2021 World Conference on Lung Cancer (Posters)-FP08. Artificial Neural Network-Based Tumour Recurrence Prediction in Non-Small Cell Lung Cancer Patients Following Radical Radiotherapy

FP08. Artificial Neural Network-Based Tumour Recurrence Prediction in Non-Small Cell Lung Cancer Patients Following Radical Radiotherapy

Back to course

Video Transcription

Video Summary

The presenter discussed using an artificial neural network (ANN) to predict tumor occurrence in non-small cell lung cancer patients receiving radical radiotherapy. The study used 451 patients for training and testing the network. The ANN outperformed classical logistic regression, with consistent accuracy, precision, sensitivity, specificity, and F1 scores above 70%. Reproducibility was assessed using Monte Carlo cross-validation, and the ANN demonstrated excellent ability to distinguish between dependent and independent variables. The presenter concluded that deep learning with ANN is a promising approach for predicting tumor recurrence and further testing in independent datasets is warranted.

Asset Subtitle

Timothy Mitchell

Meta Tag

Speaker Timothy Mitchell

Topic Multimodality of Advanced Lung Cancer

Keywords

artificial neural network

tumor occurrence

non-small cell lung cancer

radical radiotherapy

predict