High Interobserver Variability Research Articles

Stalk versus base invasion in pT1 papillary cancers of the bladder: improved substaging system predicting the risk of progression.

Pathological stage pT1 bladder cancers constitute a clinically heterogeneous group. However, current staging guidelines for superficially invasive cancers do not acknowledge the variability in type and extent of lamina propria invasion in papillary urothelial carcinomas (PUCs), and historically proposed substaging systems showed either high interobserver variation or limited value in predicting patient outcomes. The aim of this study was to reappraise pT1 PUC substaging, with the objective of identifying a novel scheme that is reproducible and prognostically meaningful. pT1 PUCs diagnosed during 1999-2015 were retrospectively reviewed and characterized as focal invasion confined to the papillary stalk, focal invasion of the tumour base, or extensive invasion of the tumour base. Cases with concurrent flat carcinoma insitu, angiolymphatic invasion, absent muscularis propria or clinically advanced disease were excluded. We calculated cumulative incidence rates of recurrence, progression and death by tumour subtype, and evaluated differential risks by using log-rank tests and Kaplan-Meier curves stratified by type and extent of invasion. Among 62 patients satisfying the inclusion criteria, 22 of 29 patients with base-extensive invasion progressed, whereas four of 13 with base-focal and none of 20 with stalk-only invasion progressed. There was strong evidence that base-extensive patients had a higher risk of progression and death resulting from bladder cancer than base-focal or stalk-only patients (P<0.0001). However, tumour subtype was not significantly associated with risk of recurrence (P=0.21). We propose an innovative substaging approach for reporting the site and extent of lamina propria invasion in patients with pT1 PUC, allowing patient stratification for risk of progression.

Light-sheet microscopy for slide-free non-destructive pathology of large clinical specimens.

For the 1.7 million patients per year in the U.S. who receive a new cancer diagnosis, treatment decisions are largely made after a histopathology exam. Unfortunately, the gold standard of slide-based microscopic pathology suffers from high inter-observer variability and limited prognostic value due to sampling limitations and the inability to visualize tissue structures and molecular targets in their native 3D context. Here, we show that an open-top light-sheet microscope optimized for non-destructive slide-free pathology of clinical specimens enables the rapid imaging of intact tissues at high resolution over large 2D and 3D fields of view, with the same level of detail as traditional pathology. We demonstrate the utility of this technology for various applications: wide-area surface microscopy to triage surgical specimens (with ~200 μm surface irregularities), rapid intraoperative assessment of tumour-margin surfaces (12.5 sec/cm2), and volumetric assessment of optically cleared core–needle biopsies (1 mm in diameter, 2 cm in length). Light-sheet microscopy can be a versatile tool for both rapid surface microscopy and deep volumetric microscopy of human specimens.

Volumetric assessment in malignant pleural mesothelioma.

Malignant pleural mesothelioma (MPM) is a highly malignant pleural neoplasm with a dismal prognosis. Multimodality approach including surgery and chemotherapy are utilized to treat patients with resectable disease. Clinical staging allows for selection of patients for treatment strategies, but has not been found to be prognostic and is plagued by high interobserver variability. Tumor volume measurement on cross-sectional imaging has emerged as a potential quantitative tool with prognostic significance. This review focuses on volumetric assessment from cross-sectional imaging (CT, MRI, 18F-FDG PET/CT) and the potential prognostic benefit and applications.

CONSISTENCY TO CONSENSUS: HOW STANDARDIZATION OF POSTNATAL MANAGEMENT OF PRENATAL HYDRONEPHROSIS IMPACTED CARE

CONSISTENCY TO CONSENSUS: HOW STANDARDIZATION OF POSTNATAL MANAGEMENT OF PRENATAL HYDRONEPHROSIS IMPACTED CARE

Usefulness of the Mitral Regurgitation Severity Index to Assess the Severity of Chronic Mitral Regurgitation.

Existing metrics for grading mitral regurgitation (MR) are limited and fraught with high interobserver variability. We developed and evaluated a Doppler-based, semiquantitative novel index (Mitral Regurgitation Severity Index [MRSI]) of MR severity. In a total of 125 patients (70 in the derivation cohort and 55 in the validation cohort), MRSI was calculated as a ratio of time velocity integral of mitral inflow (continuous-wave Doppler-TVI MV) to the time velocity integral of the left ventricle outflow (pulse-wave Doppler-TVI LVOT). Inter-rater agreement for MRSI and predictive ability of the MRSI were then assessed. In the derivation cohort, MRSI differed significantly between patients with severe MR (2.6 ± 0.51) and mild-moderate (nonsevere) MR (1.4 ± 0.18) and a cutoff of ≥1.8 was associated with optimal diagnostic accuracy. In the validation cohort, MRSI exhibited excellent agreement between a level II and a level III reader with a mean difference of-0.14 (95% confidence limit of agreement:-0.80 to 0.53), correlation coefficient of 0.88 (p <0.001), and 16% CV; and using the cut point of 1.8, it exhibited good inter-rater reproducibility with a kappacoefficient of 0.72 (p <0.001). In conclusion, MRSI appears to be a simple, quantitative, practical, color-independent metric to differentiate severe MR from nonsevere MR.

Placental recess accompanied by a T2 dark band: a new finding for diagnosing placental invasion.

Our aim was to assess the usefulness of a new magnetic resonance imaging (MRI) finding, the placental recess, for diagnosing placental invasion. This retrospective study included 51 patients (mean age 34.1years, range 26-43years) with suspected placental invasion who underwent cesarean section. Preoperative MRI was performed using a 1.5-T unit and included axial, sagittal, and coronal T2-weighted imaging (T2WI) with half-Fourier fast spin-echo sequences. Overall, 9 patients showed placental invasion, and 42 did not. Placental recess was defined as a placental deformity with contraction of the placental surface and outer rim of the uterus accompanied by a T2 dark band. Two radiologists independently assessed the presence of the placental recess and conventional findings including uterine bulging, abnormal placental vascularity, placental heterogeneous intensity on T2-weighted imaging (T2WI), and the T2 dark band. Fisher's two-sided exact test was used to compare findings between patients with and without placental invasion. Interobserver reliability was assessed using the kappa statistic. MRI features had interobserver reliability of>0.40. Placental recess yielded the highest kappa value (0.898). Significant differences were identified between patients with and without placental invasion regarding abnormal placental vascularity, placental heterogeneous intensity, a T2 dark band, and the placental recess on T2WI (p=0.0282, 0.0003, 0.0003,<0.0001, respectively). The placental recess had sensitivity, specificity, positive and negative predictive values, and accuracy of 56, 100, 100, 91, and 92%, respectively. The placental recess was useful for diagnosing placental invasion, with high interobserver variability and accuracy.

Automatic Cochlear Duct Length Estimation for Selection of Cochlear Implant Electrode Arrays.

Cochlear duct length (CDL) can be automatically measured for custom selection of cochlear implant (CI) electrode arrays. CI electrode array selection can be influenced by measuring the CDL, which is estimated based on the length of the line that connects the round window and the lateral wall of the cochlea when passing through the modiolus. CDL measurement remains time consuming and inter-observer variability has not been studied. We evaluate an automatic approach to directly measure the two-turn (2T) CDL using existing algorithms for localizing cochlear anatomy in computed tomography (CT). Pre-op CT images of 309 ears were evaluated. Two fellowship-trained neurotologists manually and independently measured CDL. Inter-observer variability between measurements across expert and automatic observers is assessed. Inter-observer differences for choice of electrode type are also investigated. Manual measurement of CDL by experts tends to underestimate cochlea size and has high inter-observer variability, with mean absolute differences between expert CDL estimations of 1.15 mm. Our results show that this can lead to a large number of cochleae for which a different electrode array type would be selected by different observers, depending on the specific threshold value of CDL used to decide between array type. Choosing the best CI electrode array is an important task for optimizing hearing outcomes. Manual cochleae length measurements are user-dependent, and errors impact upon the CI electrode array choice for certain patients. Measuring cochlea length automatically is less time consuming and generates more repeatable results. Our automatic approach could make use of CDL for patient-customized treatment more clinically adoptable.

Accuracy of vascular invasion reporting in hepatocellular carcinoma before and after implementation of subspecialty surgical pathology sign-out.

Liver cancers (including hepatocellular carcinoma [HCC] and cholangiocarcinoma) are the fifth most common cause of cancer death. The most powerful independent histologic predictor of overall survival after transplantation for HCC is the presence of microscopic vascular invasion (VI). Given that VI is known to have somewhat high interobserver variability in both HCC and other tumors, we hypothesized that pathologists with special interest and training in liver pathology would be more likely to identify and report VI in HCC than would general surgical pathologists. We searched our departmental surgical pathology archives for transplant hepatectomies performed for HCC. We identified 143 such cases with available sign-out reports and hematoxylin and eosin-stained slides. Kappa results (level of agreement) were calculated. Before surgical pathology subspecialty sign-out (SSSO) implementation, 49 of 88 HCC cases were reported as negative for VI; on rereview, 20 of these had VI. After SSSO implementation, 39 of 55 cases were reported as negative for VI; on our review, 8 of these had VI. Kappa (agreement) between general SO and subspecialty rereview was 0.562 (95% confidence interval [CI] = 0.411-0.714) "weak agreement." Kappa (agreement) between SSSO and rereview by select liver pathologists was 0.693 (95% CI = 0.505-0.880) "moderate agreement." Our study is one of only a few so far that have suggested improved accuracy of certain parameters under SSSO.

Prevalence of Extracranial Venous Narrowing on Magnetic Resonance Venography Is Similar in People With Multiple Sclerosis, Their Siblings, and Unrelated Healthy Controls: A Blinded, Case-Control Study

PurposeThe study sought to assess and compare the prevalence of narrowing of the major extracranial veins in subjects with multiple sclerosis and controls, and to assess the sensitivity and specificity of magnetic resonance venography (MRV) for describing extracranial venous narrowing as it applies to the chronic cerebrospinal venous insufficiency theory, using catheter venography (CV) as the gold standard. MethodsThe jugular and azygos veins were assessed with time-of-flight MRV in this assessor-blinded, case-control study of subjects with multiple sclerosis, their unaffected siblings, and unrelated controls. The veins were evaluated by diameter and area, and compared with CV. Collateral vessels were also analyzed for maximal diameter, as a potential indicator of compensatory flow. ResultsA high prevalence of extracranial venous narrowing was demonstrated in all study groups, collectively up to 84% by diameter criteria and 90% by area, with no significant difference between the groups when assessed independently (P = .34 and .63, respectively). There was high interobserver variability in the reporting of vessel narrowing (kappa = 0.32), and poor vessel per vessel correlation between narrowing on MRV and CV (kappa = 0.064). Collateral neck veins demonstrated no convincing difference in maximum size or correlation with jugular narrowing. ConclusionThere is a high prevalence of narrowing of the major extracranial veins on MRV in all 3 study groups, with no significant difference between them. These findings do not support the chronic cerebrospinal venous insufficiency theory. Although MRV has shown a high sensitivity for identifying venous narrowing, time-of-flight imaging demonstrates poor interobserver agreement and poor specificity when compared with the gold standard CV.

Abstract 12945: A Novel Index for Quantification of Interventricular Septal Flattening in Pediatric Patients With Right Ventricular Hypertension

Introduction: Systolic flattening of the inter-ventricular septum (IVS) is a marker of elevated right ventricular pressure (RVp). It can be non-invasively quantified by a left ventricular eccentricity index (LVEI), RV to left ventricular (LV) ratio and a novel index - the systolic flattening angle (SFA). Data evaluating the relationship between these indices and invasive RVp in pediatrics are limited. Methods: Patients referred for right and left heart cardiac catheterizations (CC) were enrolled and underwent echocardiograms after induction of anesthesia. The SFA, RV:LV ratio and LVEI were measured in end-systole. Peak systolic RVp and aortic pressure (pAo) were measured simultaneously during CC. The likelihood of SFA to predict an RVp:pAo ratio of &gt; 0.5 was assessed using receiver-operator curve (ROC) analysis. Bland-Altman plots were used to assess inter-observer agreement. Results: Forty-one unique subjects were enrolled [56% female, median age 2.7 years (0.7, 7.1) and median weight 12.1kg (6.0, 22.3)]. The median mid-ventricular (MV), apical, and sub-costal LVEI was 1.18 (0.9, 1.7), 1.18 (1.0, 1.7), and 1.13 (1.0, 1.5), respectively. The mean SFA was 8.5±9.6°, the median RV:LV ratio was 0.71 (0.46, 1.23), and the mean RVp:pAo ratio was 0.68 ± 0.5. There was a significant negative correlation (r = -0.75, p &lt;0.001) between SFA and RVp:pAo ratio and a modest correlation with other non-invasive indices (Figure). ROC analysis demonstrated an area under the curve = 0.90 for prediction of RVp:pAo &gt; 0.50 by SFA. An SFA &lt; 10.6 had a sensitivity of 88% and specificity of 79% in predicting RVp:pAo &gt; 0.50. SFA was modestly reproducible with a Bland-Altman plot demonstrating a bias of 4.14 and limits of agreement of -7.8 and 16.1. Conclusions: SFA has the best correlation with RVp:pAO, good sensitivity and specificity, but relatively high inter-observer variability. Taken together, these non-invasive measures may help minimize the need for invasive assessment of RVp.

Postediting prostate magnetic resonance imaging segmentation consistency and operator time using manual and computer-assisted segmentation: multiobserver study.

Prostate segmentation on T2w MRI is important for several diagnostic and therapeutic procedures for prostate cancer. Manual segmentation is time-consuming, labor-intensive, and subject to high interobserver variability. This study investigated the suitability of computer-assisted segmentation algorithms for clinical translation, based on measurements of interoperator variability and measurements of the editing time required to yield clinically acceptable segmentations. A multioperator pilot study was performed under three pre- and postediting conditions: manual, semiautomatic, and automatic segmentation. We recorded the required editing time for each segmentation and measured the editing magnitude based on five different spatial metrics. We recorded average editing times of 213, 328, and 393s for manual, semiautomatic, and automatic segmentation respectively, while an average fully manual segmentation time of 564s was recorded. The reduced measured postediting interoperator variability of semiautomatic and automatic segmentations compared to the manual approach indicates the potential of computer-assisted segmentation for generating a clinically acceptable segmentation faster with higher consistency. The lack of strong correlation between editing time and the values of typically used error metrics ([Formula: see text]) implies that the necessary postsegmentation editing time needs to be measured directly in order to evaluate an algorithm's suitability for clinical translation.

Fibrous Cap Thickness by Optical Coherence Tomography In Vivo

BackgroundOptical coherence tomography (OCT) imaging is considered to be the only imaging modality with sufficient resolution to measure fibrous cap thickness (FCT) in vivo. However, reproducibility of the measurements in vivo has been unsatisfactory. ObjectivesThe authors aimed to investigate whether satisfactory reproducibility of FCT measurements by OCT in vivo can be achieved between independent observers. MethodsOne hundred seventy OCT pullbacks were analyzed by 2 independent observers with intravascular imaging expertise in accordance with current guidelines to assess the interobserver variability of FCT measurement by intraclass correlation coefficient (ICC). The main sources of the variability were analyzed and incorporated in lesion assessment criteria. The same 170 OCT pullbacks were reanalyzed by the same observers using the developed criteria, and the interobserver reproducibility of the measurements was reassessed. On the basis of the developed criteria, a third independent observer interpreted all 170 OCT images. Assessment of the maximal lipid arc was also undertaken similarly before and after the development of interpretation criteria. ResultsThe original ICC of the FC thickness was 0.56 (95% confidence interval [CI]: 0.38 to 0.69). The poor definition of necrotic core facing border of FC and the neointimal presence of macrophages and calcification contributed to the high interobserver variability of FCT measurement. The ICC of FCT measurements by OCT in vivo was 0.88 (95% CI: 0.80 to 0.93) after we developed lesion assessment criteria. The ICC for the maximal lipid arc assessment before and after was 0.76 and 0.82 respectively. The third independent observer was extensively coached and returned the ICC of 0.82 (95% CI: 0.74 to 0.87) with observer 1 and 0.90 (95% CI: 0.86 to 0.94) with observer 2. ConclusionsCareful consideration of OCT features mimicking fibroatheroma lesions and imaging artifacts contributed to significantly higher levels of interobserver agreement. Interobserver variation can be partially resolved by development of standard interpretation algorithms.

Automatic basal slice detection for cardiac analysis.

Identification of the basal slice in cardiac imaging is a key step to measuring the ejection fraction of the left ventricle. Despite all the effort placed on automatic cardiac segmentation, basal slice identification is routinely performed manually. Manual identification, however, suffers from high interobserver variability. As a result, an automatic algorithm for basal slice identification is required. Guidelines published in 2013 identify the basal slice based on the percentage of myocardium surrounding the blood cavity in the short-axis view. Existing methods, however, assume that the basal slice is the first short-axis view slice below the mitral valve and are consequently at times identifying the incorrect short-axis slice. Correct identification of the basal slice under the Society for Cardiovascular Magnetic Resonance guidelines is challenging due to the poor image quality and blood movement during image acquisition. This paper proposes an automatic tool that utilizes the two-chamber view to determine the basal slice while following the guidelines. To this end, an active shape model is trained to segment the two-chamber view and create temporal binary profiles from which the basal slice is identified. From the 51 tested cases, our method obtains 92% and 84% accurate basal slice detection for the end-systole and the end-diastole, respectively.

Interobserver variability in identification of breast tumors in MRI and its implications for prognostic biomarkers and radiogenomics.

To assess the interobserver variability of readers when outlining breast tumors in MRI, study the reasons behind the variability, and quantify the effect of the variability on algorithmic imaging features extracted from breast MRI. Four readers annotated breast tumors from the MRI examinations of 50 patients from one institution using a bounding box to indicate a tumor. All of the annotated tumors were biopsy proven cancers. The similarity of bounding boxes was analyzed using Dice coefficients. An automatic tumor segmentation algorithm was used to segment tumors from the readers' annotations. The segmented tumors were then compared between readers using Dice coefficients as the similarity metric. Cases showing high interobserver variability (average Dice coefficient <0.8) after segmentation were analyzed by a panel of radiologists to identify the reasons causing the low level of agreement. Furthermore, an imaging feature, quantifying tumor and breast tissue enhancement dynamics, was extracted from each segmented tumor for a patient. Pearson's correlation coefficients were computed between the features for each pair of readers to assess the effect of the annotation on the feature values. Finally, the authors quantified the extent of variation in feature values caused by each of the individual reasons for low agreement. The average agreement between readers in terms of the overlap (Dice coefficient) of the bounding box was 0.60. Automatic segmentation of tumor improved the average Dice coefficient for 92% of the cases to the average value of 0.77. The mean agreement between readers expressed by the correlation coefficient for the imaging feature was 0.96. There is a moderate variability between readers when identifying the rectangular outline of breast tumors on MRI. This variability is alleviated by the automatic segmentation of the tumors. Furthermore, the moderate interobserver variability in terms of the bounding box does not translate into a considerable variability in terms of assessment of enhancement dynamics. The authors propose some additional ways to further reduce the interobserver variability.

Classification of fatty and dense breast parenchyma: comparison of automatic volumetric density measurement and radiologists’ classification and their inter-observer variation

Automatically calculated breast density is a promising alternative to subjective BI-RADS density assessment. However, such software needs a cutoff value for density classification. To determine the volumetric density threshold which classifies fatty and dense breasts with highest accuracy compared to average BI-RADS density assessment, and to analyze radiologists' inter-observer variation. A total of 537 full field digital mammography examinations were randomly selected from a population based screening program. Five radiologists assessed density using the BI-RADS density scale, where BI-RADS I-II were classified as fatty and III-IV as dense. A commercially available software (Quantra) calculated volumetric breast density. We calculated the cutoff (threshold) values in volumetric density that yielded highest accuracy compared to median and individual radiologists' classification. Inter-observer variation was analyzed using the kappa statistic. The threshold that best matched the median radiologists' classification was 10%, which resulted in 87% accuracy. Thresholds that best matched individual radiologist's classification had a range of 8-15%. A total of 191 (35.6 %) cases were scored both dense and fatty by at least one radiologist. Fourteen (2.6 %) cases were unanimously scored by the radiologists, yet differently using automatic assessment. The agreement (kappa) between reader's median classification and individual radiologists was 0.624 to 0.902, and agreement between median classification and Quantra was 0.731. The optimal volumetric threshold of 10% using automatic assessment would classify breast parenchyma as fatty or dense with substantial accuracy and consistency compared to radiologists' BI-RADS categorization, which suffers from high inter-observer variation.

Radiofrequency ablation for low-grade dysplasia in Barrett's esophagus.

The diagnosis and management of low-grade dysplasia (LGD) in Barrett's esophagus continue to evolve and vary in practice. Radiofrequency ablation (RFA) is now an acceptable option for the treatment of LGD. Therefore, the purpose of this review article is to present current literature on the diagnosis and management of LGD, and to explore when to optimally pursue RFA for the treatment of LGD. The challenge of the management of LGD begins with its diagnosis. Because of high interobserver variability among pathologists in the diagnosis of dysplasia, it is recommended that an expert gastrointestinal pathologist confirms the diagnosis of LGD. Endoscopic mucosal resection is indicated to remove visible lesions in the setting of dysplasia to obtain an accurate diagnosis, especially regarding T staging. Management options include intense surveillance and endoscopic eradication therapy. RFA provides a reasonable method to eradicate flat Barrett's mucosa. Endoscopic eradication of confirmed LGD has been demonstrated to reduce the risk of progression to esophageal cancer. Additional data about the risk factors associated with progression, and the benefits and risks of treatment are discussed, and can be incorporated in patient counseling and decision making. Endoscopic eradication therapy with RFA may be an appropriate treatment option for LGD in Barrett's esophagus after the benefits and risks have been discussed in detail with the patient.

Comparison of prognostic and predictive impact of genomic or central grade and immunohistochemical subtypes or IHC4 in HR+/HER2- early breast cancer: WSG-AGO EC-Doc Trial

Potential prognostic and predictive markers in early, intermediate-risk breast cancer (BC) include histological grade, Ki-67, genomic signatures, e.g. genomic grade index (GGI), and intrinsic subtypes. Their prognostic/predictive impact in hormone receptor (HR: ER and/or PR) positive/HER2- BC is controversial. WSG-AGO EC-Doc demonstrated superior event-free survival (EFS) in patients with 1-3 positive lymph node receiving epirubicin/cyclophosphamide-docetaxel (EC-Doc) versus 5-fluoruracil/epirubicin/cyclophosphamide (FEC). In a representative trial subset, we quantify concordance among factors used for clinical chemotherapy indication. We investigate the impact of central histology (n = 772), immunohistochemistry for intrinsic subtyping and IHC4, and dichotomous (GG) or continuous (GGI) genomic grade (n = 472) on patient outcome and benefit from taxane chemotherapy, focusing on HR+/HER2- patients (n = 459). Concordance of local grade (LG) with central (CG) or genomic grade was modest. In HR+/HER2- patients, low (GG-1: 16%), equivocal (GG-EQ: 17%), and high (GG-3: 67%) GG were associated with respective 5-year EFS of 100%, 93%, and 85%. GGI was prognostic for EFS within all LG subgroups and within CG3, whereas IHC4 was prognostic only in CG3 tumors.In unselected and HR+/HER2- patients, CG3 and luminal-A-like subtype entered the multivariate EFS model, but not IHC4 or GG. In the whole population, continuous GGI entered the model [hazard ratio (H.R.) of 75th versus 25th = 2.79; P = 0.01], displacing luminal-A-like subtype; within HR+/HER2- (H.R. = 5.36; P < 0.001), GGI was the only remaining prognostic factor.In multivariate interaction analysis (including central and genomic grade), luminal-B-like subtype [HR+ and (Ki-67 ≥20% or HER2+)] was predictive for benefit of EC-Doc versus FEC in unselected but not in HR+/HER2- patients. In the WSG-AGO EC-Doc trial for intermediate-risk BC, CG, intrinsic subtype (by IHC), and GG provide prognostic information. Continuous GGI (but not IHC4) adds prognostic information even when IHC subtype and CG are available. Finally, the high interobserver variability for histological grade and the still missing validation of Ki-67 preclude indicating or omitting adjuvant chemotherapy based on these single factors alone. The WSG-AGO/EC-Doc is registered at ClinicalTrials.gov, NCT02115204.

Image-based computational quantification and visualization of genetic alterations and tumour heterogeneity.

Recent large-scale genome analyses of human tissue samples have uncovered a high degree of genetic alterations and tumour heterogeneity in most tumour entities, independent of morphological phenotypes and histopathological characteristics. Assessment of genetic copy-number variation (CNV) and tumour heterogeneity by fluorescence in situ hybridization (ISH) provides additional tissue morphology at single-cell resolution, but it is labour intensive with limited throughput and high inter-observer variability. We present an integrative method combining bright-field dual-colour chromogenic and silver ISH assays with an image-based computational workflow (ISHProfiler), for accurate detection of molecular signals, high-throughput evaluation of CNV, expressive visualization of multi-level heterogeneity (cellular, inter- and intra-tumour heterogeneity), and objective quantification of heterogeneous genetic deletions (PTEN) and amplifications (19q12, HER2) in diverse human tumours (prostate, endometrial, ovarian and gastric), using various tissue sizes and different scanners, with unprecedented throughput and reproducibility.

Toward automated classification of acetabular shape in ultrasound for diagnosis of DDH: Contour alpha angle and the rounding index

Background and objectivesThe diagnosis of Developmental Dysplasia of the Hip (DDH) in infants is currently made primarily by ultrasound. However, two-dimensional ultrasound (2DUS) images capture only an incomplete portion of the acetabular shape, and the alpha and beta angles measured on 2DUS for the Graf classification technique show high inter-scan and inter-observer variability. This variability relates partly to the manual determination of the apex point separating the acetabular roof from the ilium during index measurement. This study proposes a new 2DUS image processing technique for semi-automated tracing of the bony surface followed by automatic calculation of two indices: a contour-based alpha angle (αA), and a new modality-independent quantitative rounding index (M). The new index M is independent of the apex point, and can be directly extended to 3D surface models. MethodsWe tested the proposed indices on a dataset of 114 2DUS scans of infant hips aged between 4 and 183 days scanned using a 12MHz linear transducer. We calculated the manual alpha angle (αM), coverage, contour-based alpha angle and rounding index for each of the recordings and statistically evaluated these indices based on regression analysis, area under the receiver operating characteristic curve (AUC) and analysis of variance (ANOVA). ResultsProcessing time for calculating αA and M was similar to manual alpha angle measurement, ∼30s per image. Reliability of the new indices was high, with inter-observer intraclass correlation coefficients (ICC) 0.90 for αA and 0.89 for M. For a diagnostic test classifying hips as normal or dysplastic, AUC was 93.0% for αA vs. 92.7% for αM, 91.6% for M alone, and up to 95.7% for combination of M with αM, αA or coverage. ConclusionsThe rounding index provides complimentary information to conventional indices such as alpha angle and coverage. Calculation of the contour-based alpha angle and rounding index is rapid, shows potential to improve the reliability and accuracy of DDH diagnosis from 2DUS, and could be extended to 3D ultrasound in future.

Acute Abdominal Pathology in Pregnant Women

Imaging pregnant patients comes with several challenges, in terms of the type of pathology (maternal, fetal or placental), choosing the optimal imaging modality and minimizing the radiation risk. Radiation exposure information is sometimes limited and confusing, for both referring physicians and patients, and may lead to uninformed decisions on the appropriateness and risks of the exam. Overestimating a negligible risk following a low radiation dose can induce excessive and unnecessary anxiety among patients or delay diagnosis and treatment. Underestimation of the risk can result in fetal non-carcinogenic (teratogenic) effects and both fetal and maternal carcinogenic effects. Fetal radiation doses, above 100 and 150 mGy respectively, bear a potential and high non-carcinogenic risk. The reported lifetime attributable risk of cancer incidence for fetuses exposed to ionizing radiation (X-ray and computed tomography) ranges between 2.5 and 5% per 100 mSv. Magnetic resonance (MR) imaging (1.5 and 3.0 Tesla) can be performed in all trimesters if deemed clinically necessary (and no delay until completion is possible) and ideally as an adjunct to initial ultrasound evaluation. Although no teratogenic adverse effects are reported for iodinated and gadolinium based contrast agents, informed consent is recommended when their use is considered clinically necessary. A typical example of the use of abdominal computed tomography in pregnant patients is a severe/high-velocity trauma. Placental pathology is often less known by radiologists. Following an inconclusive sonographic examination, MR is the imaging modality of choice for the placenta. A minimal MR checklist should examine the location and morphology of the placenta, document placental and umbilical cord anomalies and look for placental invasion. The MR key signs of placental invasion are an abnormal uterine bulging, heterogeneity of the signal intensity within the placenta, and the presence of dark intraplacental bands on T2-images. One should note that radiological techniques often fail to accurately determine the extent of placental invasion. Fetopelvic disproportion, particularly in primigravida, typically presents late in pregnancy. Maternal pelvimetry might be a helpful adjunct to clinical parameters and examination. X-ray pelvimetry should be avoided, because of the relatively low accuracy of the modality, the high interobserver variability and the use of ionizing radiation. MR is the modality of choice to measure the pelvic diameters and if not available, low-dose computed tomography might be an alternative. When using pelvimetry reference values, one should pay attention to the number of participants included in the study, the imaging modality used in the study and the reason for pelvimetric imaging of the patient group (e.g. spontaneous deliveries, suspected fetal-pelvic disproportion, all referred patients, etc.).