Comparison of ELM, LSTM, and CNN Models in Breast Cancer Classification

• Main Authors: Silvia, Ratna, M., Muflih, Haldi, Budiman, Usman, Syapotro, Muhammad, Hamdani
• Format: Article
• Language: English; English
• Published: INTI International University 2024
• Subjects: QA75 Electronic computers. Computer science; QA76 Computer software; RC0254 Neoplasms. Tumors. Oncology (including Cancer); T Technology (General)
• Online Access: http://eprints.intimal.edu.my/2054/1/jods2024_55.pdf; http://eprints.intimal.edu.my/2054/2/595; http://eprints.intimal.edu.my/2054/; http://ipublishing.intimal.edu.my/jods.html

Classification can significantly impact treatment decisions and patient outcomes. This study evaluates and compares the performance of three machine learning models Extreme Learning Machine (ELM), Long Short-Term Memory (LSTM), and Convolutional Neural Networks (CNN) in breast cancer classification. ELM, known for its fast-learning speed and strong generalization, is compared with LSTM, which is effective in capturing long-term dependencies in sequential data, and CNN, which is renowned for its ability to automatically extract features from images and structured data. The models were trained and tested on a breast cancer dataset, focusing on accuracy and computational efficiency. The results revealed that while CNNs demonstrated better accuracy in feature-rich data, LSTMs excelled in handling sequential data patterns. On the other hand, ELM offers a good balance between training speed and classification performance. This comparative analysis provides valuable insights into the strengths and limitations of each model, contributing to the development of more effective breast cancer diagnostic tools. In this case, LSTM outperformed ELM by 0.91%, outperformed CNN significantly by 3.72%, and outperformed Improved LSTM by 0.91%. This indicate that the LSTM model shows higher accuracy in breast cancer classification.