AI as a New Frontier in Contrast Media Research

Haubold, Johannes; Hosch, René; Jost, Gregor; Kreis, Felix; Forsting, Michael; Pietsch, Hubertus; Nensa, Felix

doi:10.1097/rli.0000000000001028

Cited by 3 publications

(2 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This is particularly important in the topic area of reducing the amount of contrast agent through AI-based contrast enhancement, as separate algorithms are often trained for each contrast phase. [55][56][57]…”

Section: Discussionmentioning

confidence: 99%

Addressing the Contrast Media Recognition Challenge

Baldini,

Hosch,

Schmidt

et al. 2024

Invest Radiol

Self Cite

View full text Add to dashboard Cite

Objectives Accurately acquiring and assigning different contrast-enhanced phases in computed tomography (CT) is relevant for clinicians and for artificial intelligence orchestration to select the most appropriate series for analysis. However, this information is commonly extracted from the CT metadata, which is often wrong. This study aimed at developing an automatic pipeline for classifying intravenous (IV) contrast phases and additionally for identifying contrast media in the gastrointestinal tract (GIT). Materials and Methods This retrospective study used 1200 CT scans collected at the investigating institution between January 4, 2016 and September 12, 2022, and 240 CT scans from multiple centers from The Cancer Imaging Archive for external validation. The open-source segmentation algorithm TotalSegmentator was used to identify regions of interest (pulmonary artery, aorta, stomach, portal/splenic vein, liver, portal vein/hepatic veins, inferior vena cava, duodenum, small bowel, colon, left/right kidney, urinary bladder), and machine learning classifiers were trained with 5-fold cross-validation to classify IV contrast phases (noncontrast, pulmonary arterial, arterial, venous, and urographic) and GIT contrast enhancement. The performance of the ensembles was evaluated using the receiver operating characteristic area under the curve (AUC) and 95% confidence intervals (CIs). Results For the IV phase classification task, the following AUC scores were obtained for the internal test set: 99.59% [95% CI, 99.58–99.63] for the noncontrast phase, 99.50% [95% CI, 99.49–99.52] for the pulmonary-arterial phase, 99.13% [95% CI, 99.10–99.15] for the arterial phase, 99.8% [95% CI, 99.79–99.81] for the venous phase, and 99.7% [95% CI, 99.68–99.7] for the urographic phase. For the external dataset, a mean AUC of 97.33% [95% CI, 97.27–97.35] and 97.38% [95% CI, 97.34–97.41] was achieved for all contrast phases for the first and second annotators, respectively. Contrast media in the GIT could be identified with an AUC of 99.90% [95% CI, 99.89–99.9] in the internal dataset, whereas in the external dataset, an AUC of 99.73% [95% CI, 99.71–99.73] and 99.31% [95% CI, 99.27–99.33] was achieved with the first and second annotator, respectively. Conclusions The integration of open-source segmentation networks and classifiers effectively classified contrast phases and identified GIT contrast enhancement using anatomical landmarks.

show abstract

Section: Discussionmentioning

confidence: 99%

Addressing the Contrast Media Recognition Challenge

Baldini,

Hosch,

Schmidt

et al. 2024

Invest Radiol

Self Cite

View full text Add to dashboard Cite

show abstract

“…Last in order but of great importance today (and potential impact upon the use of MR contrast media) is the use of artificial intelligence (AI) techniques in magnetic resonance and specifically their impact on contrast utilization—whether it be reduction of dose, 30–33 allowing new applications, 34 or the elimination of use in specific areas 35–37 . For the latter, it is still too early to come to any conclusions, and likely the answer is not straightforward.…”

mentioning

confidence: 99%

A New Era in Magnetic Resonance Contrast Media

Runge,

Heverhagen

2023

Invest Radiol

View full text Add to dashboard Cite

Next-generation gadolinium-based contrast agents (GBCAs), including both high relaxivity agents and targeted agents, and manganese-based agents with a high probably of commercial success are discussed in some depth. It is highly likely that gadopiclenol and gadoquatrane, both next-generation high relaxivity gadolinium-based compounds, will come in time to replace the current macrocyclic gadolinium chelates, despite the wide acceptance, very high safety profile, and high stability of the latter group. Current research has also made possible the development of 2 new targeted gadolinium chelates, which look very promising, with the potential to improve cancer detection (for both MT218 and ProCA32.collagen) as well as diseases of collagen (for the latter agent). Further work with manganese-based compounds, a topic left fallow for more than 20 years, has also now produced 2 agents with high potential for clinical use, one (manganese chloride tetrahydrate, administered orally) developed primarily for imaging of the liver and the other (Mn-PyC3A, administered intravenously) as a gadolinium-free replacement for the GBCAs. New detail has recently emerged regarding specific circumscribed subregions of the brain with specialized cytoarchitecture and functions in which high gadolinium concentrations are seen following injection of the linear agent gadodiamide. These findings pave the way for tailored functional neurological testing, specifically in patients at potential risk due to the continued wide use in many countries across the world of the linear GBCAs. The impact of artificial intelligence is also critically discussed, with its most likely applications being dose reduction and new clinical indications.

show abstract