AI Blog

Taking artificial intelligence out of the black box: An “interpretable” deep learning system for liver tumour diagnosis

European Radiology May 2019

Convolutional neural networks (CNN) have demonstrated the potential to become effective and accurate decision support tools for radiologists. A major barrier to clinical translation, however, is that the majority of such algorithms currently function like a “black box”. After training a CNN with a large set of input and output data, its internal layers are automatically adjusted to mathematically “map”

AI and radiologists working together: an interview with Professor Elmar Kotter

Alex Vogel May 2019

We spoke with Professor Elmar Kotter from the Medical Center – University of Freiburg to explore his thoughts on artificial intelligence (AI) in radiology, the challenges it poses, the legal issues associated with these new technologies, and what the future holds for radiologists. AI taking over the tasks of the radiologist “In order to adapt to the increasing ubiquity of

Using machine learning to predict cervical lymph node metastasis on dual-energy CT

Reza Forghani May 2019

In this article, we extracted “hand-crafted” radiomic features from dual-energy CT (DECT) virtual monochromatic images (VMIs) reconstructed at different energies and used machine learning to construct prediction models that use the radiomic features of head and neck squamous cell carcinoma (HNSCC) to predict associated nodal metastases. This proof of concept study demonstrated that (1) HNSCC radiomic features can predict associated

Assessing the impact of AI in radiology management

Melisande Rouger May 2019

As the demand for time-consuming imaging services keeps on growing, radiology departments are increasingly vulnerable to staff shortages. Artificial intelligence (AI) can be a solution in many management scenarios, but leaders must address the enduring apprehension and define which tools are relevant when making the most of the new technology. A world-famous leader shared timeless advice on how to tackle

CT reconstruction algorithms affect histogram and texture analysis: evidence for liver parenchyma, focal solid liver lesions, and renal cysts

European Radiology May 2019

The aim of this study was to determine the effects of different reconstruction algorithms on histogram and texture features in different targets. The authors reconstructed CT images with filtered back-projection (FBP), hybrid iterative reconstruction (HIR), and iterative model reconstruction (IMR) algorithms and performed computerized histogram and texture analyses by extracting 11 features. The results enabled the authors to conclude that

AI-assisted Multi-Organ Image Interpretation

ESR May 2019

Advertorial The broader implementation of artificial intelligence (AI) will be based on versatile systems that can be seamlessly integrated into existing workflows and IT architectures. Assessing multiple anatomical structures and organs on a chest CT more quickly and precisely would be one strategy to make AI support a self-evident aspect of image interpretation. A crucial prerequisite for advancing the implementation

Differentiation of clear cell and non-clear cell renal cell carcinomas by all-relevant radiomics features from multiphase CT: a VHL mutation perspective

European Radiology April 2019

The main goal of this retrospective two-center study was to develop a radiomics model with all-relevant imaging features from multiphasic computed tomography (CT) for differentiating clear cell renal cell carcinoma (ccRCC) from non-ccRCC. The authors also set out to investigate the possible radiogenomics link between the imaging features and a key ccRCC driver gene—the von Hippel-Lindau (VHL) gene mutation. Using

What the radiologist should know about artificial intelligence – an ESR white paper

Daniel Pinto dos Santos April 2019

Did you ever get lost between all the buzz words floating around like artificial intelligence, imaging informatics, radiomics, imaging biobanks, neural networks, clinical decision support, etc.? And what are the ethical and medico-legal implications of all these new developments? Then you may want to read the ESR white paper that was recently published by the ESR’s eHealth and Informatics subcommittee

How machine learning-based high-dimensional CT texture analysis is influenced by segmentation margin

Burak Kocak April 2019

Radiomic workflows include various challenging steps. One of the most demanding steps in radiomics is the segmentation process. Particularly for the renal cell carcinomas, most of the studies used manual tumour contour delineation. In this work, our group wanted to perform an experiment by changing the segmentation margin a little bit, that is, just 2 mm, to see what happens

Strategic research agenda for biomedical imaging

Daniel Pinto dos Santos April 2019

Over the last years, medicine has been moving further towards providing a more tailored, patient-centric approach by taking into account as much information as possible to deliver personalised solutions for the individual patient; certainly, radiology and other imaging-based technologies have facilitated this to a great extent. But what will be the way for the future? What are the challenges that

Welcome to the blog on Artificial Intelligence of
the European Society of Radiology

Latest posts