Superficial Endometriosis at Ultrasound Examination-A Diagnostic Criteria Proposal.

The actual prevalence of superficial endometriosis is not known. However, it is considered the most common subtype of endometriosis. The diagnosis of superficial endometriosis remains difficult. In fact, little is known about the ultrasound features of superficial endometriotic lesions. In this study, we aimed to describe the appearance of superficial endometriosis lesions at ultrasound examination, with laparoscopic and/or histologic correlation. This is a prospective study on a series of 52 women with clinical suspicion of pelvic endometriosis who underwent preoperative transvaginal ultrasound and received a confirmed diagnosis of superficial endometriosis via laparoscopy. Women with ultrasound or laparoscopic findings of deep endometriosis were not included. We observed that superficial endometriotic lesions may appear as a solitary lesions, multiple separate lesions, and cluster lesions. The lesions may exhibit the presence of hypoechogenic associated tissue, hyperechoic foci, and/or velamentous (filmy) adhesions. The lesion may be convex, protruding from the peritoneal surface, or it may appear as a concave defect in the peritoneum. Most lesions exhibited several features. We conclude that transvaginal ultrasound may be useful for diagnosing superficial endometriosis, as these lesions may exhibit different ultrasound features.

AI: Can It Make a Difference to the Predictive Value of Ultrasound Breast Biopsy?

(1) Background: This study aims to compare the ground truth (pathology results) against the BI-RADS classification of images acquired while performing breast ultrasound diagnostic examinations that led to a biopsy and against the result of processing the same images through the AI algorithm KOIOS DS TM (KOIOS). (2) Methods: All results of biopsies performed with ultrasound guidance during 2019 were recovered from the pathology department. Readers selected the image which better represented the BI-RADS classification, confirmed correlation to the biopsied image, and submitted it to the KOIOS AI software. The results of the BI-RADS classification of the diagnostic study performed at our institution were set against the KOIOS classification and both were compared to the pathology reports. (3) Results: 403 cases were included in this study. Pathology rendered 197 malignant and 206 benign reports. Four biopsies on BI-RADS 0 and two images are included. Of fifty BI-RADS 3 cases biopsied, only seven rendered cancers. All but one had a positive or suspicious cytology; all were classified as suspicious by KOIOS. Using KOIOS, 17 B3 biopsies could have been avoided. Of 347 BI-RADS 4, 5, and 6 cases, 190 were malignant (54.7%). Because only KOIOS suspicious and probably malignant categories should be biopsied, 312 biopsies would have resulted in 187 malignant lesions (60%), but 10 cancers would have been missed. (4) Conclusions: KOIOS had a higher ratio of positive biopsies in this selected case study vis-à-vis the BI-RADS 4, 5 and 6 categories. A large number of biopsies in the BI-RADS 3 category could have been avoided.

Advances in Imaging for Assessing Pelvic Endometriosis.

In recent years, due to the development of standardized diagnostic protocols associated with an improvement in the associated technology, the diagnosis of pelvic endometriosis using imaging is becoming a reality. In particular, transvaginal ultrasound and magnetic resonance are today the two imaging techniques that can accurately identify the majority of the phenotypes of endometriosis. This review focuses not only on these most common imaging modalities but also on some additional radiological techniques that were proposed for rectosigmoid colon endometriosis, such as double-contrast barium enema, rectal endoscopic ultrasonography, multidetector computed tomography enema, computed tomography colonography and positron emission tomography-computed tomography with 16α-[18F]fluoro-17ß-estradiol.

Extra-Gynecological Pelvic Pathology: A Challenge in the Differential Diagnosis of the Female Pelvis.

Ultrasound technology with or without color Doppler allows a real-time evaluation of the entire female pelvis including gynecologic and non-gynecological organs, as well as their pathology. As ultrasound is an accurate tool for gynecological diagnosis and is less invasive and less expensive than other techniques, it should be the first imaging modality used in the evaluation of the female pelvis. We present a miscellany of non-gynecological pelvic images observed during the realization of gynecological ultrasound. Transvaginal and transabdominal ultrasound is the first choice among diagnostic techniques for the study of the female pelvis, providing information about gynecological and extra-gynecological organs, allowing for an orientation toward the pathology of a specific organ or system as well as for additional tests to be performed that are necessary for definitive diagnosis.

Conservative Management of Cesarean Scar Pregnancy: A Case Report and Literature Review.

Cesarean scar pregnancy (CSP) is a rare form of ectopic pregnancy located in the lower uterine segment. The current increase in the percentage of cesarean sections is accompanied by significant growth in the incidence of CSP, while advances in ultrasound diagnostic techniques have led to a greater number of CSP diagnoses. A misdiagnosed CSP, or one that is diagnosed too late, is life-threatening to the pregnant patient and predisposes her to complications such as uterine bleeding or rupture, which often require hysterectomy and thus result in the irreversible loss of fertility. We present the case of a 50-year-old woman with a history of undiagnosed CSP after multiple consultations for intermittent bleeding and hemorrhage. She was diagnosed by ultrasound and the diagnosis was confirmed by hysteroscopy. She underwent conservative medical treatment that was successful.

The Reproducibility of Ultrasonographic Findings of Rectosigmoid Endometriosis Among Examiners With Different Level of Expertise.

OBJECTIVE: To analyze the reproducibility of ultrasonographic (US) findings of rectosigmoid endometriosis among examiners with different level of expertise using stored three-dimensional (3D) volumes of the posterior compartment of the pelvis as a part of SANABA (Sardinia-Navarra-Barcelona) collaborative study. MATERIALS AND METHODS: Six examiners in 3 academic Department of Obstetrics and Gynecology, with different levels of experience and blinded to each other, evaluated 60 stored 3D volumes from the posterior compartment of the pelvis and looked for the presence or absence of features of rectosigmoid endometriotic lesions defined as an irregular hypoechoic nodule with or without hypoechoic foci at the level of the muscularis propria of the anterior wall rectum sigma. Multiplanar view and virtual navigation were used. All examiners had to assess the 3D volume of posterior compartment of the pelvis and classify it as present or absent disease. To analyze intra-observer and the inter-observer agreements, each examiner performed the assessment twice with a 2-week interval between the first and second assessments. Reproducibility was assessed by calculating the weighted Kappa index. RESULTS: Intra-observer reproducibility was moderate to very good for all observers (Kappa index ranging from 0.49 to 0.96) associated with a good diagnostic accuracy of each reader. Inter-observer reproducibility was fair to very good (Kappa index range: 0.21-0.87). CONCLUSIONS: The typical US sign of rectosigmoid endometriosis is reasonably recognizable to observers with different level of expertise when assessed in stored 3D volumes.

Artificial intelligence (AI) in the detection of rectosigmoid deep endometriosis.

OBJECTIVES: The aim of this study was to compare the accuracy of seven classical Machine Learning (ML) models trained with ultrasound (US) soft markers to raise suspicion of endometriotic bowel involvement. MATERIALS AND METHODS: Input data to the models was retrieved from a database of a previously published study on bowel endometriosis performed on 333 patients. The following models have been tested: k-nearest neighbors algorithm (k-NN), Naive Bayes, Neural Networks (NNET-neuralnet), Support Vector Machine (SVM), Decision Tree, Random Forest, and Logistic Regression. The data driven strategy has been to split randomly the complete dataset in two different datasets. The training dataset and the test dataset with a 67 % and 33 % of the original cases respectively. All models were trained on the training dataset and the predictions have been evaluated using the test dataset. The best model was chosen based on the accuracy demonstrated on the test dataset. The information used in all the models were: age; presence of US signs of uterine adenomyosis; presence of an endometrioma; adhesions of the ovary to the uterus; presence of "kissing ovaries"; absence of sliding sign. All models have been trained using CARET package in R with ten repeated 10-fold cross-validation. Accuracy, Sensitivity, Specificity, positive (PPV) and negative (NPV) predictive value were calculated using a 50 % threshold. Presence of intestinal involvement was defined in all cases in the test dataset with an estimated probability greater than 0.5. RESULTS: In our previous study from where the inputs were retrieved, 106 women had a final expert US diagnosis of rectosigmoid endometriosis. In term of diagnostic accuracy the best model was the Neural Net (Accuracy, 0.73; sensitivity, 0.72; specificity 0.73; PPV 0.52; and NPV 0.86) but without significant difference with the others. CONCLUSIONS: The accuracy of ultrasound soft markers in raising suspicion of rectosigmoid endometriosis using Artificial Intelligence (AI) models showed similar results to the logistic model.

Ultrasonography and Atypical Sites of Endometriosis.

In the present pictorial we show the ultrasonographic appearances of endometriosis in atypical sites. Scar endometriosis may present as a hypoechoic solid nodule with hyperechoic spots while umbilical endometriosis may appear as solid or partially cystic areas with ill-defined margins. In the case of endometriosis of the rectus muscle, ultrasonography usually demonstrates a heterogeneous hypoechogenic formation with indistinct edges. Inguinal endometriosis is quite variable in its ultrasonographic presentation showing a completely solid mass or a mixed solid and cystic mass. The typical ultrasonographic finding associated with perineal endometriosis is the presence of a solid lesion near to the episiotomy scar. Under ultrasonography, appendiceal endometriosis is characterized by a solid lesion in the wall of the small bowel, usually well defined. Superficial hepatic endometriosis is characterized by a small hypoechoic lesion interrupting the hepatic capsula, usually hyperechoic. Ultrasound endometriosis of the pancreas is characterized by a small hypoechoic lesion while endometriosis of the kidney is characterized by a hyperechoic small nodule. Diaphragmatic endometriosis showed typically small hypoechoic lesions. Only peripheral nerves can be investigated using ultrasound, with a typical solid appearance. In conclusion, ultrasonography seems to have a fundamental role in the majority of endometriosis cases in "atypical" sites, in all the cases where "typical" clinical findings are present.

Ultrasonographically diagnosed dermoid cysts do not influence ovarian stimulation response in an in vitro fertilization cycle.

To determine if patients with a DC respond similarly to ovarian stimulation when compared to patients without a DC. Infertility patients with a DC that underwent IVF between January 2009 and December 2016 were included. A cystic mass with mixed echogenicity, internal echoes similar to thick bands, fatty-fluid level, or an echogenic tubercle with acoustic shadow (Rokitansky nodule) within two years of the cycle characterized the diagnosis. The z-score compared the standard deviations (SDs) in patients with/without a DC and were compared to a nomogram (expected oocytes minus oocytes obtained divided by the SD), adjusted for age and number of oocytes retrieved, built utilizing cycles from noninfertile female patients. Thirty-nine patients with DC and 7839 patients without DC were identified. The mean number of oocytes (8.6 ± 5.8 vs. 8.5 ± 7.7, p = .43) and MIIs (6.7 ± 4.7 vs. 7.0 ± 6.7, p = .74) retrieved were similar. When cycles with and without a DC were compared to the nomogram (z-score of 0), cycles with a DC presented a z-score for ovarian response of 0.1921 SDs from the mean, and patients without DC presented a z-score of -0.2065 SDs from the mean (similar and less than -1.0). After building a population 'normal' response as a template, patients with and without a DC responded similar to COS.

Endometrial cancer off-line staging using two-dimensional transvaginal ultrasound and three-dimensional volume contrast imaging: Intermethod agreement, interrater reliability and diagnostic accuracy.

OBJECTIVES: The aim is to estimate agreement between two-dimensional transvaginal ultrasound (2D-TVS) and three-dimensional volume contrast imaging (3D-VCI) in diagnosing deep myometrial invasion (MI) and cervical stromal involvement (CSI) of endometrial cancer and to compare the two methods regarding inter-rater reliability and diagnostic accuracy. METHODS: Fifteen ultrasound experts assessed off-line de-identified 3D-VCI volumes and 2D-TVU video clips from 58 patients with biopsy-confirmed endometrial cancer regarding the presence of deep (≥50%) MI and CSI. Video clips and 3D volumes were assessed independently. Interrater reliability was measured using kappa statistics. Histological diagnosis after hysterectomy served as gold standard. Accuracy measurements were correlated to rater experience using Spearman's rank correlation coefficient (ρ). RESULTS: Agreement between 2D-TVU and 3D-VCI for diagnosing MI was median 76% (range 64-93%) and for CSI median 88% (range 79-97%). Interrater reliability was better for 2D-TVU than for 3D-VCI (Fleiss' kappa 0.41 vs. 0.31 for MI and 0.55 vs. 0.45 for CSI). Median accuracy for diagnosing deep MI was 76% (range 59-84%) with 2D-TVU and 69% (range 52-83%) for 3D-VCI; the corresponding figures for CSI were 88% (range 81-93%) and 86% (range 72-95%). Accuracy was significantly correlated to how many cases the raters assessed annually. CONCLUSIONS: Off-line assessment of MI and CSI in women with endometrial cancer using 3D-VCI has lower interrater reliability and lower accuracy than 2D-TVU video clip assessment. Since accuracy was correlated to the number of cases assessed annually it is advised to centralize these examinations to high-volume centres.

Long-term Results for Expectant Management of Ultrasonographically Diagnosed Benign Ovarian Teratomas.

OBJECTIVE: To assess the natural history of ultrasonographically diagnosed benign ovarian teratomas in asymptomatic women. METHODS: We conducted a retrospective observational cohort study of 408 women (mean age 36.6 years, range 14-81 years) diagnosed as having an ovarian teratoma by transvaginal ultrasonography (except eight who only had a transabdominal study done) between January 2003 and December 2013 at a single tertiary care institution. Six hundred thirteen women were diagnosed with ovarian teratoma of whom 205 were promptly treated surgically, leaving 408 patients followed conservatively with follow-up scans at 3 and 6 months from diagnosis and then yearly. The ultrasonographic diagnosis of a benign ovarian teratoma required at least one of the following features: a cystic mass with mixed echogenicity, thick band-like echoes, a fat-fluid level, or echogenic tubercle with posterior shadowing. Clinical, ultrasonographic, and histologic data (in case of surgery) were retrieved for analysis. RESULTS: During follow-up, 130 of 408 (31.8%) women underwent surgery. The main reason for surgery was the physician's recommendation according to our protocol (n=115). One patient had adnexal torsion. Most surgeries (112/130 [86.2%]) were performed within the first 5 years after diagnosis. The remainder (278/408) is still being followed (median time 45.6 months, range 6-147 months). The vast majority of these lesions had no change and women remain asymptomatic. Histologic diagnosis of tumors removed surgically revealed a benign ovarian teratoma in 103 of 130 (79.2%) of the women. There were two borderline tumors, four endometriomas, three fibromas, seven serous cysts, two mucinous cysts, two stroma ovarii, seven other benign, and no case of malignant tumor. CONCLUSION: Our results demonstrate that expectant management might be a reasonable option for managing asymptomatic women who receive a ultrasonographic diagnosis of a benign ovarian teratoma. The risk of undergoing surgery for this lesion decreases significantly after 5 years to follow-up. With careful observation, the risk of missing a diagnosis of malignancy is low.

Reproducibility of the International Endometrial Analysis Group Color Score for Assigning the Amount of Flow Within the Endometrium Using Stored 3-Dimensional Volumes.

OBJECTIVES: To estimate intraobserver and interobserver reproducibility for assigning an International Endometrial Tumor Analysis (IETA) group color score for endometrial vascularization on color Doppler imaging. METHODS: Sixty-eight endometrial 3-dimensional volumes from endometrial color Doppler assessments of women with different endometrial disorders were evaluated by 8 different examiners (4 skilled examiners and 4 obstetric and gynecologic trainees). One skilled examiner who did not participate in the assessments selected the 68 volumes from a database to select a balanced number of each IETA score. Each examiner evaluated the 68 endometrial volumes to assign the IETA color score (1, absence of vascularization; 2, low vascularization; 3, moderate vascularization; or 4, abundant vascularization) using tomographic ultrasound imaging. The analysis was repeated 4 weeks later, and interobserver and intraobserver reproducibility was analyzed by calculating the weighted κ index. The second of the measurements made by each observer was used to estimate interobserver reproducibility. RESULTS: The intraobserver reproducibility was very good for all examiners, with a weighted κ index ranging from 0.84 to 0.91. The interobserver reproducibility was good or very good for all estimated comparisons, with a weighted κ index ranging from 0.77 to 0.96, regardless of experience level. CONCLUSIONS: The reproducibility of assigning the IETA color score for assessing endometrial vascularization using 3-dimensional volumes is good or very good regardless of the experience of the examiner.

Malignancy risk of sonographically benign appearing purely solid adnexal masses in asymptomatic postmenopausal women.

OBJECTIVE: To assess the natural history of benign appearing purely solid ovarian lesions in asymptomatic postmenopausal women. METHODS: Retrospective observational cohort study comprising 99 women (mean age, 58.2 years, ranging from 50 to 77 years) diagnosed as having a purely solid ovarian lesion at transvaginal ultrasound between April 2001 and October 2015. Inclusion criteria were as follows: asymptomatic postmenopausal women with a well-defined purely solid ovarian lesion with International Ovarian Tumor Analysis color score 1 or 2, without ascites and/or signs of carcinomatosis. Clinical, sonographic, biochemical (CA-125), and histologic data (in case of surgery) were retrieved for analysis. Patients who were managed conservatively were assessed by transvaginal sonography every 6 months for a minimum of a year. In case of bilateral lesions we used the largest one for analysis. RESULTS: Five women (5.1%) had bilateral lesions. Mean size of the lesion was 2.9 cm (range, 1.0-7.8 cm). Most lesions were homogeneous (96.0%). Acoustic shadowing was present in 59.6% of cases. International Ovarian Tumor Analysis color score was 1 in 77.8% and 2 in 22.2% of the cases, respectively. Median CA-125 was 10.8 IU/mL (range, 3.0-403.0 IU/mL). Forty-two women underwent surgery after diagnosis (histologic diagnoses were as follows: fibroma (n = 26), fibrothecoma (n = 5), dermoid (n = 3), Brenner tumor (n = 3), endometrioma (n = 2), thecoma (n = 1), primary invasive cancer (n = 2). One case of invasive cancer CA-125 was 403.0 IU/mL and in the other case CA-125 was 6.0 IU/mL. They both were stage 1. Fifty-seven women were managed with serial follow-up. With a median follow-up time of 36 months (range, 12-142 months) all these lesions had no change and women remain asymptomatic. Considering all 99 cases the risk of malignancy is 2% (95% CI, 0.1-7.5). CONCLUSIONS: The risk of malignancy of benign appearing purely solid adnexal masses in asymptomatic postmenopausal women is low. Conservative management of these lesions might be an option.

A Simple Method for Excluding Uterine Canalization Defects Using Two-Dimensional Ultrasound in Infertile Women.

Objective: To compare the uterine transverse diameter (UTD) in women with normal uteri and women with uterine canalization defects as well as to assess its performance for ruling out such defects. Study Design: Retrospective analysis of prospectively collected data in a series of selected women with primary or secondary infertility. Measurement of UTD and 3D volume acquisition for subsequent off-line analysis was performed in order to identify possible canalization defects. UTD of the normal uterus, measured by 2D ultrasound, was compared to that of arcuate, subseptate, and septate uteri. ROC curve was plotted to determine the best UTD cutoff for differentiating normal from abnormal uteri. Results: A total of 421 women were ultimately evaluated. UTD was significantly larger in women with arcuate (53.3 mm, SD 6.3, p<0.05), subseptate (55.0 mm, SD 6.7, p<0.05), and septate (56.0 mm, SD 4.8, p<0.05) uterus as compared with the normal uterus (45.9 mm, SD 7.1). ROC curve showed that the best UTD cutoff for ruling out the presence of a uterine canalization defect was 45 mm (AUC 0.809, 95% CI 0.768­0.849). Conclusion: Measurement of UTD may be a simple and practical method for ruling out a uterine canalization defect in infertile women.

Diagnosis and laparoscopic management of a rudimentary horn pregnancy: Role of three-dimensional ultrasound.

Rudimentary horn pregnancy occurs in 1 in 76,000-150,000 pregnancies and causes uterine rupture in about 80% of cases. The use of three-dimensional transvaginal ultrasound seems to be useful for its early detection. We present a case of an 8-week pregnancy in a rudimentary horn, managed by laparoscopic excision. © 2016 Wiley Periodicals, Inc. J Clin Ultrasound 45:112-115, 2017.

Training Performance in Diagnosis of Congenital Uterine Anomalies With 3-Dimensional Sonography: A Learning Curve Cumulative Summation Analysis.

OBJECTIVES: We evaluated learning curve cumulative summation (CUSUM) of 3-dimensional (3D) sonography for diagnosis of congenital uterine anomalies and the deviations of the level of trainees' performance at the control-stage CUSUM. METHODS: First-year (R1), second-year (R2), and third-year (R3) residents in obstetrics and gynecology received a training program to learn how to analyze 3D sonographic volumes and to classify congenital uterine anomalies. Each trainee worked on 155 3D sonographic volumes from preselected patients. Their results were evaluated by learning curve CUSUM and standard CUSUM. The time for each volume analysis was calculated for the expert examiner and the trainees. RESULTS: Trainees R1, R2, and R3 reached competence at the 85th, 58th, and 40th evaluations, respectively, with success rates of 80%, 81%, and 85%, and kept the process under control with error levels of less than 4.5% until the end of the test. The trainees significantly reduced the average time of the evaluation per volume (P < .001). CONCLUSIONS: Learning curve CUSUM provided quantitative indicators of the learning evolution of 3D sonography for diagnosis of congenital uterine anomalies by obstetrics and gynecology residents. The training received by the residents was adequate for diagnosis of congenital uterine anomalies using 3D sonography.

Three-dimensional sonography for diagnosis of rectovaginal septum endometriosis: interobserver agreement.

OBJECTIVES: The aim of this study was to evaluate the interobserver agreement for diagnosis of deep endometriosis of the rectovaginal septum using introital 3-dimensional (3D) sonography. METHODS: Two experienced observers (observers A and B) performed a retrospective review of stored 3D sonographic volumes from a sample of 84 consecutive patients with a clinical suspicion of endometriosis. Each observer, independently and blinded to each other, evaluated the presence or absence of involvement of the rectovaginal septum. When no lesion was seen, the observers were asked to judge whether the acquisition of the volume was suboptimal for interpretation or whether no lesion on the rectovaginal septum was detectable. One inadequate acquisition case was discarded; a total of 83 cases were evaluated. To calculate the performance of introital 3D sonography, 7 discordant cases were reviewed by a third observer. Interobserver agreement was assessed by calculating the κ index, and the sensitivity, specificity, positive predictive value, and negative predictive value for the 3 observers were also determined. RESULTS: Interobserver agreement was 0.816 (95% confidence interval, 0.69-0.93), representing very good agreement. Sensitivity was 74.1%; specificity, 85.7%; positive predictive value, 71.4%; and negative predictive value, 87.3%. CONCLUSIONS: Our results show that introital 3D sonography for diagnosis of deep endometriosis of the rectovaginal septum is reproducible, with very good interobserver agreement.

Normal and abnormal images of intrauterine devices: Role of three-dimensional sonography.

The purpose of this pictorial essay is to describe the diagnostic value of two-dimensional ultrasound (2DUS) and the additional information that three-dimensional ultrasound (3DUS) provides in the assessment of location, type and complications of IUDs.

Intra- and interobserver variability of 2D and 3D transvaginal sonography in the diagnosis of benign versus malignant adnexal masses.

PURPOSE: The purpose of this study was to investigate the intraobserver and interobserver agreements in the diagnosis of malignant versus benign adnexal masses using two-dimensional ultrasonography (2D US) and three-dimensional ultrasonography (3D US). METHODS: Two experienced sonographers performed a retrospective review of digitally stored 2D images and 3D data from a sample of 41 consecutive patients with a diagnosis of adnexal mass. Each observer independently, and blinded to each other, evaluated the 2D static images of each adnexal mass and then the 3D volumes 1 week later. The observers were required to classify the adnexal lesion as benign or malignant according to pattern recognition analysis. Intraobserver and interobserver agreement were assessed by calculating the kappa index (κ). RESULTS: Intraobserver agreement between 2D US and 3D US for the observer A was 1.00 and for the observer B was 0.69. Interobserver agreement was 0.69 for 2D US and 1.00 for 3D US (p > 0.05) CONCLUSIONS: Although 3D US in the diagnosis of adnexal masses appeared more reproducible than 2D US, the difference was not statistically significant.

The diagnosis of ovarian cancer: is color Doppler imaging reproducible and accurate in examiners with different degrees of experience?

OBJECTIVE: To evaluate the reproducibility and accuracy of color Doppler flow location in indeterminate masses after a gray-scale sonography in the diagnosis of ovarian malignancy. METHODS: Digitally stored color Doppler sonographic images from a random sample of 130 women with an indeterminate adnexal mass submitted to surgery were evaluated by six different examiners with different degrees of experience. A mass was graded malignant if flow was shown within the excrescences or solid areas. Intraobserver agreement and interobserver agreement according to the level of experience were assessed by calculating the kappa index. RESULTS: Intraobserver agreement was good for all examiners with different degrees of experience (kappa 0.72-0.89). Interobserver agreement was good to moderate for all operators (kappa 0.48-0.71) irrespective of degree of experience. The accuracy was comparable among different operators. CONCLUSIONS: Our results indicate that color Doppler imaging for detection of adnexal malignancy seems to be a reproducible method even in moderately experienced examiners.