Artificial Intelligence‐Driven Decisions in Breast Cancer Diagnosis

Abstract

Artificial intelligence (AI) is a disruptive innovation, involving the development and deployment of algorithms to classify data, and to find the abstract relationships between different data points. AI, its subtype machine learning (ML), and advanced version of ML that is deep learning (DL) have shown commendable progress in the last decade in various fields like marketing, banking, e-commerce platforms, streaming platforms, self-driving cars, retrosynthesis of chemicals, clinical trials, drug discovery, and others. This remarkable change has been attributed to the availability of high-speed internet, the internet of things, the huge amount of data, and most importantly cutting-edge computation tools like graphics processing units (GPUs), referred to as GPUs. Although the success of AI in medical imaging is limited initially, now stakeholders are focusing on the deployment of full-fledged AI systems by riding on the success of convolutional neural networks (CNNs) in other areas like image identification through computer vision, generation of an entirely new set of images, and videos based on training and testing datasets. Breast cancer is one of the most common cancers in women worldwide. Various diagnostic tools are there to conform positive or negative cases, still, sure-shot reliance on these tools is doubtful owing to unintentional flaws in scan interpretation by radiologists. Here, in such circumstances, experts are now leveraging AI to aid humans with faster, accurate, and bias-free interpretation much to the relief of patients. In this chapter, how AI is shaping and redefining entire mammogram interpretation has been elaborated besides throwing light on classical tools for diagnosis and interpretation of breast cancer.