AI Blog

AI support in MR imaging of incidental renal masses

Alexander Marka September 2024

Our study explores the integration of artificial intelligence (AI) into magnetic resonance (MR) imaging to enhance the differentiation between benign and malignant renal lesions. The findings suggest that AI can significantly improve diagnostic accuracy and cost-effectiveness, addressing a crucial need in radiology. AI has the potential to alleviate pressures on healthcare systems by improving diagnostic efficiency and accuracy. By incorporating

AI applied to MRI reliably detects the presence of meniscus tears

European Radiology August 2024

It is known that meniscus tears are difficult to diagnose on knee MRIs. Therefore, this study reviews and compares the accuracy of convolutional neural networks (CNNs). The authors assessed databases including PubMed, MEDLINE, EMBASE, and Cochrane, finding eleven articles to include in the final review, consisting of over 13,000 patients and over 57,000 images. They concluded that CNN is accurate

Detection of femoropopliteal arterial steno-occlusion at MR angiography

European Radiology Experimental August 2024

This single-centre retrospective study aimed to evaluate a deep learning (DL) algorithm for detecting vessel steno-occlusions in peripheral arterial disease patients (PAD). The authors’ findings suggest that the proposed DL model is promising and an effective tool to assist in the detection of arterial steno-occlusions in PAD patients. Key points: Article: Detection of femoropopliteal arterial steno-occlusion at MR angiography: initial

CT-based radiomics combined with hematologic parameters for survival prediction

Insights into Imaging August 2024

This study aimed to investigate the prognostic significance of radiomics in conjunction with hematological parameters relating to the overall survival (OS) of patients diagnosed with esophageal squamous cell carcinoma (ESCC) after undergoing definitive chemoradiotherapy (dCRT). Utilizing radiomics and hematologic parameters, the authors were able to develop a prognostic model that enabled the prediction of OS in ESCC patients. This approach

On Artificial Intelligence: An interview with Hyun Soo Ko

Alex Vogel, Hyun Soo Ko August 2024

This week we spoke to Hyun Soo Ko, a frequent contributor to the ESR’s blog on artificial intelligence and a radiologist in the Department of Cancer Imaging at the Peter MacCallum Cancer Centre in Melbourne, Australia, as well as the Department of Diagnostic and Interventional Radiology at University Hospital Bonn in Bonn, Germany. What is your background/experience with artificial intelligence

AI exhibits high sensitivity and specificity in breast cancer screening program

Erkin Aribal August 2024

Our recent study analyzed a subset of 5,136 mammograms, due to data constraints, from a decade-long screening program (five rounds) involving over 22,000 mammograms from around 9,000 women. Our objective was to evaluate the AI’s detection capabilities and assess various reading scenarios combining AI with human reading. The results highlighted the AI system’s performance with an AUC of 90%. However,

Deep learning for automatic bowel-obstruction identification on abdominal CT

Quentin Vanderbecq July 2024

In our recent study, we developed an automated system for evaluating abdominal computed tomography (CT) scans to assist radiologists in managing their substantial workloads, thus improving patient outcomes. Our machine-learning model focuses on reliably identifying suspected bowel obstruction (BO) on abdominal CT scans. We used an internal dataset of 1,345 annotated CT scans, of which only 670 were re-annotated by

AI-based algorithm offers reliable Cobb angle measurement on routine MRI for degenerative scoliosis patients

European Radiology July 2024

Due to how the severity of degenerative scoliosis is assessed, this retrospective study aimed to develop and evaluate the reliability of a novel automatic method that measured Cobb angles on lumbar MRI in degenerative scoliosis (DS) patients. The authors developed a 3D artificial intelligence algorithm that was trained on 447 lumbar MRI. The study concluded that the AI-based algorithm offered

Shallow and deep learning classifiers in medical image analysis

European Radiology Experimental July 2024

The authors of this review aimed to give educational insight into the most accessible and widely employed classifiers in the field of radiology, distinguish between “shallow” learning algorithms, as well as look into “deep” learning architectures such as convolutional neural networks and vision transformers. This review found that machine learning classifiers offer vital information for the development of clinical decision

Deep learning-based 3D cerebrovascular segmentation workflow on bright and black blood sequences MRA

Insights into Imaging July 2024

Cerebrovascular diseases are seen as a significant threat to human life and health, and the segmentation of brain blood vessels has become a scientific challenge. Therefore, the authors of this study aimed to develop a fully automated deep learning workflow capable of accurate 3D segmentation of cerebral blood vessels using convolutional neural networks (CNNs) and transformer models. The study, conducted

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

Latest posts