Machine Learning Algorithms for Smart Healthcare: Breast Cancer Prediction

Background: Breast cancer continues to be a significant concern for women worldwide, emphasizing the crucial role of early detection in improving treatment outcomes. The machine learning techniques play an important role in early detection of breast cancer. Detection of breast cancer with traditional techniques is time-consuming and does not provide accurate results. So to enhance the accuracy of detection and to decrease time taken, an automatic breast cancer technique is required. This work introduced innovative machine-learning algorithms for the detection of breast cancer. Machine learning algorithms used in this paper are Neural Networks (NN), Logistic Regression (LR), Support Vector Machine (SVM), Decision Tree (DT), Random Forest (RF), and K-Nearest Neighbor (KNN) with a widespread set of related parameters. The dataset taken for the work contains many patient demographics, clinical details, and histopathological information to train and validate machine learning algorithms. For the evaluation of the work accuracy and loss are calculated. The Logistic Regression outperformed with remarkable accuracy 97.4%, over the mentioned five other algorithms. All five models are compared to verify and validate the experimental results. The proposed work requires very less human intervention and attained high accuracy in considerable time. In future, for validation and its applications in clinical settings, the author can design a hybrid technique that can handle real-time clinical data.