AI Blog

Deep learning–based identification of spine growth potential on EOS radiographs

European Radiology November 2023

In this study, the authors developed a deep learning-based algorithm, which is able to mimic human judgment, in order to help clinicians assess the potential of spine growth based on EOS radiographs. The outcome of the study showed that their deep learning method achieved comparable, and even superior, results compared to those of clinicians, which should have positive applications in

Beyond diagnosis: is there a role for radiomics in prostate cancer management?

Andrea Ponsiglione November 2023

At present, therapeutic and prognostic recommendations for prostate cancer (PCa) predominantly hinge on risk-stratification tools that are built upon clinical parameters. Recent evidence indicates that incorporating imaging can enhance the precision of prognostic models based on clinical factors. However, challenges like subjective interpretation, variability in image analysis, and the absence of reliable quantitative measures need to be overcome to fully

Does deep learning improve the consistency and performance of radiologists in assessing bi-parametric prostate MRI?

ESR November 2023

Our study aimed to evaluate whether deep learning (DL) software could enhance the consistency and performance of radiologists in assessing bi-parametric prostate MRI scans. Intriguingly, our findings revealed that the DL software did not significantly improve the Prostate Imaging-Reporting and Data System (PI-RADS) scoring consistency or the detection performance of clinically significant prostate cancer (csPCa) among radiologists with varying levels

Are deep models in radiomics performing better than generic models?

Aydin Demircioğlu October 2023

In radiomics, deep learning methods are increasingly used because they promise higher predictive performance than models based on generic, hand-crafted features. However, sample sizes in radiomics are often small, and it is known that the performance of deep learning models is often critically dependent on sample size. Therefore, it is unclear whether deep models can outperform generic models. In our

AI in thoracic imaging: the transition from research to practice

European Radiology October 2023

This commentary explores the use of artificial intelligence tools that are increasingly being implemented into clinical practice within the field of radiology. From using deep learning as a second reader and assistant to radiologists in reading chest x-rays to the financial impact of these tools, the authors look at the adoption of AI in radiology and its usefulness. Article: Artificial intelligence

QuantImage v2: physician-centered cloud platform for radiomics and machine learning research

European Radiology Experimental October 2023

The authors of this study proposed a physician-centered vision of radiomics research to help aid the translation of radiomics prediction models into clinical practice and workflow. Free-to-access radiomics tools and frameworks were reviewed to help identify best practices and gaps in the existing software. The authors designed QuantImage V2 (QI2) to help implement this vision and address any issues they

A deep learning model using chest X-ray to identify TB and NTM-LD patients

Insights into Imaging October 2023

The authors of this study aimed to evaluate whether artificial intelligence, specifically a deep neural network (DNN), was able to distinguish between tuberculosis (TB) or nontuberculous mycobacterial lung disease (NTM-LD) patients through chest X-rays (CXRs) from suspected mycobacterial lung disease. A total of 1,500 CXRs from two hospitals were retrospectively collected and evaluated. They determined that the developed DNN model

Evaluating the impact of a novel denoising algorithm on image quality in dual-energy abdominal CT of obese patients

Fides Regina Schwartz September 2023

In the ever-evolving landscape of radiology, the quest for enhanced image quality and reduced noise, particularly in obese patients, remains an enduring challenge. It is within this context that a novel algorithm for noise reduction in dual-source dual-energy (DE) CT imaging is a promising development. In a retrospective study involving seventy-nine patients with contrast-enhanced abdominal imaging, this novel algorithm was

Unlocking the Potential of Radiomics: Ongoing Challenges Are Revolving Around Methodology and Reproducibility

Hyun Soo Ko September 2023

Over a decade in the making, the novel concept of radiomics has been silently brewing, promising to reshape the landscape of personalised and precision medicine. So, why has radiomics not made its clinical debut yet, despite its innovative and logical approach? The answer lies in the intricate world of advanced computation that forms the basis of radiomics, which in essence

Evaluation of techniques to improve a deep learning algorithm for the automatic detection of intracranial haemorrhage on CT head imaging

European Radiology Experimental September 2023

This study evaluates deep learning (DL) algorithms that are playing an increasingly important role in automatic medical image analysis. The DL algorithm used was trained and externally evaluated on open-source, multi-centre retrospective data that contained radiologist-annotated non-contrast CT head studies. The authors concluded that the DL model has applications in a triage role with the potential to improve diagnostic yield

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

Latest posts