Detection and quantification of microglial cells and dopaminergi̇c neurons in microscopy images using deep learning approaches

In this study, machine learning approaches such as Faster-RCNN, RetinaNET, and YoloV3 were employed for the automatic detection and quantification of microglial cells and dopaminergic neurons in 3D microscopy images, which are crucial for gaining insights into neurological diseases and assessing therapeutic strategies. The artificial intelligence algorithm exhibited good performance when compared to conventional methods. The AI model was trained using datasets from rodents. Parkinson's disease was investigated using a rodent model. The algorithm achieved a level of accuracy similar to manual counting for microglial cells and dopaminergic neurons across different species. The machine learning method employed demonstrates the ability to capture subtle variations within the datasets. The applicability of this method in neurological research has been substantiated