Eastern Nazarene College Assistant Professor of Biology, Robert Logan, was the invited leading author of a high impact review paper recently published by the journal Frontiers in Aging Neuroscience as part of the special topic series “Progress of Translational Medicine in Alzheimer’s disease”. Frontiers in Aging Neuroscience is the 5th most cited open-access journal within the neurosciences, and the Frontiers in Neuroscience journal series are the highest cited journals within the neuroscience discipline. Professor Robert Logan’s paper details the recent advancements in deep learning for the classification of Alzheimer’s disease from clinical image data. Non-invasive neuroimaging, such as from magnetic resonance imaging (MRI) and position emission tomography (PET), hold the promise of establishing an early diagnosis of neurodegeneration, even before symptoms appear.
“There is an urgent need for the development of robust early detection mechanisms for chronic diseases that have no known cure, such as neurodegeneration. Once symptoms appear in patients with Alzheimer’s disease or Parkinson’s disease, the damage to the brain is too significant to be reversed, halted or even reasonably slowed down with the currently available medications. It is imperative that early screening and detection become commonplace in healthcare so that measures can be taken early enough in the pathological development to effectively treat neurodegeneration before it becomes too late”, says Robert Logan.
The study was an exhaustive review and analysis of how machine learning can be applied to identifying pathological brain changes among patients even before the onset of clinical symptoms. The paper covered topics such as the use of multilayer perceptron neural networks, convolutional neural networks, generative adversarial networks, and ensemble learning in the detection of Alzheimer’s disease. Although the focus of the paper was on machine learning, it was written for an audience of neuroscientists and clinicians rather than computer scientists, in the hope of encouraging a more interdisciplinary approach to healthcare practices.
“This publication offers a step forward in tackling Alzheimer’s disease, which is a very complex disease without an available cure. Therefore, it continues to aggressively claim lives on a global scale. The fruitful and valuable collaboration between ENC and PDx (PluriPotent Diagnostics Corp.) takes a novel interdisciplinary approach for early disease detection. Our models incorporate the most advanced technologies to date, to provide the most accurate brain scan detection of Alzheimer’s disease development. Early diagnosis allows for early intervention.”, says Sean J. Miller, the senior author of the paper.
Robert Logan and Sean J. Miller met each other while they both worked at the Massachusetts General Hospital and Harvard Medical School. Dr. Miller is the C.E.O. and C.S.O. of PluriPotent Diagnostics Corporation, based out of the Molecular Medicine Research Institute in Sunnyvale, California. Professor Robert Logan is the C.O.O. of PluriPotent Diagnostic Corporation. Dr. Miller and Dr. Logan are both co-founders of PluriPotent Diagnostics Corporation and Dr. Miller is also a co-investigator in the Logan Lab at ENC.
Citation: Logan R, Williams BG, Ferreira da Silva M, Indani A, Schcolnicov N, Ganguly A and Miller SJ (2021) Deep Convolutional Neural Networks With Ensemble Learning and Generative Adversarial Networks for Alzheimer’s Disease Image Data Classification. Front. Aging Neurosci. 13:720226. doi: 10.3389/fnagi.2021.720226
Pluripotent Diagnostics Corp.: www.pluripotentdiagnostics.com
The Logan Lab at ENC: www.theloganlab.com