CT

Photon-counting computed tomography demonstrated over 20 percent accuracy at the patient level and nearly 14 percent higher accuracy level at the vessel level in comparison to energy-integrating detector CT for detecting > 70 percent stenosis, according to a newly published study.

The launch of the AVI imaging platform reportedly facilitates direct integration into PACS and RS systems, and automated access to adjunctive CT-based AI applications for a variety of conditions including pulmonary embolism, intracranial hemorrhage and large vessel occlusion

Offering a “safety net” in emergency and inpatient settings, the Chest CT software reportedly provides enhanced triage capabilities and adjunctive detection of 167 radiological features.

New research shows that incorporating the CT vascular sign into assessments of pulmonary nodules led to significant increases in specificity and accuracy as well as reduced false positive rates for differentiating between malignant and benign solid pulmonary nodules.

The Harrison.ai Open Platform reportedly emphasizes open architecture, customer ROI and elimination of costly AI platform fees.

The AI-powered cvi42 | Plaque software enables adjunctive on-site interpretation of coronary CT angiography (CCTA) scans for plaque quantification.

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

The newly launched PlaqueIQ™ image analysis modality is reportedly the first CT-based software indicated for the quantification and classification of plaque morphology in the carotid arteries.

For patients with esophageal cancer, preoperative MRI demonstrated an average AUC above 94 percent for detecting tracheal invasion in contrast to an AUC range between 52.9 and 70.6 percent for CT, according to a recently published study.

New research on photon-counting computed tomography in 2025 reveals increasing capabilities of the technology to enhance CT visualization and characterization with significant reductions in radiation dosing in cardiovascular imaging, abdominal imaging and other diagnostic applications.

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

In a review of nearly 1,300 emergency referrals for computed tomography (CT) scans, approximately 28 percent were deemed to have adequate requisition quality according to a RI-RADS analysis.

Facilitating timely triage, the qER-CTA software enables AI assessment of the internal carotid artery and M1 segment of the middle cerebral artery for possible large vessel occlusions (LVOs).

Through the use of a multimodal system, Viz Assist provides generative AI summaries of patient histories that may ehance care coordination and facilitate more timely interventions.

The AI-enabled LungQ® 4 software reportedly offers enhanced segmentation of peripheral airways and estimates of chronic perfusion deficits based on analysis of chest CT scans.

Facilitating timely detection and follow-up management of patients with aortic aneurysms, the PRAEVAorta2 software reportedly enables automated measurement of aortic diameters on computed tomography (CT) scans.

A deep learning AI platform, which incorporated radiomic features including CT attenuation metrics, demonstrated a 93.6 percent AUC for detecting invasive adenocarcinoma on chest CT.

In a study involving over 500 patients, researchers noted that the CT-based AI software led to a 24 percent increase of evaluations for aortic abdominal aneurysms within six months of detection.

The AI-powered Bunkerhill MAC software reportedly facilitates chest CT detection and quantification of mitral annular calcification, which has been linked to higher cardiovascular risk and related mortality.

Based on survey findings and subsequent voting from radiologists at nine facilities, the Society of Abdominal Radiology has issued new consensus recommendations on the use of adult abdominal photon-counting CT.

Researchers found comparable 10-year overall survival rates for people who had computed tomography (CT) surveillance (94.7 percent) and those who had surgery for ground-glass nodules (97.6 percent), according to a new prospective study involving over 680 participants.

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

The Planmed XFI full-body weightbearing cone-beam CT scanner reportedly enables upright and supine patient positioning.

Providing adjunctive AI detection for multiple conditions on CT through a single workflow, the multi-triage platform would be available through Aidoc’s aiOS platform.

While pooled data from 38 studies showed markedly higher sensitivity rates for the use of adjunctive AI for ischemic stroke detection on non-contrast CT exams, researchers noted a high degree of bias and lack of external validation in many of the studies.