AI Blog

Motion-corrected coronary calcium scores by a CNN: a robotic simulating study

European Radiology November 2019

The authors of this study aimed to classify motion-induced blurred images of calcified coronary plaques, in order to correct coronary calcium scored on non-triggered chest computed tomography (CT). They did so by using a deep convolutional neural network (CNN) which was trained using a selection of images of motion artifacts. Key points A deep CNN architecture trained by CT images

Why I Stopped Worrying And Love The Intelligent Machine

ESR November 2019

The future will always be just that…until it becomes the present. That is especially true when the future involves science. What if science fiction has a way of turning – suddenly – into scientific fact? Like flying machines and spacecraft, one day they existed only in theory, and the next, they became a reality. At the annual meeting of the

Fully automated convolutional neural network-based affine algorithm improves liver registration and lesion co-localization on hepatobiliary phase T1-weighted MR images

European Radiology Experimental November 2019

The authors of this study aimed to assess the performance of a convolutional neural network (CNN) algorithm to register cross-sectional liver imaging series and its performance to manual image registration. The study included three hundred fourteen patients who underwent gadoxetate disodium-enhanced magnetic resonance imaging (MRI) and were retrospectively selected. Key points Image registration across series can improve lesion co-localization and

Looking outside the box: using AI to help elephants, tracking bird populations, and fighting food waste

ESR November 2019

This week in artificial intelligence (AI) news, we take a look at African parks using AI to fight poachers and help the declining elephant population, monitoring bird populations using deep learning, and how a company in the United Kingdom is using technology to combat food waste more efficiently. In this article from NPR, researchers in Africa are taking advantage of

A machine learning model for the prediction of survival and tumor subtype in pancreatic ductal adenocarcinoma from preoperative diffusion-weighted imaging

European Radiology Experimental November 2019

The aim of this study was to develop a supervised machine learning (ML) algorithm that would use diffusion-weighted imaging-derived radiomic features to predict median overall survival in patients with pancreatic ductal adenocarcinoma. Based on the evaluation of 132 patients, it was determined that the use of ML allowed the prediction of overall survival with high diagnostic accuracy. Key points Pancreatic

Deep learning to convert unstructured CT pulmonary angiography reports into structured reports

John Nance October 2019

We believe that the possibilities for artificial intelligence (AI) over the coming years will be limited only by our imagination. While there is a tremendous amount of warranted excitement for disease detection, characterization, and quantification with AI algorithms, less rousing but still valuable efforts can and should be made to improve operational efficiency and even reduce the growing problem of

Looking outside the box: anxiety over future employment, tricking AI systems, and an autonomous journey across the Atlantic

ESR October 2019

This week in artificial intelligence (AI) news, we take a look at a poll showing the lack of confidence individuals have in their preparation for a world run by AI, the fragility of deep neural networks and how to fool them, and autonomously retracing the journey of the Mayflower. In a recent poll by Northeastern University and Gallup, a disconcerting

MRI-based radiomics nomogram may predict the response to induction chemotherapy and survival in locally advanced nasopharyngeal carcinoma

European Radiology October 2019

This study integrated the clinical data and radiomics signature generated by a support vector machine to establish a radiomics nomogram for prediction of induction chemotherapy response and survival in nasopharyngeal carcinoma patients. The results proved that multiparametric MRI-based radiomics could be helpful for personalized risk stratification in patients receiving induction chemotherapy. Key points MRI Radiomics can predict IC response and

AI is a means, not a goal

Melisande Rouger October 2019

Artificial intelligence (AI) software solutions developed for radiology are increasingly focused on answering precise clinical questions, rather than just detecting lesions. This is the most noticeable trend in medical imaging AI, which continues to bring new products to the market with clockwork regularity. Whether it feeds on machine or deep learning, AI in radiology really boils down to one thing:

Noncontrast computer tomography–based radiomics model for predicting intracerebral hemorrhage expansion: preliminary findings and comparison with conventional radiological model

European Radiology October 2019

The authors developed a radiomics model for predicting hematoma expansion in patients with intracerebral haemorrhage (ICH) and compared its predictive performance with a conventional radiological feature-based model. Through retrospective analysis and noncontrast computed tomography (NCCT) assessment, it was found that an NCCT-based radiomics model showed better performance in the prediction of early hematoma expansion in ICH patients. Key points Radiomics

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

Latest posts