Arteriovenous malformations (AVM) of the uterus can cause life-threatening hemorrhage. Unexplained, heavy vaginal bleeding in a reproductive age woman should raise suspicion for an AVM. Here a 37-year-old woman had increasingly severe vaginal bleeding for 15 days. Serum β-hCG was elevated. Two-dimensional transvaginal ultrasound suggested retained products of conception. Before dilation and curettage (D&C), color Doppler and three-dimensional (3D) power Doppler demonstrated findings indicative of uterine AVM. A bilateral uterine artery embolization was performed without complications. Three months after uterine artery embolization, 3D power Doppler ultrasonography found complete resolution of the AVM. This case illustrates the importance of assessing both gray-scale and 3D power Doppler, and the ability of postprocedure Doppler to assess resolution.

During the last two decades three-dimensional (3D) ultrasound has become increasingly popular around the world. There have been a great number of research papers studying its feasibility as well as reliability and accuracy. Compared to two-dimensional (2D), 3D ultrasound produces better visualization with minimal probe manipulation. The sonographer performing 2D ultrasound has to mentally assemble crosssectional images of a structure and/or object into a 3D image. In cases with uncertain anatomic orientation less experienced examiners may experience difficulties in envisioning three dimensions of the visualized structure, lesion or organ. Patients and parents are often faced with similar problems. A number of studies have shown that novice operators can efficiently store and interpret the volume data by 3D ultrasound, which may have many positive implications in the practice.

Purpose: This study evaluates the accuracy and reliability of off-line volume measurement using virtual organ computer-aided analysis (VOCAL TM ) and two-dimensional (2D) ultrasound by novices.Methods: Thirty medical students, 15 sonographers and three experts were recruited to measure a liver, a gestational sac and a breast cancer lesion using 2D ultrasound and threedimensional (3D) VOCAL TM method. The VOCAL TM measurements produced by the experts were used as the gold standard for comparison. Accuracy and reliability were assessed by analysis of variance followed by a mean separation procedure. Results:The t-test revealed with statistical significance that in measuring the liver, students outperformed sonographers regardless of 2D or VOCAL TM methods; also, VOCAL TM generated more accurate measurements than 2D ultrasound in both groups. In regards to the gestational sac, students continued to perform well with both 2D and VOCAL TM while the sonographers were accurate by using VOCAL TM only. Both groups of novices were not accurate with measuring the breast lesion. VOCAL TM increased the novices' interobserver reliability but not intraobserver reliability. Conclusion:Students and sonographers can measure regularly shaped structures, such as the gestational sac accurately and reliably using VOCAL TM . VOCAL TM increased interobserver reliability in all groups of operators.