations

• Study Contact: Name: Leila Mureebe, MD; Email: leila.mureebe@duke.edu

Study Locations

• United States
  • North Carolina
    • Durham, North Carolina, United States, 27710
      • Recruiting
      • Duke University Medical Center
      • Contact: Leila Mureebe, MD; Email: leila.mureebe@duke.edu

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Child; Adult; Older Adult
• Accepts Healthy Volunteers: Yes

• Sampling Method: Non-Probability Sample

Study Population

Patients with a clinical indication and order for non-invasive vascular testing

Description

Inclusion Criteria:

• A clinically driven request for non-invasive vascular testing must be present

Exclusion Criteria:

• None (other than patient declines to participate)

Study Plan

How is the study designed?

Design Details

• Observational Models: Cohort
• Time Perspectives: Prospective

Number of groups / cohorts

1

Cohorts and Interventions

Group / Cohort	Intervention / Treatment
Non-invasive vascular testing; All patients undergoing non-invasive vascular testing will be eligible for this study. The official results will be used to develop the algorithm and to evaluate the accuracy of the algorithm	Device: Non-invasive vascular testing; Results of clinically indicated non-invasive vascular testing will be used to develop a machine learning algorithm; Other Names: Continuous wave Doppler; plethysmography; Device: machine-learning algorithm

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Time Frame
Algorithm generated Doppler classification	1 year

Secondary Outcome Measures

Outcome Measure	Time Frame
Presence or absence of pulse	1 year
Quality of pulse	1 year
Presence or absence of Doppler signal	1 year
Quality of Doppler signal	1 year

Collaborators and Investigators

• Sponsor: Duke University

Study record dates

Study Major Dates

• Study Start (Actual): September 7, 2016
• Primary Completion (Estimated): December 31, 2026
• Study Completion (Estimated): December 31, 2026

Study Registration Dates

• First Submitted: September 19, 2016
• First Submitted That Met QC Criteria: October 12, 2016
• First Posted (Estimated): October 13, 2016

Study Record Updates

• Last Update Posted (Actual): March 5, 2026
• Last Update Submitted That Met QC Criteria: March 4, 2026
• Last Verified: March 1, 2026

More Information

Terms related to this study

• Keywords: Arteries
• Additional Relevant MeSH Terms: Vascular Diseases; Cardiovascular Diseases; Arteriosclerosis; Arterial Occlusive Diseases; Wounds and Injuries; Atherosclerosis; Diagnostic Techniques and Procedures; Diagnosis; Diagnostic Techniques, Cardiovascular; Plethysmography
• Other Study ID Numbers: Pro00070090

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: NO

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No