Sanjay Gandhi
Southmead Hospital, UK
Title: The role of computer-assisted detection (CAD), artificial intelligence and machine learning in the medical imaging data overload problem: Current practice, limitations and new developments
Biography
Biography: Sanjay Gandhi
Abstract
The demand for diagnostic imaging has continued to increase dramatically over the past few decades. More than 5 billion diagnostic tests are performed globally each year. Replacement of older modalities such as Barium enema with CT colonoscopy and isotopeVQ-scans with CT pulmonary angiography has also massively contributed to increased image datasets per examination. As a result, the majority of Radiologists and other healthcare professionals have to review tens of thousands of images every day. Fortunately, technologies such as Computer Aided Detection (CAD), 3D processing and automated image analysis have also continued to develop. These are becoming increasingly more reliable and affordable. Computer-assisted polyps and cancers detection on virtual-colonography, nodules on lung cancer screening and analysis of breast lumps on MR mammography are just a few of the examples. We will discuss the accuracy and use of different CAD programmes. Our research has shown that a large variation exists in sensitivity and PPV of commercially available software. Some programmes suffer from very long analysis times. Hence, companies producing CAD tools need to address these issues. The role and potential of new technologies such as Artificial Intelligence (AI) and machine learning in coping with the massive increase in the medical imaging workload will be explored. This talk will cover the pitfalls and provide practical tips on the use of these techniques. Such information is useful for Radiologists and Radiographers/Technicians. In addition, CAD developers and other healthcare sector’s entrepreneurs might find this discussion useful in order to develop future products.