Diagnosing the neurodegenerative disorder called Parkinson’s disease (PD) is based on the symptoms that could be found only when 60% of dopamine content is lost (advanced stage) in the midbrain. The disease has no clinical tests in its earlier stage which leads to investigate an innovative method to diagnose PD in its early stage. The novelty of the work is to frame a mathematical model for the neurotransmitter called Gamma Amino Butyric Acid (GABA) to diagnose PD in an early stage. The mathematical model is framed using Regression Analysis (RA) and Neural Network (NN) as a fitting tool, considering Striatal Binding Ratio value (SBR), a quantitative measure of dopamine, as input parameters for the model. The model performance is valued using normal probability curve, analysis of variance (ANOVA), and residual plots for RA, regression, and Mean Square Error (MSE) plot for NN fitting method. This ensures that the model is highly significant in identifying the disease. Extreme Learning Machine (ELM) with the kernel function called Radial Basis Function (RBF) offers better performance of accuracy 95.34% than the other classifiers. The mathematical equation is constructed which aids the neurologists to find out the early stage disease.

诊断称为帕金森病 (PD) 的神经退行性疾病是基于只有当中脑中 60% 的多巴胺含量丢失（晚期）时才能发现的症状。该疾病在其早期没有临床测试，这导致研究一种在早期诊断 PD 的创新方法。这项工作的新颖之处在于为称为伽马氨基丁酸 (GABA) 的神经递质构建了一个数学模型，以在早期诊断 PD。使用回归分析 (RA) 和神经网络 (NN) 作为拟合工具构建数学模型，将纹状体结合率值 (SBR)（一种多巴胺的定量测量）作为模型的输入参数。使用正态概率曲线、方差分析 (ANOVA) 和 RA、回归、和 NN 拟合方法的均方误差 (MSE) 图。这确保了该模型在识别疾病方面非常重要。具有称为径向基函数 (RBF) 核函数的极限学习机 (ELM) 提供比其他分类器更好的 95.34% 精度性能。数学方程的构建有助于神经科医生找出早期疾病。