An Ensemble Learning Model for Multi-Type Cancer Prediction in Clinical Diagnostic Decision Support Systems

Résumé fourni par la source

Different organs of the human body are targeted by cancer, leading to varied effects that cause numerous severe health conditions.The early detection of cancer is essential, as it provides timely clinical admission of the patient, which is necessary for successful treatment.Machine Learning (ML) and Deep Learning (DL) algorithms are being widely used to detect and identify cancer cases, relying on data from multiple disciplines, including medical, biomedical, and bioinformatics.These algorithms can identify meaningful diagnostic patterns and detect cancer cases in complex cancer databases.In this paper, we have suggested an ensemble learning approach that consists of five common ML and DL algorithms to build an ensemble-based Multi-Type Cancer Prediction (eMTCP) model.The algorithms used in the development of the eMTCP model are Naive Bayes (NB), Random Forest (RF), Support Vector Machines (SVM), Convolutional Neural Networks (CNNs), and Long Short-Term Memory (LSTM).The eMTCP is utilized in the development of a Cancer Diagnostic Clinical Decision Support System (CDCDSS).Four cancer diagnostic datasets, namely liver cancer, breast cancer, brain cancer, and cervical cancer, are then used to compare the performance of the eMTCP model.The algorithm with the best potential is eMTCP (stacked ensemble), which yields the best F1-scores across all datasets: breast (0.979), liver (0.765), brain (0.898), and cervical (0.482), demonstrating superior performance in multi-type cancer prediction.CNN and LSTM achieved high stability with superior F1-scores of 0.957 (breast), 0.669 (liver), and 0.853 (brain).CNN outperformed in the case of liver cancer and showed similar performance in other cancer types.The SVM ML model has the lowest Keywords Machine learning (ML), ensemble learning (EL), ensemble learning, Decision Support System (DSS), multi-cancer diagnosis, breast cancer, liver cancer, cervical cancer, brain cancer scores of all: 0.969 (breast), 0.716 (liver), 0.844 (brain), and only 0.153 (cervical), indicating that its representation in the clinical-only dataset is not very reliable.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.