A Review Paper on Innovative Deep Learning and Machine Learning Algorithms for Improved Cervical Cancer Identification
Prachi N. Shah, Rais Abdul Hamid Khan · 2024
Cervical cancer is a major problem, especially in nations with low or middle incomes where access to cutting-edge healthcare and regular screening is constrained. Early detection and precise diagnosis are essential for effective therapy and higher survival rates. The identification of cervical cancer has shown outstanding and more accurate results with recent methods in machine learning (ML) and deep learning (DL). More accurate and efficient diagnostic tools are required since the use of conventional screening methods, like Pap smears and HPV tests, is restricted. The research looks into feature extraction techniques like Histogram of Oriented Gradients (HOG) and Gray Level Co-occurrence Matrix (GLCM) in addition to machine learning approaches like support vector machines (SVM), random forests, and k-nearest neighbors (k-NN). Convolution neural networks (CNNs) and transfer learning are addressed in order to achieve better performance on fewer datasets. Recurrent neural networks (RNNs) and generative adversarial networks (GANs), two cutting-edge deep learning techniques, are also studied. Some of the future directions include the development of individualized medical strategies, the application of ML and DL models to clinical practice, and the fusion of multiple communication data sources to increase diagnosis accuracy.