Search research articles

Search research articles

Related Experiment Videos

Deep multi-modal features based spatio-temporal video regression for non-invasive hemoglobin estimation.

Ata Güneş¹, Ufuk Bal², Selami Beyhan³

¹Department of Electrical and Electronics Engineering, Faculty of Engineering, Izmir Democracy University, İzmir, Turkey.

Medical & Biological Engineering & Computing

|July 2, 2026

Summary

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Refugees, Trauma, and Positive Psychological Change: Mindfulness as a Moderator for Posttraumatic Growth.

Healthcare (Basel, Switzerland)·2026

Same author

Estimation of Total Hemoglobin (SpHb) from Facial Videos Using 3D Convolutional Neural Network-Based Regression.

Biosensors·2025

Same author

A deep learning feature extraction-based hybrid approach for detecting pediatric pneumonia in chest X-ray images.

Physical and engineering sciences in medicine·2023

Same author

Noninvasive Hemoglobin Measurement Reduce Invasive Procedures in Thalassemia Patients.

Hemoglobin·2022

Same author

The Effects of Occupational Therapy and Psychosocial Interventions on Interpersonal Functioning and Personal and Social Performance Levels of Corresponding Patients.

Noro psikiyatri arsivi·2017

Same author

Non-contact estimation of heart rate and oxygen saturation using ambient light.

Biomedical optics express·2015

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

This summary is machine-generated.

Non-invasive hemoglobin estimation from facial videos offers a promising alternative to blood tests. This machine learning approach uses facial video analysis for accurate anemia screening and remote health monitoring.

Area of Science:

Biomedical Engineering
Artificial Intelligence in Healthcare
Medical Diagnostics

Background:

Hemoglobin concentration is crucial for diagnosing anemia and monitoring health.
Conventional blood sampling is invasive, limiting accessibility and patient comfort.
Non-invasive methods using machine learning are emerging for point-of-care and remote healthcare.

Purpose of the Study:

To develop and validate a non-invasive method for estimating hemoglobin levels from facial videos.
To explore multi-modal feature extraction and ensemble learning for improved accuracy.
To assess the system's potential for anemia screening and remote patient monitoring.

Main Methods:

A dataset of 260 participants was used.
Features were extracted using pre-trained convolutional neural networks (MobileNetV2, ResNet152), remote photoplethysmography (rPPG) signals, and color statistics.

Keywords:

Anemia diagnosis Computer vision Deep learning Hemoglobin concentration estimation Machine learning Non-invasive measurement Remote photoplethysmography Video regression

Related Experiment Videos

Machine learning models including XGBoost, Random Forest, and Stacking Regressor were employed for hemoglobin estimation.

Main Results:

The Stacking Regressor achieved the best performance (MAE: 0.7754 g/dL, R²: 0.5852).
Combining ResNet152 features with XGBoost yielded comparable results (MAE: 0.6635 g/dL, R²: 0.4977).
Multi-modal feature strategies significantly outperformed single-modality approaches.

Conclusions:

The proposed video-based hemoglobin estimation system achieves clinically relevant accuracy.
The system demonstrates potential for anemia screening and remote patient monitoring.
This non-invasive approach shows promise compared to existing literature methods and point-of-care instruments.