
A team of researchers has developed an AI platform to detect neurodegenerative diseases in human brain tissue samples. The AI can detect, for example, chronic traumatic encephalopathy and Alzheimer’s disease.
The researchers, from the Icahn School of Medicine at Mount Sinai, Boston University School of Medicine, and the University of Texas Southwestern Medical Center, wrote about their work in the journal Laboratory Investigation (citation below).
The authors say that their discovery will help researchers develop targeted treatment and biomarkers. They also believe it will result in more accurate diagnoses of complex brain diseases, which hopefully will improve patients’ outcomes.
One Alzheimer’s disease feature is the accumulation of abnormal tau proteins in the brain in neurofibrillary tangles. This also occurs in other neurodegenerative diseases, such as chronic traumatic encephalopathy and other age-related conditions.
Accurately diagnosing neurodegenerative diseases is ‘challenging’
Accurate diagnoses of neurodegenerative diseases are extremely challenging and require highly-trained specialists.
The authors developed and used the Precise Informatics Platform to apply powerful machine learning approaches to digitize microscopic slides. These slides were prepared using the tissue samples of patients who suffered from a wide range of neurodegenerative diseases.
Applying deep learning, these slides were used to develop a convolutional neural network that could identify neurofibrillary tangles directly from digitized images with a high degree of accuracy.
Study leader, John Crary, MD, PhD, a Professor of Pathology and Neuroscience at the Icahn School of Medicine at Mount Sinai, said:
“Utilizing artificial intelligence has great potential to improve our ability to detect and quantify neurodegenerative diseases, representing a major advance over existing labor-intensive and poorly reproducible approaches.”
“Ultimately, this project will lead to more efficient and accurate diagnosis of neurodegenerative diseases.”
According to a Mount Sinai press release:
“This is the first framework available for evaluating deep learning algorithms using large-scale image data in neuropathology.”
“The Precise Informatics Platform allows for data managements, visual exploration, object outlining, multi-user review, and evaluation of deep learning algorithm results.”
Using AI to classify different diseases
Engineers at the Center for Computational and Systems Pathology at Mount Sinai have used mathematical techniques and advanced computer science, together with AI, computer vision, and state-of-the-art microscope technology to more accurately classify many different diseases.
Carlos Cordon-Cardo, MD, PhD, Chair of the Department of Pathology at the Mount Sinai Health System, said:
“Mount Sinai is the largest academic pathology department in the country and processes more than 80 million tests a year, which offers researchers access to a broad set of data that can be used to improve testing and diagnostics, ultimately leading to better diagnosis and patient outcomes.”
Dr. Cordon-Cardo is also a Professor of Pathology, Genetics and Genomic Sciences, and Oncological Sciences at the Icahn School of Medicine.
What is artificial intelligence?
Artificial Intelligence refers to software that makes ‘smart’ devices think like us (humans). It also makes them behave like we do. If it cannot perform at least as well as a human, we should not call it AI, say many experts.
Machine learning is an artificial intelligence application that gives intelligent machines the ability to learn from experience. In other words, they learn and get better (upgrade) on their own, without human help.
Citation
“Artificial intelligence in neuropathology: deep learning-based assessment of tauopathy,” Maxim Signaevsky, Marcel Prastawa, Kurt Farrell, Nabil Tabish, Elena Baldwin, Natalia Han, Megan A. Iida, John Koll, Clare Bryce, Dushyant Purohit, Vahram Haroutunian, Ann C. McKee, Thor D. Stein, Charles L. White III, Jamie Walker, Timothy E. Richardson, Russell Hanson, Michael J. Donovan, Carlos Cordon-Cardo, Jack Zeineh, Gerardo Fernandez, and John F. Crary. Laboratory Investigation (2019). DOI: https://doi.org/10.1038/s41374-019-0202-4.