Table 4.

Examples of different applications of machine learning applications for different fields of neuropathological challenges

Name of study	Reference	Model	Outcome	Relevant field
3D segmentation of glial cells using fully convolutional networks and k-terminal cut	(37)	CNN	High accuracy (F1 0.89) in 3D segmentation of glial cells	Neuropathology, basic research
Code-free machine learning for classification of central nervous system histopathology images	(80)	CNN	High precision in detecting various brain tumor entities (e.g. glioma, subtypes of glioma, metastasis)	Neuropathology, neurooncology
Deep neural networks segment neuronal membranes in electron microscopy images	(38)	CNN	Superhuman pixel error rate (60 × 10−3) on segmenting neuronal membranes for studying connectomes	Basic research
Artificial intelligence in neuropathology: deep learning-based assessment of tauopathy	(81)	CNN	High accuracy in detecting neurofibrillary tangles for diagnosis of tauopathies	Neuropathology, neurodegeneration
Automated brain histology classification using machine learning	(82)	CNN	Low- versus high-grade glioma classification of brain histology slides	Neuropathology, neurooncology
Near real-time intraoperative brain tumor diagnosis using stimulated Raman histology and deep neural networks	(83)	CNN	Intraoperative diagnosis of brain tumors from hematoxylin and eosin-stained (H&E) specimens	Neurooncology, neurosurgery, neuropathology

The papers listed here show some of the diverse use cases of neural networks in supporting researchers and pathologists in their work and research projects.

CNN, convolutional neural network.