The term artificial intelligence (AI) refers to one branches of computer science that is dedicated create systems able to perform tasks that usually require human intelligence. In recent years, AI has gained relevance in medical imaging. For instance, in radiology the use of AI has been promising results in a wide range of applications, from image processing and acquisition, data storage, computer aided diagnosis, and many others. Machine learning (ML) is a subfield of artificial intelligence that focuses on develop algorithms that can learn and adapt without following explicit instructions. Neural networks, more specifically, artificial neural networks (ANN) are an approach widely used in AI approaches. Their structure and name were inspired by the human brain because it tries to mimic the way that biological neurons send signals to each other. ANN consist of input, hidden, and output layers with interconnected group of nodes or neurons to simulate the human brain. Neural Networks make up the backbone of deep learning (DL) algorithms. Deep learning is a neural network with three or more layers. In fact, the depth, or the number of node layers of ANNs that distinguishes a single neural network from a deep learning algorithm.

The success of deep learning in many pattern recognition applications has been revolutionary because it has brought excitement and high expectations in health field, specifically in medical imaging. The potential of applying deep-learning-based medical image analysis to computer-aided diagnosis (CAD) provides decision support to clinicians. Additionally, it can improve the efficiency and accuracy in the diagnosis and subsequent treatment processes. In the last couple of years, new research and development efforts in CAD has been focused on medical imaging such as tomography (CT) and magnetic resonance imaging (MRI), radiography, mammography, and ultrasound are also interest medical to explore segmentation and classification task in AI.

With the advent of AI in medical imaging, radiologists do not need to know the deepest details of these algorithms. However, it is necessary to know the technical vocabulary used by machine learning engineers and data scientists to efficiently communicate with them to work together to go one step beyond in radiology.

Sources:

European Society of Radiology (ESR) communications@ myesr. org Neri Emanuele de Souza Nandita Brady Adrian Bayarri Angel Alberich Becker Christoph D. Coppola Francesca Visser Jacob. (2019). What the radiologist should know about artificial intelligence–an ESR white paper. Insights into imaging, 10(1), 44.

Shen, D., Wu, G., & Suk, H. I. (2017). Deep learning in medical image analysis. Annual review of biomedical engineering, 19, 221.

Alvaro Andres Sandino
Data Scientist, IMEXHS
Ms.C Biomedical engineer
Universidad Nacional de Colombia
Data Scientist