Abstract
BACKGROUND: This study aimed to compare sonographer assessment of scan quality using contrast-to-noise ratio (CNR) for obstetric ultrasound images to identify the maternal and obstetric factors that could affect scan quality. METHODS: This retrospective study included 187 consecutive ultrasound scans at 18 weeks of gestation, 205 fetuses, and 565 ultrasound images. Ultrasound scans were performed by a single operator using a General Electric ultrasound machine and a 2-7-MHz RM6C convex probe with the same ultrasound settings. Two independent, blinded sonographers rated the contrast and noise of the abdominal perimeter, four-chamber, and sagittal spine views on a scale of 1-3. The sonographer scores were categorized as poor, fair, good, or excellent, respectively. Patient scans were considered unsatisfactory if they had at least one poor CNR; good if they had more than three CNRs classified as good or excellent; and acceptable otherwise. RESULTS: Interobserver agreement for CNR was moderate to good with 0.53< intraclass correlation coefficient 0.68; P < 0.0001 for the small and large structures of the three views. Half of the unsatisfactory scans were found in obese patients (P = 0.001), and no other factors were significant. CONCLUSION: We demonstrated good interoperator reproducibility of CNR. Maternal obesity was the main factor affecting fetal ultrasound quality.