Related Experiment Video
Updated: Nov 1, 2025
Supervised Machine Learning for Semi-Quantification of Extracellular DNA in Glomerulonephritis
Published on: June 18, 2020
Deep learning-based framework for the distinction of membranous nephropathy: a new approach through hyperspectral
Tianqi Tu1,2, Xueling Wei3, Yue Yang2
1Peking University China-Japan Friendship School of Clinical Medicine, Beijing, China.
Background:
Common subtypes seen in Chinese patients with membranous nephropathy (MN) include idiopathic membranous nephropathy (IMN) and hepatitis B virus-related membranous nephropathy (HBV-MN). However, the morphologic differences are not visible under the light microscope in certain renal biopsy tissues.
Methods:
We propose here a deep learning-based framework for processing hyperspectral images of renal biopsy tissue to define the difference between IMN and HBV-MN based on the component of their immune complex deposition.
Results:
The proposed framework can achieve an overall accuracy of 95.04% in classification, which also leads to better performance than support vector machine (SVM)-based algorithms.
Conclusion:
IMN and HBV-MN can be correctly separated via the deep learning framework using hyperspectral imagery. Our results suggest the potential of the deep learning algorithm as a new method to aid in the diagnosis of MN.
Insights
A novel deep learning framework accurately distinguishes between idiopathic membranous nephropathy (IMN) and hepatitis B virus-related membranous nephropathy (HBV-MN) using hyperspectral imaging, aiding in diagnosing membranous nephropathy (MN).
Area of Science:
- Nephrology
- Digital Pathology
- Artificial Intelligence
Background:
- Membranous nephropathy (MN) subtypes, including idiopathic (IMN) and hepatitis B virus-related (HBV-MN), present diagnostic challenges.
- Morphologic distinctions between IMN and HBV-MN are often indiscernible via light microscopy in renal biopsy tissues.
Purpose of the Study:
- To develop a deep learning framework for differentiating IMN and HBV-MN.
- To analyze immune complex deposition components in renal biopsy hyperspectral images.
Main Methods:
- A deep learning framework was designed to process hyperspectral images of renal biopsy tissue.
- The framework was trained to classify samples based on immune complex deposition characteristics.
Main Results:
- The deep learning framework achieved a 95.04% classification accuracy.
- Performance surpassed traditional support vector machine (SVM)-based algorithms.
Conclusions:
- Deep learning with hyperspectral imagery effectively separates IMN and HBV-MN.
- This approach shows promise as an adjunctive diagnostic tool for membranous nephropathy.
