Open Access

ARTICLE

Prediction of distant recurrence in breast cancer using a deep neural network

Balqis Mohd Azman^1,2, Saiful Izzuan Hussain^1,2, Nor Aniza Azmi³, Muhammad Zahin Athir Abd Ghani³, Nor Irfan Danial Norlen³

1 Department of Mathematical Sciences, Faculty of Science and Technology
2 Universiti Kebangsaan Malaysia, 43600 Bangi, Selangor, Malaysia
3 Diagnostic Imaging and Radiotherapy Program, School of Diagnostic and Applied Health Sciences, Faculty of Health Sciences, Universiti Kebangsaan Malaysia, Jalan Raja Muda Abdul Aziz, 50300 Kuala Lumpur, Malaysia

* Corresponding Author: Saiful Izzuan Hussain ()

Revista Internacional de Métodos Numéricos para Cálculo y Diseño en Ingeniería 2022, 38(1), 1-10. https://doi.org/10.23967/j.rimni.2022.03.006

Accepted 09 March 2022; Issue published 22 March 2022

Abstract

Breast cancer is the most common cancer diagnosed in women, and it is ranked as the second highest cancer with high mortality rate. Breast-cancer recurrence is the cancerous tumor that returned after treatment. Cancer treatments such as radiotherapy are performed mainly to kill cancer cells; however, some cells may have survived and multiply themselves at the same area as the original cancer (local recurrence) or to any other part (distant recurrence). Distant recurrence occurs when cancer cells spread to other parts of the body, most commonly to bone, breast, liver, and lungs. This study employed an Artificial Neural Network of the deep learning approach to predict distant recurrence of breast cancer. Factors that contribute to the risk of recurrence are age, type of surgery performed, tumor size, breast subtype, estrogen receptor, progesterone receptor, undergoing chemotherapy or not, and lymph node involvement. The actual value of distant recurrence is also considered to be a variable. Principal Component Analysis using five and three principal components was conducted. The outcome indicates that the model has accuracy of up to 0.80 using three principal components.

---

Keywords

---