Comparative Study of Machine Learning and Deep Learning Models for Early Prediction of Ovarian Cancer

Ovarian cancer remains one of the most difficult gynecological cancers to detect early, often resulting in poor survival rates. This study presents a comparative analysis of machine learning (ML) and deep learning (DL) models for the early prediction of ovarian cancer using clinical and biomarker data. The dataset undergoes comprehensive preprocessing, including handling missing values, outlier removal, normalization, and dimensionality reduction via PCA. Feature selection methods such as Feature Importance, Recursive Feature Elimination (RFE), and autoencoder-based techniques are employed to enhance model performance. Various classifiers, including KNN, SVM, Logistic Regression, Random Forest, and deep networks like ANN, FNN, CNN, and RNN, are evaluated. Ensemble models such as Bagging, AdaBoost, Stacking, and XGBoost are also implemented. Our results show that the Feedforward Neural Network (FNN), combined with autoencoder-based feature selection, achieved the highest accuracy (85.71%), indicating its potential as a reliable predictive model for ovarian cancer. This comparative study highlights the significance of integrating optimized preprocessing, feature engineering, and model selection for effective early diagnosis.