Projects

Selected research and development projects spanning medical imaging, biomedical signal processing, applied artificial intelligence, real-time deployment, and remote sensing.

Overview

My projects focus on translating AI and signal processing methods into practical tools for health and biomedical applications. This includes ultrasound guidance, Doppler and echocardiography analysis, voice and breathing-based monitoring, smartphone and real-time AI deployment, and earlier work in multispectral image fusion and remote sensing.

Featured Research Projects

AI-Guided Ultrasound and Operator Support

2025 – Present
AI-Guided Ultrasound project figure
Example graphical abstract or representative ultrasound workflow.

Developing machine learning tools that help guide ultrasound acquisition, support minimally trained users, and improve image standardization in real-world clinical environments.

  • Under development section
  • Under development section
Paper Code Dataset

Automated Venous Gas Emboli Detection

2021 – Present
Venous Gas Emboli project figure
VGE detection example in real-time on echocardiogram data.

Building deep learning systems for automatic grading and detection of venous gas emboli from Doppler ultrasound audio and echocardiographic data, including real-time deployment.

  • Deep Learning-Based Venous Gas Emboli Grade Classification in Doppler Ultrasound Audio Recordings [Article] [Code]
  • Free Software for Automated Grading of Venous Gas Emboli in Doppler Ultrasound [Article] [Code]
  • Development of a graphical user interface for automatic separation of human voice from Doppler ultrasound audio in diving experiments [Article] [Code]

Speech and Audio Biomarkers for Health

2023 – Present
Speech and audio biomarker project figure
Voice, breathing, and signal biomarkers for health monitoring.

Investigating speech, breathing, and acoustic biomarkers for digital health, including respiratory stress, edema-related signatures, and enhancement methods for noisy medical audio.

  • Improving deep speech denoising by noisy2noisy signal mapping [Article]
  • Consensus-Based Definitions for Vocal Biomarkers: The International VOCAL Initiative [Article]
  • Acoustic Signatures of Edema: A Voice-Based Approach to Respiratory Stress in Extreme Physiology, In Preparation.

Multispectral Image Fusion and Remote Sensing

2017 – 2021
Multispectral image fusion project figure
Deep learning and optimization for multispectral image fusion.

Earlier work focused on deep learning and optimization-based methods for multispectral image fusion, pansharpening, and remote sensing image analysis.

  • Convolutional Autoencoder-Based Multispectral Image Fusion [Article] [Code]
  • Image Fusion in Remote Sensing: Conventional and Deep Learning Approaches [Book]
  • Application of fractional-order differentiation in multispectral image fusion [Article] [Code]

Selected Student and Mentored Projects

Deep Learning for Diabetic Retinopathy Detection

Mentored student work on detection of five stages of diabetic retinopathy with an emphasis on real-time deployment on smartphones.

Abrams Scholars project

Speech Enhancement for Hearing-Impaired Patients

Mentored a project focused on deep learning-based speech enhancement and real-time smartphone deployment for hearing-related applications.

Abrams Scholars project

Real-Time Automated Bubble Detection in Echocardiography

Undergraduate research project evaluating real-time automated bubble detection workflows for echocardiographic imaging.

Turtle Ultrasound GUI and Decompression Analysis

Developed and mentored work on a graphical user interface for analysis of turtle ultrasound data and decompression sickness-related imaging studies.

Instantaneous Heart Rate from Doppler Ultrasound Audio

Research project focused on extracting heart rate information from Doppler ultrasound audio recordings using signal processing and machine learning.

Research Sponsors