Artificial Intelligence and Machine Learning in the Diagnosis and Management of Gastroenteropancreatic Neuroendocrine Neoplasms—A Scoping Review

Abstract

Neuroendocrine neoplasms (NENs) and tumors (NETs) are rare neoplasms that may affect any part of the gastrointestinal system. In this scoping review, we attempt to map existing evidence on the role of artificial intelligence, machine learning and deep learning in the diagnosis and management of NENs of the gastrointestinal system. After implementation of inclusion and exclusion criteria, we retrieved 44 studies with 53 outcome analyses. We then classified the papers according to the type of studied NET (26 Pan-NETs, 59.1%; 3 metastatic liver NETs (6.8%), 2 small intestinal NETs, 4.5%; colorectal, rectal, non-specified gastroenteropancreatic and non-specified gastrointestinal NETs had from 1 study each, 2.3%). The most frequently used AI algorithms were Supporting Vector Classification/Machine (14 analyses, 29.8%), Convolutional Neural Network and Random Forest (10 analyses each, 21.3%), Random Forest (9 analyses, 19.1%), Logistic Regression (8 analyses, 17.0%), and Decision Tree (6 analyses, 12.8%). There was high heterogeneity on the description of the prediction model, structure of datasets, and performance metrics, whereas the majority of studies did not report any external validation set. Future studies should aim at incorporating a uniform structure in accordance with existing guidelines for purposes of reproducibility and research quality, which are prerequisites for integration into clinical practice.

Updates on lung neuroendocrine neoplasm classification

“…Using digital quantification algorithms, the assessment of Ki‐67 in NENs have shown similar, if not better, performances compared to manual counting in many studies, 78,80–82 including in those exclusively conducted on pulmonary NENs 83–85 . Similarly, computer‐assisted mitotic count has also shown good performance in many tumour entities, 86,87 including NENs 81 …”

“…78 In these settings, positivity for Ki-67 or PHH3 is analysed through algorithms detecting nuclei by morphological shape and size and then classifying cells as positive or negative based on pixel colour and intensity. 79 Using digital quantification algorithms, the assessment of Ki-67 in NENs have shown similar, if not better, performances compared to manual counting in many studies, 78,[80][81][82] including in those exclusively conducted on pulmonary NENs. [83][84][85] Similarly, computer-assisted mitotic count has also shown good performance in many tumour entities, 86,87 including NENs.…”

“…[83][84][85] Similarly, computer-assisted mitotic count has also shown good performance in many tumour entities, 86,87 including NENs. 81 These promising data notwithstanding, it is important to remember that the implementation of DIA algorithms in routine clinical practice is still far from being achieved, with further studies needed to validate standard models, define image storage policies and assess the pathologist's responsibilities. 88…”

Molecular factors, diagnosis and management of gastrointestinal tract neuroendocrine tumors: An update

“…The role of artificial intelligence and machine learning and a deep understanding of the diagnosis and management of NENs of the gastrointestinal system have been recently evaluated. A standard of practice has not yet been established; however, it may serve as a useful adjunct in current practices[ 105 ].…”

Diagnostic Management of Gastroenteropancreatic Neuroendocrine Neoplasms: Technique Optimization and Tips and Tricks for Radiologists

“…Gastroenteropancreatic neuroendocrine neoplasms (GEP-NENs) account for approximately 1.5% of all gastrointestinal and pancreatic malignancies [ 1 ]. The word “neuroendocrine” derives from the similarity of these cells with the neural crest for the expression of, for example, synaptophysin, neurospecific enolase and chromogranin proteins [ 2 , 3 ]. Although it is known that the incidence is low, the growth is slow, and the prevalence is high, in recent decades, there has been an increase in their incidence rate, also due to the improvement of radiological, nuclear medical, and endoscopic imaging techniques [ 4 , 5 ].…”