Data Science is all Set to Revolutionize the Ultrasound Diagnosis in Medical Health Care

Abstract

Ultrasound (US) with CAD system imaging is a medical diagnostic tool which has become the most commonly performed cross-sectional diagnostic imaging technique in the practice of medicine, and also the one of the most common schemes in the medical diagnostic area to detect diseases in the clinical practice. It is also cheapest, non-ionizing, reliable, portable, and capable of real-time image acquisition, safety, convenience, and low risk, live display makes it acceptance worldwide. US is oldest but with the cult of computer science, data science (machine learning, deep learning) and chip level advancement in the ultrasound devices. This Ultrasound is all set to upgrade and change according to the advancement in evolving technology, information technology with significant challenges and opportunities. High inter- and intra-operator inconsistency and limited image quality management are the few trending issues which are prime focus of the researchers in present days. As mentioned earlier the US devices due to cult introduction of the computer science and data science has become smaller, enhanced computational capability, and become very advanced. Ultrasound also can contribute significantly to decrease changeability with its advanced image processing. However, reading ultrasound imaging is little tricky, in a proposal to make it little supportive to the clinicians and to reduce the load of doctors, many ultrasound computer-aided diagnosis (CAD) systems with different segmentation, classifiers techniques have been launched. In recent years, the success of data science especially deep learning in the image classification and segmentation is very successful. This led to more and more scholars and researchers to think unanimously realizing that improvement in accuracy, Specificity, and sensitivity of the disease finding can be brought by utilizing the deep learning in the ultrasound CAD system. This paper reviews the research which focuses on the ultrasound CAD system utilizing Data science in recent years. Ultrasound CAD system are mainly 1. traditional ultrasound CAD system and deep learning ultrasound CAD system. The different segmentation techniques and the classifier employed in ultrasound CAD system are discussed. Performance of the different CAD systems with different techniques has accounted in the research paper in contrast accuracy, sensitivity, and Specificity in disease finding is noted. This paper will be useful for the researchers who focus on the ultrasound CAD system more over it is the base of the academic research for the fulfillment of the PhD research work.