Imagine a world where Parkinson's disease could be detected much before the initial symptoms start appearing. This is the promise of researchers Racha Soubra, RHU alumna (class 2017 MS biomedical engineering), and Aly Chkeir, both assistant professors at the University of Technology of Troyes-France (UTT). They are working on a project called AMPIATI "Anticipation of Parkinson's Disease by Artificial Intelligence and Image Processing" that can use artificial intelligence (AI) and medical imaging to intervene at an early stage and change the course of Parkinson's disease affecting millions of people across the globe.
"We are currently working on the identification of various biomarkers that will enable us to classify the five stages of the disease progression after diagnosis," explained Racha and Aly. "Our objective is to find an indicator of stage 0, which refers to the preclinical stage. This indicator will be utilized for screening the disease before the first neurons that produce dopamine are lost."
To find their target biomarkers, analyzing the volume and the morphology of critical brain structures such as substantia nigra and the striatum using medical brain images has been proposed as a first step to gain insights into the disease's progression and early-stage detection.
During their master's internship at UTT, two graduates from RHU, Jad Baalbaki and Mohamed El Zein, have accomplished the challenge of automatically segmenting the striatum. Their work has resulted in a scientific paper entitled "3D MRI Volume Segmentation Using 2D U-Net Models: A Focus on Deep Brain Structures of the Striatum," which has been submitted and accepted for publication and presentation at the MeMeA conference.