Assess Interobserver Variability Research Articles

The Diagnosis of Endometrial Carcinomas With Clear Cells by Gynecologic Pathologists: An Assessment of Interobserver Variability and Associated Morphologic Features

The Diagnosis of Endometrial Carcinomas With Clear Cells by Gynecologic Pathologists: An Assessment of Interobserver Variability and Associated Morphologic Features

An In Vitro Comparison of the Accuracy of Measurements Obtained from High- and Low-resolution Cone-beam Computed Tomography Scans

IntroductionThis study aimed to investigate in an ex vivo model the reduction in patient radiation dose while maintaining accurate linear measurements by comparing cone-beam computed tomography (CBCT) scans taken at 360° versus 180° rotation, with porcine jaw specimens as a reference standard. MethodsCBCT scans of 12 sectioned porcine hemimandibles at 360° and 180° rotations were taken with standardized clinical exposure parameters. To assess interobserver variability, 6 assessors who were blinded to the degree of rotation took linear measurements of anatomic structures on each scan. The measurements were repeated after 2 weeks to assess intraobserver variability. Accuracy of measurement was judged against the corresponding measurements taken from the porcine jaw specimens. ResultsIntraclass correlations signaled good-to-excellent intraobserver and interobserver agreement (0.65–0.98 and 0.79–0.98), respectively. Mixed regression analysis found no significant difference between the measurements from 180° or 360° rotations and no difference between the 2 rotations and porcine jaw specimens. ConclusionsA CBCT image sufficient to make accurate clinical measurements with a reduced radiation exposure may be obtained by using 180° rotation of the CBCT tube head.

Interobserver Variability in the CT Assessment of Honeycombing in the Lungs

To quantify observer agreement and analyze causes of disagreement in identifying honeycombing at chest computed tomography (CT). The institutional review board approved this multiinstitutional HIPAA-compliant retrospective study, and informed patient consent was not required. Five core study members scored 80 CT images with a five-point scale (5 = definitely yes to 1 = definitely no) to establish a reference standard for the identification of honeycombing. Forty-three observers from various subspecialties and geographic regions scored the CT images by using the same scoring system. Weighted κ values of honeycombing scores compared with the reference standard were analyzed to investigate intergroup differences. Images were divided into four groups to allow analysis of imaging features of cases in which there was disagreement: agreement on the presence of honeycombing, agreement on the absence of honeycombing, disagreement on the presence of honeycombing, and other (none of the preceding three groups applied). Agreement of scores of honeycombing presence by 43 observers with the reference standard was moderate (Cohen weighted κ values: 0.40-0.58). There were no significant differences in κ values among groups defined by either subspecialty or geographic region (Tukey-Kramer test, P = .38 to >.99). In 29% of cases, there was disagreement on identification of honeycombing. These cases included honeycombing mixed with traction bronchiectasis, large cysts, and superimposed pulmonary emphysema. Identification of honeycombing at CT is subjective, and disagreement is largely caused by conditions that mimic honeycombing.

Visibility and measurement of Cesarean section scars in pregnancy: a reproducibility study

To evaluate the visibility of cesarean section (CS) scars by transvaginal sonography (TVS) in pregnant women, to apply a standardized approach for measuring CS scars and to test its reproducibility throughout the course of pregnancy. In this observational cohort study, 320 consecutive pregnant women with a previous cesarean delivery were examined to assess scar visibility by two independent examiners. TVS was carried out at 11-13, 19-21 and 34-36 weeks' gestation. A scar was defined as visible when an area of hypoechogenicity representing myometrial discontinuity at the anterior wall of the lower uterine segment was identified. In a subset of patients (n = 111), visible scars were measured by two independent examiners in three dimensions: scar width, depth and length as well as the residual myometrial thickness (RMT). Descriptive analysis was used to assess scar visibility, and the intraclass correlation coefficient (ICC) was calculated to show the strength of absolute agreement between two examiners for scar measurements. For RMT, a cut-off of 2.4 mm was used and measurement agreement was assessed using Cohen's kappa coefficient. The scar was visible in 284/320 cases (88.8%). Visible scars were significantly associated with anteverted uteri (P < 0.0001). Both examiners had 100% agreement on scar visibility at 12 and 20 weeks' gestation, while agreement was 96% at 34 weeks. The intra- and interobserver agreements for scar measurements were generally good (ICC 0.86 and 0.89, respectively). The kappa coefficient for the RMT was 0.27 in the first trimester, compared with 0.51 and 0.72 in the second and third trimesters, respectively. CS scars remain visible in the majority of women throughout pregnancy. They can be reproducibly measured in three dimensions when assessed by TVS in all trimesters of pregnancy. The agreement between two observers for CS scar measurement can be considered good for the first trimester, compared with relatively moderate agreement for the second and third trimesters.

Pretransplant prediction of microvascular invasion in patients with hepatocellular carcinoma: Added value of diffusion-weighted magnetic resonance imaging

The goals of the meticulous selection of patients withhepatocellular carcinoma (HCC) for liver transplanta-tion (LT) are to minimize the incidence of tumor recur-rence after transplantation and to improve patients’survival. Tumor invasion into vascular structures is aknown risk factor for tumor recurrence after LT. Mac-rovascular invasion, which is defined as the presenceof a tumor thrombus in a hepatic or portal vein, canbe detected on preoperative cross-sectional imaging. Itis considered an absolute contraindication to LT. Con-versely, microvascular invasion (MVI) by definitioncan be definitively diagnosed only by histopathology.Several studies have shown that MVI is associatedwith HCC recurrence within 1 year of transplanta-tion.

Guidelines for the investigation and management of mantle cell lymphoma.

The guideline group was selected to be representative of UK-based medical experts and patients representatives. Ovid MEDLINE, EMBASE and NCBI Pubmed were searched systematically for publications in English from 1980 to 2011 using the MeSH subheading 'lymphoma, mantle cell' and 'lymphoma, mantle cell' as a keyword, as well as all subheadings. In addition, all references to mantle cell lymphoma in the World Health Organization Classification of Tumours of Haematopoietic and Lymphoid Tissue (Swerdlow et al, 2008) and the British Committee for Standards in Haematology (BCSH) Guideline: Best Practice in Lymphoma Diagnosis and Reporting (Parker et al, 2010) have been included. The writing group produced the draft guideline, which was subsequently revised by consensus by members of the Haemato-oncology Task Force of the BCSH. The guideline was then reviewed by a sounding board of approximately 50 UK haematologists, the BCSH and the British Society for Haematology Committee and comments incorporated where appropriate. The 'GRADE' system was used to quote levels and grades of evidence, details of which can be found in Appendix I. The objective of this guideline is to provide healthcare professionals with clear guidance on the investigation and management of patients with mantle cell lymphoma. The guidance may not be appropriate to patients with other lymphoma sub-types and in all cases individual patient circumstances may dictate an alternative approach.

Comparaison de différentes méthodes de segmentation en TEP/TDM pour le ciblage en radiothérapie des carcinomes épidermoïdes des VADS

IntroductionTo assess interobserver variability for biological target volume (BTV) delineation and to compare the reproductibility of different semiautomatic segmentation methods in pretreatment 18-fluorodeoxyglucose positron emission tomography (PET/CT) of head and neck squamous cell carcinoma (HNSCC). Patients and methodsPatients with histologically proved HNSCC referred to the nuclear medicine service in Brest for pretreatment PET/CT were prospectively included from February 2009 to June 2010. Three nuclear medicine physicians (two specialized in oncology) delineated manually and independently BTV on each primary tumor. Four semiautomatic segmentation methods have been studied; three using a fixed threshold and one applying an adaptive threshold based on the signal-to-background ratio (Daisne). The variability between κ observers and/or methods has been assessed. The concordance between the various BTV intersections and unions has been also assessed. ResultsThirty patients (29M; 1F) were included. The primary site location was oropharynx in six patients, oral cavity in 10 patients, hypopharynx in five patients and larynx in nine patients. A statistically significant global interobserver variability (P=0.01) was showed, but without statistically difference between the two experienced oncologists (P=0.15). The maximal concordance of the two experienced observers with the semiautomatic methods was found for the Daisne method (CI=61.5%; κ=0.68), expressing a good agreement according to the Landis and Koch criteria, better than with the segmentation method using a fixed threshold with 40% of maximal signal intensity (CI=52.1%; κ=0.53). ConclusionOur results suggest the feasibility of achieving HNSCC BTV delineation by PET/CT using semiautomatic methods, in particular those which apply an adaptative threshold but under the supervision of an experienced operator.

The effect of fetal congenital heart disease on in utero urine production rate

To evaluate the influence of congenital heart disease (CHD) or congestive heart failure (CHF) on fetal urine production rate (UPR) and to establish normal reference intervals. A cross-sectional study was performed on normal fetuses (n = 314) and on fetuses with CHD (n = 49). CHD cases were divided into groups on the basis of heart function as defined by Huhta score: with CHF (n = 11) and without CHF (n = 38). Fetal bladder volume was measured by two-dimensional as well as three-dimensional ultrasound with the Virtual Organ Computer-aided Analysis technique. In normal fetuses, UPR increased with gestational age from a mean value of 4.5 mL/h at 21 weeks to 66.5 mL/h at 40 weeks. There were no significant differences in UPR between CHD cases without CHF and controls (P = 0.581). However, fetuses with CHF had decreased UPR values (Controls vs. CHF, P = 0.007) compared with the controls. The interobserver and intraobserver variability for assessment of UPR was better using three-dimensional rather than two-dimensional US. Fetal UPR can be reproducibly measured by the Virtual Organ Computer-aided Analysis technique. It appears that heart function, rather than CHD, affects fetal UPR. Fetal UPR may therefore be an additional indicator of CHF.

Inter‐observer variability in the histological assessment of colorectal polyps detected through the NHS Bowel Cancer Screening Programme

Although effective clinical management of colorectal polyps detected through the National Health Service (NHS) Bowel Cancer Screening Programme (BCSP) is dependent on the quality of pathological diagnosis, there have been few attempts to formally evaluate inter-observer variability in histological assessment. The aim of this study was to examine the impact of inter-observer variability on the reported prevalence of prognostic features in a large series of screen-detected colorectal polyps. A retrospective series of 1329 screen-detected polyps (2008-10) was identified from computerized records at two histopathology departments participating in the NHS BCSP. Slides from a sample of 239 polyps were exchanged between centres for independent review and measurement of inter-observer (kappa) agreement. There were significant between-centre differences in the prevalence of polyps with high-risk histological features. Diagnostic review demonstrated good reliability with respect to the assessment of adenomatous change (κ = 0.83), excision margin status (κ = 0.74), high-grade dysplasia (0.61) and invasive malignancy (κ = 0.84). By contrast, there were significant inter-observer differences in the classification of villous lesions (0.18) despite recent efforts to standardize reporting practice. Inter-observer variability in the assessment of screen-detected colorectal polyps limits the prognostic value of histological subtyping and highlights the need for clarification of existing diagnostic criteria.

Multigate Quality Doppler Profiles and Morphological/Hemodynamic Alterations in Multiple Sclerosis Patients

Venous echo-color-Doppler (ECD) showed that chronic cerebrospinal venous insufficiency (CCSVI) syndrome is related to multiple sclerosis (MS). Study aims were to assess interobserver variability in ultrasound evaluation of MS patients and to relate echo-markers to MS clinical symptoms and the disability degree. 277 MS patients (117 men, mean age 43.05+10.04 years) admitted to the Neurology Department of Bari University General Hospital, underwent clinical, Expanded Disability Status Scale (EDSS) evaluation, and a cerebro-venous system ECD evaluation. Two operators reevaluated 32 patients to calculate interobserver variability. McNemar test confirmed the procedure reproducibility between two operators (p=ns). Septa/membranes correlated with deep cerebral veins reflux [right: 16% absence vs. 58% presence, p < 0.0001; left: 26% vs. 50%, p < 0.0001]; their absence in Primary Progressive (PP) MS form [right: 11% vs. 2%, p < 0.001; left: 12% vs. 2%, p < 0.001]. Internal jugular veins (IJVs) reflux absence was in Relapsing-remitting (RR) form [right: 60% vs. 74%, p=0.036; left: 56% vs. 85%, p < 0.0001] like hemodynamically significant stenosis [right: 57% vs. 69%, p=0.033; left: 49% vs. 73%, p < 0.001] not present in PP [right: 11% vs. 2%, p < 0.001; left: 10% vs. 3%, p=0.009]. A supine IJVs blocked flow was related to the EDSS class [right: 4.8±1.5 vs. 5.4±1.4, p=0.006; left: 4.7±1.6 vs. 5.5±1.2, p < 0.0001]; its absence was linked to RR [right: 60% vs. 76%, p=0.016; left: 58% vs. 79%, p < 0.001]. ECD has an important value in MS patients with IJV anomalies detection and a good interobserver procedure reproducibility. MS is associated with CCSVI, although further studies are needed.

Spectral CT with Metal Artifacts Reduction Software for Improvement of Tumor Visibility in the Vicinity of Gold Fiducial Markers

To evaluate spectral computed tomography (CT) with metal artifacts reduction software (MARS) for reduction of metal artifacts associated with gold fiducial seeds. Thirteen consecutive patients with 37 fiducial seeds implanted for radiation therapy of abdominal lesions were included in this HIPAA-compliant, institutional review board-approved prospective study. Six patients were women (46%) and seven were men (54%). The mean age was 61.1 years (median, 58 years; range, 29-78 years). Spectral imaging was used for arterial phase CT. Images were reconstructed with and without MARS in axial, coronal, and sagittal planes. Two radiologists independently reviewed reconstructions and selected the best image, graded the visibility of the tumor, and assessed the amount of artifacts in all planes. A linear-weighted κ statistic and Wilcoxon signed-rank test were used to assess interobserver variability. Histogram analysis with the Kolmogorov-Smirnov test was used for objective evaluation of artifacts reduction. Fiducial seeds were placed in pancreas (n = 5), liver (n = 7), periportal lymph nodes (n = 1), and gallbladder bed (n = 1). MARS-reconstructed images received a better grade than those with standard reconstruction in 60% and 65% of patients by the first and second radiologist, respectively. Tumor visibility was graded higher with standard versus MARS reconstruction (grade, 3.7 ± 1.0 vs 2.8 ± 1.1; P = .001). Reduction of blooming was noted on MARS-reconstructed images (P = .01). Amount of artifacts, for both any and near field, was significantly smaller on sagittal and coronal MARS-reconstructed images than on standard reconstructions (P < .001 for all comparisons). Far-field artifacts were more prominent on axial MARS-reconstructed images than on standard reconstructions (P < .01). Linear-weighted κ statistic showed moderate to perfect agreement between radiologists. CT number distribution was narrower with MARS than with standard reconstruction in 35 of 37 patients (P < .001). Spectral CT with use of MARS improved tumor visibility in the vicinity of gold fiducial seeds.

514 P16INK4a is a marker of good prognosis in primary invasive penile squamous cell carcinoma: A multi-institutional study

Purpose: We assessed the prognostic role of p16INK4a expression in penile cancer with respect to cancer specific survival.Materials and Methods: Based on a multi-institutional collaboration wax embedded tissues from 92 surgically treated patients, including 27 with total and 65 with partial penectomy, were retrospectively evaluated. After a central histopathological review by 1 pathologist a tissue microarray was constructed for p16INK4a immunostaining. Two independent pathologists evaluated p16INK4a expression, which was correlated with cancer specific survival. The κ statistic was used to assess interobserver variability. Univariate and multivariate Cox proportional hazards analysis was applied to assess the independent effects of prognostic factors on cancer specific survival during a median postoperative followup of 32 months (IQR 6–66).Results: The κ statistic revealed excellent interobserver agreement (κ 0.934, p <0.001). Two and 5-year cancer specific survival rates for the entire study cohort were...

P16INK4a is a Marker of Good Prognosis for Primary Invasive Penile Squamous Cell Carcinoma: A Multi-Institutional Study

We assessed the prognostic role of p16(INK4a) expression in penile cancer with respect to cancer specific survival. Based on a multi-institutional collaboration wax embedded tissues from 92 surgically treated patients, including 27 with total and 65 with partial penectomy, were retrospectively evaluated. After a central histopathological review by 1 pathologist a tissue microarray was constructed for p16(INK4a) immunostaining. Two independent pathologists evaluated p16(INK4a) expression, which was correlated with cancer specific survival. The κ statistic was used to assess interobserver variability. Univariate and multivariate Cox proportional hazards analysis was applied to assess the independent effects of prognostic factors on cancer specific survival during a median postoperative followup of 32 months (IQR 6-66). The κ statistic revealed excellent interobserver agreement (κ 0.934, p <0.001). Two and 5-year cancer specific survival rates for the entire study cohort were 86% and 74%, respectively. The 2 and 5-year rates for patients without and with p16(INK4a) expression differed significantly (73% and 57% vs 95% and 85%, respectively, p = 0.011). Univariate analysis revealed p16(INK4a) expression as a significant prognostic factor with respect to cancer specific survival (p = 0.018). Multivariate analysis identified koilocytosis (HR 0.24, 95% CI 0.07-0.83, p = 0.024), p16(INK4a) expression (HR 0.44, 95% CI 0.23-0.84, p = 0.013), and histological stage (HR 3.54, 95% CI 1.88-6.67, p <0.001) and grade (HR 2.47, 95% CI 1.00-6.09, p = 0.049) as independent prognostic factors for cancer specific survival. Results show that p16(INK4a) seems to be a prognostic parameter for primary invasive penile cancer with excellent interobserver reproducibility. At pathology laboratories without antibodies against p16(INK4a) conventional histological determination of koilocytosis by the pathologist also appears to provide important prognostic information for cancer specific survival.

Assessment of Interobserver Variability in Mitotic Figure Counting in Different Histological Grades of Oral Squamous Cell Carcinoma

Mitotic counting is often used for classification, grading and prognosis of tumors. The count usually stands as a decision point for treatment as well. The easiest way of counting the number of mitoses is done by screening routine H&E stained slides. However, for proper mitotic counting, certain strict protocols should be taken into consideration. This study on 30 cases of different grades of oral squamous cell carcinoma was undertaken to determine the interobserver variations in two different groups: Group1 (A1, A2), who were given certain criteria to be followed during the counting of the mitotic figures and group 2 investigators (B1, B2) who were unaware of such criteria. The paired t-test gives a correlation of 0.988 and a significant difference of 0.000 between the two investigators in group 1. The correlation was 0.650 with a significant difference of 0.058 between two investigators in-group 2, indicating that group 1 observers exhibit good interobserver agreement. The results emphasize that following of strict protocols are of great help in determining the accuracy of mitotic counting.

Expression of p53, p21 and cyclin D1 in penile cancer: p53 predicts poor prognosis

AimsTo evaluate the role of p53, p21 and cyclin D1 expression in patients with penile cancer (PC).MethodsParaffin-embedded tissues from PC specimens from six pathology departments were subjected to a central...

Variabilidade interobservadores no diagnóstico de lesões precursoras do câncer anal: estudo do cenário habitual

To assess interobserver variability in the diagnosis of anal cancer precursor lesions in the usual scenario of a service consisting of pathologists without previous experience in the diagnosis of these lesions. Five hundred and two anal specimens taken from 372 HIV-positive and HIV-negative patients were analyzed at the Pathology Department of the Tropical Medicine Foundation of Amazonas by three pathologists with extensive experience in the diagnosis of infectious and tropical diseases, but without significant prior experience in the diagnosis of anal cancer precursor lesions. The individual readings of each pathologist were compared to the one following the consensus diagnosis in shared optical microscope by kappa statistics. The absolute agreement between each individual diagnosis and corresponding consensus was poor (kappa = -0.002). Considering only the positive or negative results for anal squamous intraepithelial lesions, we obtained a fair agreement between observers (kappa = 0.35), while the agreement was moderate when the histopathological findings were considered positive or negative for high-grade squamous intraepithelial lesion or cancer (kappa = 0.52). The interobserver variability in histopathologic diagnosis of anal cancer and its precursor lesions among pathologists with little experience in the area is such that the diagnoses in this field and this scenario should always be a consensus.

Preliminary intraobserver and interobserver variability in wall stress and rupture risk assessment of abdominal aortic aneurysms using a semiautomatic finite element model

We investigated the intraobserver and interobserver variability of using semiautomatic finite element analysis to calculate the von Mises stress and peak wall rupture risk (PWRR) in patients with an abdominal aortic aneurysm (AAA) in longitudinal studies. Four independent observers made 3-dimensional (3D) reconstructions, with minimal manual adjustments, of small AAAs (<5.0 cm) in 17 patients and processed finite element analysis. We used semiautomatic diagnostic software with a finite element model (A4research, VASCOPS GmbH, Graz, Austria). The finite element method was used to calculate von Mises stress and PWRR, which are indicators for wall stress. The differences of each pair of measurements of von Mises stress and PWRR were plotted against their mean and the difference of the mean, according to Bland-Altman analysis. The intraobserver variability had an overall mean percentage difference of 6.86% ± 6.46% for the von Mises stress and 7.70% ± 6.26% for PWRR. The interobserver variability for the four observers showed an overall mean percentage difference of 7.09% ± 6.16% for the von Mises stress and 9.47% ± 8.18% for the PWRR measurement. No significant differences were found (P < .05), for the von Mises stress and PWRR for all observers. The von Mises stress and PWRR of small AAAs calculated in this semiautomatic finite element analysis program show good interobserver and intraobserver variability. It is suitable for clinical use to evaluate mechanical aortic wall characteristics and to compare it with other current methods such as maximum aortic diameter measurements.

Reproducibility and Fidelity of the R.E.N.A.L. Nephrometry Score

The R.E.N.A.L nephrometry score (NS) was developed to characterize renal tumor anatomy to facilitate standardized reporting and ultimately clinical decision making. Up to three points are assigned for each of the following criteria: Tumor size (R), exophytic vs endophytic nature (E), nearness to the collecting system (N), anterior vs posterior (A), and polar location (L), with more complex lesions receiving higher scores. There are no independent studies to date that validate the reproducibility of this scoring system. Our aim was to validate the R.E.N.A.L. NS system by assessing interobserver variability, and therefore reproducibility and fidelity of this proposed assessment tool. We reviewed our prospectively collected laparoscopic partial nephrectomy (LPN) database and identified 306 patients with available preoperative CT or MRI. Of these, 149 were independently read by two urology residents who assigned NS. The Pearson test was used to assess interobserver variability of total NS as well as each of the five components of the scoring system. Interobserver correlation of total NS calculated by the Pearson test was found to be 0.92 (P<0.001). Concordance rates for each of the individual nephrometry components R.E.N.A.L (hilar) were 96%, 92%, 86%, 96%, 89%, and 99% respectively. A t test showed no significant difference between final NS assigned by two different observers. The R.E.N.A.L. NS system is a comprehensive and reproducible tool that may aid surgeons in communicating tumor characteristics effectively. Interobserver correlation is high, rendering it a high fidelity assessment tool.

EFFECT OF ACQUISITION TIME AND CHEMICAL FAT SUPPRESSION ON MENINGEAL ENHANCEMENT ON MR IMAGING IN DOGS

Our purpose was to characterize meningeal gadolinium enhancement on magnetic resonance (MR) imaging in dogs with inflammatory and neoplastic diseases, and to assess interobserver variability and the impact of delayed acquisition and chemical fat saturation on its conspicuity. Transverse T1-weighted FLAIR images were acquired prior to, and immediately following gadolinium injection (T0), and at 5 (T5) and 15-20 min delay (TD), with and without fat suppression, in 155 consecutive dogs imaged for suspected brain disease. The agreement on meningeal enhancement was globally substantial (kappa = 0.61) and the likelihood of obtaining a definite diagnosis was significantly increased with the use of fat suppression (P < or = 0.004). Meningeal enhancement was judged definitively present by consensus in 46 of 155 (30%) dogs. Of these, meningeal enhancement was characterized qualitatively and quantitatively in 30 dogs with a clinical diagnosis (18 inflammatory, 11 neoplastic, 1 infarct), and image sequences were compared. Meningeal enhancement was more often diffuse and leptomeningeal in animals with inflammation versus neoplasia (50% vs. 42%, and 69% vs. 48%, respectively), but significant associations were not found. Meningeal thickness and contrast ratio were higher with neoplasia (P < or = 0.02), but results did not vary significantly between series for either group. Yet, images with fat suppression were most useful 50% of the time for definite diagnosis and/or characterization of meningeal enhancement. While delayed image acquisition following gadolinium injection does not improve characterization of meningeal enhancement in dogs, fat suppression is beneficial qualitatively.

Assessment of right ventricle volumes and function by cardiac MRI: Quantification of the regional and global interobserver variability

Reproducibility of the manual assessment of right ventricle volumes by short-axis cine-MRI remains low and is often attributed to the difficulty in separating the right atrium from the ventricle. This study was designed to evaluate the regional interobserver variability of the right ventricle volume assessment to identify segmentation zones with the highest interobserver variability. Short-axis views of 90 right ventricles (30 hypertrophic, 30 dilated, and 30 normal) were acquired with 2D steady-state free precession sequences at 1.5 T and were manually segmented by two observers. The two segmentations were compared and the variations were quantified with a variation score based on the Hausdorff distance between the two segmentations and the interobserver 95% limits of concordance of the global volumes. The right ventricles were semiautomatically split into four subregions: apex, mid-ventricle, tricuspid zone, and infundibulum. These four subregions represented 11%, 34%, 36%, and 19% of the volume but, respectively, yielded variation scores of 8%, 16%, 42%, and 34%. The infundibulum yielded the highest interobserver regional variability although its variation score remained comparable to the tricuspid zone due to its lower volume. These results emphasize the importance of standardizing the segmentation of the infundibulum and the tricuspid zone to improve reproducibility.