Gastroenterology Meets Machine Learning: Status Quo and Quo Vadis

Table 3

ML applications in medical domains.

Medicine domain	ML applications	References
Radiology	Radiological imaging tasks such as: (i) Risk stratification.; (ii) Therapy response.; (iii) Lesions segmentation and classification.; (iv) Multi-omics disease discovery.; (v) Discovery of radiographic imaging biomarkers.; (vi) Creating study protocols.	[22–26]
Pathology	Digital pathological image analysis notably: (i) Tissue phenomics.; (ii) Histopathological imaging analysis.; (iii) Whole Slide imaging analysis.	[14, 27, 28]
Oncology	Early cancer diagnosis and prognosis: (i) Cancer metastases detection.; (ii) Molecular subtyping of cancer.; (iii) Cancer detection from microarray gene expression data; (iv) Risk classification of cancer survival.	[29–31]
Cardiology	Early detection of cardiovascular diseases based on: (i) Electrocardiographic interpretation.; (ii) Echocardiography interpretation.; (iii) Myocardial perfusion analysis.; (iv) Discrimination of different diseases with similar symptoms like constrictive pericarditis and restrictive cardiomyopathy or hypertrophic cardiomyopathy and physiological hypertrophy.	[32–34]
Neurology	Neurological disorders identification and prediction: (i) Electroencephalography data interpretation; (ii) Electromyography data interpretation; (iii) Augmented Intelligence such as: (iv) Restoring the control of movement in patients with quadriplegia.; (v) Controlling upper-limb prostheses via Brain-computer interface.	[17, 35]