In the ever-perilous autoimmune disease world of systemic lupus erythematosus (SLE or lupus), up to 60% of adult patients and 80% of children will develop lupus nephritis (LN), and up to half of those will move on to end-stage renal disease. LN occurs when the immune system wrongly attacks the kidneys, preventing them from doing their job, i.e., cleaning blood, balancing body fluids and controlling hormones that impact blood pressure.
Credit: University of Houston
In the ever-perilous autoimmune disease world of systemic lupus erythematosus (SLE or lupus), up to 60% of adult patients and 80% of children will develop lupus nephritis (LN), and up to half of those will move on to end-stage renal disease. LN occurs when the immune system wrongly attacks the kidneys, preventing them from doing their job, i.e., cleaning blood, balancing body fluids and controlling hormones that impact blood pressure.
Unfortunately, the most precise way to diagnose LN hasn’t been all that precise. The kidney biopsy, which in itself is a painful ordeal, reaches a tipping point at the time doctors must read the biopsy report. Historically those interpretations have been imprecise and marked by significant disagreement among pathologists reading the same thing.
Enter artificial intelligence, which combines computer science and robust datasets to enable problem-solving, and two University of Houston Cullen College of Engineering faculty. Chandra Mohan, Hugh Roy and Lillie Cranz Cullen Endowed Professor of Biomedical Engineering and Hien Van Nguyen, associate professor of Electrical and Computer Engineering, have received a $3 million grant from the National Institute of Diabetes and Digestive and Kidney Diseases to bring AI into the diagnostic picture.
“Given that this critical diagnostic step – which is important for planning treatment – is highly variable and imprecise, we sought out alternative approaches,” said Mohan. “This funding allows us to use artificial intelligence approaches to train a ‘neural network’ to learn how to read and classify lupus nephritis biopsy slides.”
The goal of using AI to classify lupus nephritis in an automated fashion with high accuracy will translate to better treatment for lupus nephritis, according to researchers.
Whereas Mohan is known internationally for his work on lupus nephritis, Nguyen already leads several projects to fully realize the benefits of AI in improving medical diagnosis. The UH team will work closely with renal pathologists including Jan Becker, Cologne, Germany; Luan Truong and Sadhna Dhingra, Houston Methodist; Qi Cai, UT Southwestern, Dallas, Texas; and Surya Seshan, Cornell University, Ithaca, New York.
“By leveraging the power of computer vision and deep learning, a branch of machine learning, we will build classifiers that rival the best renal pathologists in making a diagnosis using current criteria. This could dramatically improve patient management and long-term renal and patient outcome,” said Mohan.
“This collaborative effort exemplifies how AI and medical expertise can intersect to drive innovation, and I want to extend my gratitude to the hard-working team members who are committed to pushing the boundaries of what AI can do in the field of lupus nephritis,” said Nguyen.
