AI

In addition to a variety of tools to promote radiology workflow efficiencies, the integration of the Gravity AI tools into the PowerServer RIS platform may reduce time-consuming prior authorizations to minutes for completion.

Catch up on the top AI-related news and research in radiology over the past month.

Adjunctive use of the AI-powered software led to an average 38.6 percent increase in the detection of pneumothorax by general radiologists, according to a 2023 study.

Attenuation-based biomarkers on computed tomography (CT) scans demonstrated a 93 percent interclass correlation coefficient (ICC) agreement across three pancreatic segmentation algorithms for predicting diabetes, according to a study involving over 9,700 patients.

Catch up on the top radiology content of the past week.

The Invenia ABUS Premium device reportedly offers consistent high-resolution imaging and up to a 40 percent increase in scan speeds for women with dense breasts.

While recalls initiated by one of two reviewing radiologists after screening mammography were nearly 10 percent higher than recalls initiated by an AI software, the AI-initiated recalls had an 85 percent higher positive predictive value for breast cancer, according to a new study.

What would it be like if we could use AI-powered voice modes as filters for radiology dictations?

Catch up on the top radiology content of the past week.

For immunocompromised individuals, denoised ultra-low-dose CT offers similar detail as normal-dose CT scans for pneumonia detection, according to a recent study.

In addition to a significant reduction of X-ray reading time, researchers found the use of multimodal generative AI assessment led to enhanced detection of pleural lesions and widened mediastinal silhouettes.

Through AI-powered assessment of coronary computed tomography angiography (CCTA) scans, the CaRi-Plaque software may provide improved clarity on the development of coronary plaques and luminal stenosis.

Researchers noted the combination of AI and bpMRI had an average AUC of 91 percent for csPCa detection in contrast to 78 percent for unassisted radiologist interpretation in a recent meta-analysis.

In a study involving over 24,000 women, two-thirds of whom had dense breasts, adjunctive AI led to a significantly higher rate of cancer detection for breast radiologists with no difference in recall rates.

The F.A.S.T. aiCockpit CT Lung Nodule reportedly facilitates enhanced CT workflow and automated reporting for evaluation of lung nodules

A deep learning-enhanced ultra-fast bpMRI protocol offered similar sensitivity for csPCa as mpMRI with an 80 percent reduction in scan time, according to research findings presented at the European Congress of Radiology (ECR) conference.

Catch up on the top radiology content of the past week.

In a multicenter study involving over 1,000 patients, a deep learning software offered comparable sensitivity and specificity for Gleason grade group > 2 tumors in comparison to radiologist interpretation.

Catch up on the most-well viewed radiology content in February 2025.

The updated software reportedly enables a threefold improvement in MRI scan time and enhanced image sharpness.

In patients who had at least four cycles of 177Lu-PSMA-I&T for mCRPC, new research shows that a 10 percent or greater decrease in total kidney volume on CT at six months has a 90 percent AUC for predicting estimated glomerular filtration rates (eGFRs) of 30 percent or greater at one year.

Catch up on the top AI-related news and research in radiology over the past month.

Recent multicenter research showed a 22 percent improvement in the detection of fetal anomalies with the Sonio Suspect AI-powered ultrasound module.

Catch up on the top radiology content of the past week.

Researchers found that the use of seven-minute threefold parallel imaging-accelerated deep learning 3T MRI had 89 percent sensitivity for supraspinatus-infraspinatus tendon tears and 93 percent sensitivity for superior labral tears.

The updated AI-enabled software reportedly facilitates enhanced MRI imaging of the brain, abdomen, and pelvis.

The open-source, deep learning MRI segmentation tool reportedly offers over a 10 percent higher Dice score than similar segmentation models for 40 anatomical structures.

In a study involving assessment of over 1,000 thyroid nodules, researchers found the machine learning model led to substantial increases in sensitivity and specificity for estimating the risk of thyroid malignancy over traditional TI-RADS and guidelines from the American Thyroid Association.

Catch up on the top radiology content of the past week.

Offering an array of AI-powered automation and image optimization, the new Elevate software is geared to maximizing workflow efficiencies for the EPIQ Elite and Affiniti ultrasound systems.