Moving towards automated digitised image interpretation. Friend or foe?


  • Riaan van de Venter Nelson Mandela University, Department of Radiography


artificial intelligence, deep learning algorithms, machine learning, enhanced patient outcomes, support, digitised pattern recognition, ethico-legal implications


Radiology and radiography have always been at the forefront of rapid technological advancements, requiring flexibility to adapt to different workflow approaches and systems in a short time span. This is no different with the near-reality of the introduction of artificial intelligence (AI), in radiology. Artificial intelligence is automating pattern recognition and image interpretation currently done by humans, and providing a digitised interpretation to the observer. It is not going to replace the knowledge and/or expertise of radiologists or radiographers who still need to analyse and interpret radiographic images. Instead, AI fulfils a supporting function to assist in enhancing accuracy of image interpretation and pattern recognition to ultimately improve patient outcomes. However, the full extent of implications for practice is not yet fully explored; many challenges still require attention and deliberation.


Author Biography

Riaan van de Venter, Nelson Mandela University, Department of Radiography

Associate lecturer at Nelson Mandela University, Department of Radiography






Article of Interest