[Paper Review] Artificial Neural Network Based Breast Cancer Screening: A Comprehensive Review

Subrato Bharati, Prajoy Podder+1 more|arXiv (Cornell University)|May 29, 2020

AI in cancer detection80 references69 citations

TL;DR

A systematic review of ANN-based models for breast cancer screening via mammography, detailing architectures (SNN, DBN, CNN, MLNN, SAE, SDAE) and concluding that ResNet-50 and ResNet-101 CNNs often achieve best performance on public datasets.

ABSTRACT

Breast cancer is a common fatal disease for women. Early diagnosis and detection is necessary in order to improve the prognosis of breast cancer affected people. For predicting breast cancer, several automated systems are already developed using different medical imaging modalities. This paper provides a systematic review of the literature on artificial neural network (ANN) based models for the diagnosis of breast cancer via mammography. The advantages and limitations of different ANN models including spiking neural network (SNN), deep belief network (DBN), convolutional neural network (CNN), multilayer neural network (MLNN), stacked autoencoders (SAE), and stacked de-noising autoencoders (SDAE) are described in this review. The review also shows that the studies related to breast cancer detection applied different deep learning models to a number of publicly available datasets. For comparing the performance of the models, different metrics such as accuracy, precision, recall, etc. were used in the existing studies. It is found that the best performance was achieved by residual neural network (ResNet)-50 and ResNet-101 models of CNN algorithm.

Motivation & Objective

• Survey and synthesize literature on ANN-based breast cancer diagnosis from mammography.
• Describe advantages and limitations of different ANN models.
• Summarize datasets and evaluation metrics used in the literature.
• Identify leading architectures with best reported performance.
• Discuss future directions for ANN-based breast cancer screening.

Proposed method

• Systematic literature review of ANN models used in mammography for breast cancer screening.
• Categorization of models (SNN, DBN, CNN, MLNN, SAE, SDAE) and their characteristics.
• Review of publicly available datasets and common evaluation metrics (e.g., accuracy, precision, recall).
• Comparison of model performance across studies, highlighting trends and gaps.

Experimental results

Research questions

• RQ1 What ANN architectures have been applied to breast cancer screening in mammography?
• RQ2 How do these architectures perform across publicly available datasets using standard metrics?
• RQ3 What are the advantages and limitations of each ANN model in this application?
• RQ4 Which model types emerge as top-performing in the reviewed literature, and under what conditions?

Key findings

• The review describes several ANN models used for breast cancer detection in mammography, including SNN, DBN, CNN, MLNN, SAE, and SDAE.
• The literature shows that CNN-based approaches are commonly employed across datasets.
• The best reported performance is achieved by residual CNN models (ResNet-50 and ResNet-101).
• Studies utilize various public datasets and metrics such as accuracy and precision/recall to evaluate models.
• The review discusses advantages and limitations of each ANN model and highlights data availability and generalization challenges.