Related Experiment Video
Updated: Jan 3, 2026
Cerenkov Luminescence Imaging of Interscapular Brown Adipose Tissue
Published on: October 7, 2014
A novel Cerenkov luminescence tomography approach using multilayer fully connected neural network
Zeyu Zhang1, Meishan Cai, Yuan Gao
1Engineering Research Center of Molecular and Neuro Imaging of Ministry of Education, School of Life Science and Technology, Xidian University, Xi'an 710126, People's Republic of China. CAS Key Laboratory of Molecular Imaging, Beijing Key Laboratory of Molecular Imaging, The State Key Laboratory of Management and Control for Complex Systems, Institute of Automation, Chinese Academy of Sciences, Beijing 100190, People's Republic of China. These authors contributed equally to this study.
This study introduces a new multilayer fully connected neural network (MFCNN) for Cerenkov luminescence tomography (CLT). MFCNN CLT significantly improves accuracy and stability in biomedical imaging compared to traditional methods.
Area of Science:
- Biomedical Optics
- Medical Imaging
- Machine Learning in Medicine
Background:
- Cerenkov luminescence tomography (CLT) is valuable for biomedical applications.
- Severe light scattering in CLT limits its imaging performance.
- Existing CLT methods struggle with accuracy and stability.
Purpose of the Study:
- To develop a novel CLT reconstruction approach using a multilayer fully connected neural network (MFCNN).
- To train the MFCNN using Monte Carlo simulation data to learn complex source-signal relationships.
- To evaluate the performance of MFCNN-based CLT against traditional methods.
Main Methods:
- Development of a multilayer fully connected neural network (MFCNN) for CLT reconstruction.
- Training the MFCNN with Monte Carlo simulated data to map surface signals to internal light sources.
- Validation through both simulated and in vivo experiments, comparing MFCNN CLT with radiative transfer equation (RTE) based methods.
Main Results:
- The MFCNN effectively learned the complex relationship between surface luminescence signals and true source distributions.
- MFCNN CLT demonstrated superior accuracy and stability in both simulations and in vivo experiments compared to the RTE-based method.
- The proposed method significantly enhances the performance of Cerenkov luminescence tomography.
Conclusions:
- The multilayer fully connected neural network (MFCNN) offers a promising advancement for Cerenkov luminescence tomography (CLT).
- This approach addresses the limitations of severe light scattering, improving imaging accuracy and stability.
- The study highlights the potential of machine learning in advancing biomedical optical imaging techniques.
