Related Experiment Video
Updated: Oct 15, 2025
09:50
Author Spotlight: Aiding Research in Kidney Biology by Labeling Glomeruli in Cleared Tissues
Published on: February 9, 2024
1.4K
Multi-view features integrated 2D\3D Net for glomerulopathy histologic types classification using ultrasound images
Jinjin Hai1, Kai Qiao1, Jian Chen1
1Henan Key Laboratory of Imaging and Intelligent Processing, PLA Strategy Support Force Information Engineering University, China.
Computer Methods and Programs in Biomedicine
|October 25, 2021
Summary
This study introduces CD-ConcatNet, a novel deep learning model using multi-view renal ultrasound images for accurate glomerulopathy classification. The method shows promise in improving diagnostic capabilities for kidney diseases.
Area of Science:
- Nephrology
- Medical Imaging
- Artificial Intelligence
Background:
- Glomerulopathy diagnosis relies on invasive renal biopsy.
- Early diagnosis and treatment are crucial for managing glomerulopathy progression.
- Non-invasive diagnostic tools are needed to improve patient outcomes.
Purpose of the Study:
- To develop and validate a non-invasive method for histologic classification of glomerulopathy using renal ultrasonography.
- To evaluate the efficacy of a novel deep learning model, CD-ConcatNet, for this classification task.
Main Methods:
- Proposed a multi-view and cross-domain integration strategy (CD-ConcatNet) for renal ultrasound image analysis.
- Utilized 2D group convolution and 3D convolution to extract multi-view features from ultrasound images.
- Employed cross-domain concatenation for enhanced feature learning.
Main Results:
- The study included 76 adult patients (56 training, 20 validation).
- CD-ConcatNet achieved a mean accuracy of 0.83 and an AUC of 0.8667 on the validation dataset.
- Demonstrated superiority in classifying histologic types of glomerulopathy.
Conclusions:
- CD-ConcatNet demonstrates high performance in histologic classification of glomerulopathy.
- Integrating multi-view ultrasound images enhances diagnostic accuracy.
- Renal ultrasonography provides valuable discriminating information for histologic diagnosis.
