Prospective Evaluation of Artificial Intelligence Triage of Intracranial Hemorrhage on Noncontrast Head CT Examinations.

Background: Retrospective studies evaluating artificial intelligence (AI) algorithms for intracranial hemorrhage (ICH) detection on noncontrast CT (NCCT) have shown promising results but lack prospective validation. Objective: To evaluate the impact on radiologists' real-world aggregate performance for ICH detection and report turnaround times for ICH-positive examinations of a radiology department's implementation of an AI triage and notification system for ICH detection on head NCCT examinations. Methods: This prospective single-center study included adult patients who underwent head NCCT examinations from May 12, 2021 to June 30, 2021 (phase 1) or September 30, 2021 to December 4, 2021 (phase 2). Before phase 1, the radiology department implemented a commercial AI triage system for ICH detection that processed head NCCT examinations and notified radiologists of positive results through a widget with a floating pop-up display. Examinations were interpreted by neuroradiologists or emergency radiologists, who evaluated examinations without and with AI assistance in phase 1 and phase 2, respectively. A panel of radiologists conducted a review process for all examinations with discordance between the radiology report and AI and a subset of remaining examinations, to establish the reference standard. Diagnostic performance and report turnaround times were compared using Pearson chi-square test and Wilcoxon rank-sum test, respectively. Bonferroni correction was used to account for five diagnostic performance metrics (adjusted significance threshold, .01 [α=.05/5]). Results: A total of 9954 examinations from 7371 patients (mean age, 54.8±19.8 years; 3773 female, 3598 male) were included. In phases 1 and 2, 19.8% (735/3716) and 21.9% (1368/6238) of examinations, respectively, were positive for ICH (P=.01). Radiologists without versus with AI showed no significant difference in accuracy (99.5% vs 99.2%), sensitivity (98.6% vs 98.9%), PPV (99.0% vs 99.7%), or NPV (99.7% vs 99.7%) (all P>.01); specificity was higher for radiologists without than with AI (99.8% vs 99.3%, respectively, P=.004). Mean report turnaround time for ICH-positive examinations was 147.1 minutes without AI versus 149.9 minutes with AI (P=.11). Conclusion: An AI triage system for ICH detection did not improve radiologists' diagnostic performance or report turnaround times. Clinical Impact: This large prospective real-world study does not support use of AI assistance for ICH detection.

Inter-Rater Reliability and Correlation of L1 Hounsfield Unit Measurements with DXA Scores.

INTRODUCTION: While prior studies have generally reported rigorous protocols using prespecified CT scanner settings for HU measurements, the present study sought to report on the correlation between DXA and HUs recorded using several CT scanners with varying sequences, simulating measurements performed in "real-world" hospital and Emergency Department (ED) settings. METHODOLOGY: Six raters performed HU measurements of trabecular bone at the L1 vertebral body for forty consecutive patients on Phillips and General Electric (GE) abdominal CT scans obtained between 2017 and 2021. Inter-rater reliability of the HU measurements and their correlations with recorded DXA-based bone assessments were determined. Correlation coefficients were calculated for the HU measurements between scanner vendors as well as for the CT HUs with each DXA measurement. RESULTS: The ICC for L1 HUs read on the Phillips and GE scanners were 0.85 and 0.82, respectively, indicating excellent agreement. The correlation coefficient for the mean HUs on the Phillips and GE scanners was 0.92, also indicating excellent correlation. For both scanner vendors, the HU values most closely correlated with the total femur and femoral neck T-scores. CONCLUSIONS: HU values recorded on a Phillips and GE scanner both demonstrated excellent inter-rater reliability. Correlations were strongest between L1 HU values and total femur DXA T-scores. Readily available abdominal CT image data across multiple hospital settings can be utilized by providers of varying level of imaging interpretation expertise to determine vertebral body Hounsfield units that may help identify osteoporosis risk without additional radiation exposure or cost.

Idiopathic Intracranial Hypertension: Review of Clinical Syndrome, Imaging Findings, and Treatment.

Idiopathic intracranial hypertension (IIH) is a syndrome of unknown cause that is increasing in frequency. Patients who are typically women of childbearing age and obese present with headaches and may also present with visual changes that may become chronic. The purpose of this review is to describe the possible mechanisms for this disease and also to illustrate the ever increasing role of imaging in the diagnosis of this disorder. In addition, the various methods of treatment including medical and surgical will be reviewed. The fact that idiopathic intracranial hypertension has undergone many name changes over the years serves as a reminder that the underlying mechanism is still not well understood. Although there are only several possible mechanisms that can cause increased intracranial pressure, it is still not certain which of these mechanisms is involved. The role of imaging has significantly changed in the evaluation of patients with possible IIH. First, it is involved in ruling out secondary causes of increased intracranial pressure. In addition, there is now ample evidence that the previously held belief that imaging of patients with IIH should be normal is incorrect but rather that there are several subtle findings that radiologists need to look for. These findings include a partially empty sella, flattening of the posterior globe, cupping of optic disks and distension of the optic nerve sheaths. In addition, the role of intracranial venography is playing an ever increasing role due to the finding that a very high percentage of patients have dural venous sinus stenoses. It is becoming clear that there is potentially true morbidity associated with idiopathic intracranial hypertension. The earlier the disease can be diagnosed, the earlier treatment can be started to minimalize permanent visual changes including blindness. Treatment varies from institution to institution due to the fact that multiple specialists with different perspectives treat these patients. Knowledge of subtle imaging features associated with idiopathic intracranial hypertension can help radiologists establish the diagnosis earlier and potentially prevent complications of this disorder. However imaging has not as of yet been shown to be beneficial in managing patients with idiopathic intracranial hypertension.