International Journal For Multidisciplinary Research

E-ISSN: 2582-2160     Impact Factor: 9.24

A Widely Indexed Open Access Peer Reviewed Multidisciplinary Bi-monthly Scholarly International Journal

Call for Paper Volume 7, Issue 3 (May-June 2025) Submit your research before last 3 days of June to publish your research paper in the issue of May-June.

Breast Cancer Diagnosis: Integrating Constructive Deep Neural Network

Author(s) Ms. Chauhan Janhvi Arvindbhai, Mr. Modi Dhaval Maheshkumar
Country India
Abstract The Oncotype DX (ODX) breast cancer assay is the most widely used Gene Expression Profiling (GEP) test globally. It plays a significant role in guiding decisions regarding Adjuvant Chemotherapy (ACT). Despite the availability of several standard approaches for this purpose, their accuracy has yet to reach optimal levels. This paper focuses on Breast Cancer Computer-Aided Diagnosis (BC-CAD) using a Deep Constructive Neural Network to predict the Recurrence Score (RS) provided by the ODX assay. The proposed ConstDeepNet algorithm was evaluated by developing two types of classifiers: the first uses a "one-against-all" architecture, building a separate Deep Neural Network for each class, while the second employs a single DNN to classify all three classes. A separate network is constructed for each class in the first architecture, while the second architecture utilizes a single deep neural network to classify all three classes. The proposed BC-CAD algorithm was evaluated on a real-world dataset and demonstrated strong performance. The dataset consists of 92 cases of luminal B mammary carcinoma with available Oncotype DX test results collected between 2012 and 2017 from the Georges Francois Leclerc Cancer Centre.
Keywords deep learning, neural networks, breast cancer, recurrence score, oncotype DX
Field Engineering
Published In Volume 3, Issue 1, January-February 2021
Published On 2021-01-08

Share this