Artificial intelligence in predicting neuroendocrine tumour (NET) outcomes: A model-based prognostic approach

#4600

Introduction: Neuroendocrine tumours (NETs) vary widely in clinical behaviour, complicating prognosis. Traditional models often struggle with accuracy due to the complexity of NETs. Artificial intelligence (AI) offers tools for enhanced prognostic precision by analysing complex datasets. This study explores an AI-driven approach to predict outcomes in NET patients at Cantonment General Hospital, Pakistan.

Aim(s): To develop and assess a machine learning model for predicting progression-free survival (PFS) and overall survival (OS) in NET patients, using demographic, clinical, and pathological data.

Materials and methods: This retrospective study included 150 NET patients treated from 2015 to 2023. Key variables such as age, gender, tumour site, stage, grade, treatment, and biomarkers (e.g., chromogranin A, Ki-67 index) were used. Machine learning algorithms – random forest, support vector machine, and neural networks – built predictive models, validated by cross-validation and assessed by metrics like AUC-ROC.

Conference:

Presenting Author: Fatima A

Authors: Fatima A,

Keywords: artificial intelligence, machine learning, Precision medicine, Personalised treatment,