Intraobserver Variability Research Articles

Automatic segmentation-based multi-modal radiomics analysis of US and MRI for predicting disease-free survival of breast cancer: a multicenter study.

Several studies have confirmed the potential value of applying radiomics to predict prognosis of breast cancer. However, the tumor segmentation in these studies depended on delineation or annotation of breast cancer by radiologist, which is often laborious, tedious, and vulnerable to inter- and intra-observer variability. Automatic segmentation is expected to overcome this difficulty. Herein, we aim to investigate the value of automatic segmentation-based multi-modal radiomics signature and magnetic resonance imaging (MRI) features in predicting disease-free survival (DFS) of patients diagnosed with invasive breast cancer. This retrospective multicenter study included a total of 643 female patients with invasive breast cancer who underwent preoperative ultrasound (US) and MRI for prognostic analysis. Data (n = 480) from center 1 were divided into training and internal testing sets, while data (n = 163) from centers 2 and 3 were analyzed as the external testing set. We developed automatic segmentation frameworks for tumor segmentation by deep learning. Then, Least absolute shrinkage and selection operator Cox regression was used to select features to construct radiomics signature, and corresponding radiomics score (Rad-score) was calculated. Finally, six models for predicting DFS were constructed by using Cox regression and assessed in terms of discrimination, calibration, and clinical usefulness. The multi-modal radiomics signature combining intra- and peri-tumoral radiomics signatures of US and MRI achieved a higher C-index in the internal (0.734) and external (0.708) testing sets than most other radiomics signatures in predicting DFS, and successfully stratified patients into low- and high-risk groups. The multi-modal clinical imaging model combining the multi-modal Rad-score and clinical traditional MRI model-score resulted in a higher C-index (0.795) than other models in the external testing set, and it had a better calibration and higher clinical benefit. This study demonstrates that the multi-modal radiomics signature derived from automatic segmentations of US and MRI is a promising risk stratification biomarker for breast cancer, and highlights that the appropriate combination of multi-modal radiomics signature, clinical characteristics, and MRI feature can improve performance of individualized DFS prediction, which might assist in guiding decision-making related to breast cancer.

Abstract 4135318: Comparison of Right Ventricular Function Changes Between Volume Control and Pressure Support Ventilation Modes: A Prospective Study Using S' Prime Values

Background: This study assesses changes in right ventricular function by measuring S' values during the transition from volume control-assist control (VC-AC) to pressure support ventilation (PSV). Transitioning from VC-AC to PSV reduces the mean airway pressure (P mean) which causes a decrease in right ventricular afterload and also potentially improves venous return and right ventricular preload. PSV allows spontaneous breathing efforts, possibly increasing cardiac output. Thus, we expect a consistent increase in S' value when changing ventilator modes. Methods: This prospective observational study included ICU patients on invasive mechanical ventilation for over 24 hours. The clinical team decided when to switch from VC-AC to PSV. An average of three S' measurements was taken during VC-AC and within 10 minutes of transitioning to PSV. Sedation, vasopressor use, and Richmond Agitation-Sedation Scale scores were consistent throughout the measurements. The study was approved by the Albert Einstein College of Medicine IRB (#2023-15068) Results: Seventeen patients were included in the study, with an average age of 61.29 years (SD 11.39). Of these, 6 (35.3%) were female. The mean S' during VC-AC was 12.79 ± 4.52, and upon transitioning to PSV, it was 14.56 ± 3.89 (p-value 0.47). The intra-observer variability for S' was 5.3%. Of the seventeen patients, 5/17 (29%) showed an S' change of less than 5.3% and were deemed to have no significant variation; 7/17 (41%) experienced a significant decrease in S' (15.41% ± 7.29% SD), while the remaining 5/17 (29%) showed a significant increase (19.31 ± 13.23%). Conclusion: Our study demonstrates that transitioning from VC-AC to PSV results in variable changes in right ventricular systolic function, as indicated by S' values. Contrary to our expectations of a consistent increase, the interplay between S' values and ventilator mode changes appears to be non-linear, suggesting other factors may influence cardiopulmonary interactions in spontaneously breathing ventilated patients. These findings highlight the complexity of these interactions, underscoring the need for individualized patient management and paving the way for further research with a larger cohort.

AI-powered precision: breast carcinoma diagnosis through digital proliferation index (Ki-67) assessment in pathological anatomy

PurposeThe primary objective of the study is to enhance the accuracy and efficiency of assessing the proliferation index in cancer cells, specifically focusing on the role of Ki-67. The purpose is to address the limitations of traditional visual assessments conducted by pathologists by integrating AI technologies, particularly deep learning. By accurately computing the percentage of Ki-67-labeled cells, the research aims to streamline the diagnostic process, reduce subjectivity and contribute to the advancement of diagnostic precision in pathological anatomy.Design/methodology/approachThe research employs a methodological approach that integrates Ki-67, a non-histone nuclear protein, as a vital biomarker for assessing the proliferative status of cancer cells. Given the challenges associated with traditional visual assessments by pathologists, including inter- and intra-observer variability and time-consuming efforts, the study adopts a novel methodology leveraging artificial intelligence (AI) solutions. Deep learning is applied to precisely calculate the percentage of Ki-67-labeled cells. The process involves pathologists delineating the tumor area at x40 magnification, enabling the segmentation of various cell types (positive, negative and tumor-infiltrating lymphocytes). The subsequent percentage calculation enhances efficiency and minimizes subjectivity in the diagnostic process.FindingsDespite inherent errors, the research findings indicate that the model surpasses existing benchmarks, showcasing superior accuracy in terms of average error measurement. The comparison with diverse datasets and benchmarking against pathologists’ diagnoses contributes empirical evidence to support the effectiveness of the AI-based model in accurately computing the percentage of Ki-67-labeled cells. These findings signify a noteworthy advancement in diagnostic methodologies and reinforce the potential of AI technologies in improving the precision of cancer diagnostics within the realm of pathological anatomy.Originality/valueThe research contributes to the field by introducing an innovative approach that combines Ki-67 as a biomarker and AI technologies for improved diagnostic precision. The originality lies in the utilization of deep learning to calculate the percentage of labeled cells, mitigating the challenges associated with manual assessments. The validation of the model against diverse datasets and benchmarking against pathologists’ diagnoses demonstrates its superior accuracy, highlighting the value of integrating AI in pathological anatomy for enhanced diagnostic outcomes. The study represents a significant stride in original research, offering novel insights and methodologies in the pursuit of more precise cancer diagnostics.

Measuring up: Ensuring Intra- and Interobserver Reliability in Stretched Penile Length with the SPLINT Technique

ABSTRACT Background: A discrepancy between the true and measured value of stretched penile length (SPL) may be a result of errors that can either be systematic or random. Hence, it becomes important to focus on the quality of measurements to prevent any iatrogenic harm to the patients. Objective: The objective of this study was to assess the magnitude of intra- and interobserver variations in the measurement of SPL with the SPLINT technique. Materials and Methods: SPL was measured prospectively in a cohort of 449 boys aged 0–14 years including 68 infants (substratified into Group I: >4 years, Group II: 4–8 years, and Group III: >8 years) with the SPLINT technique by expert (E: E1 and E2) and trainee (T: T1 and T2) surgeons after completing a three-tiered training module. Intra- and interobserver variability was assessed through descriptive statistics, intraclass correlation (ICC), relative technical error of measurement (rTEM), and reliability or R (%). Results: Intraobserver variability: the mean difference between the two readings (E1 and E2) is 0.08 cm (95% confidence interval [CI]: 0.073–0.087), ICC was 0.998 (95% CI: 0.997–0.998), and intraobserver variability ≤0.1 cm in 85% of the participants (n = 370 of 433). The rTEM and reliability (%) were 1.82% and 98.1% (Group I), 1.65% and 98.9% (Group II), and 1.09% and 99.7% (Group III), respectively. The intraobserver variability was observed to be inversely proportional to the age of the participants (correlation coefficient = −0.56). Interobserver variability was calculated separately for expert versus trainee and trainee versus trainee (T-vs-T) measurements. For expert versus expert, ICC, rTEM, and reliability (%) were 0.984, 2.4%, and 96.8% (Group 1), 0.992, 2.07%, and 98.3% (Group 2), and 0.997, 1.38%, and 99.05% (Group 3), respectively. A similar pattern of variability was observed for T-vs-T measurements. The reliability (%) of the SPL by experts is consistently more than that of trainees across all age groups; however, the difference ameliorates with the age of participant. Conclusions: The study has validated the SPLINT technique by demonstrating a high level of intra- and interobserver reliability. The adequacy of the training modules for SPL measurements described in this study has also been established. Evidence that the SPL can be used as an objective marker of penile dimensions is herewith furnished.

The role of colposcopy in HPV vaccination era.

Colposcopy has traditionally played a crucial role in the assessment and management of abnormal cervical cytology and human papillomavirus (HPV) -related lesions. However, with the reduction in high-risk (HR) HPV types due to vaccination, the role of colposcopy in the HPV vaccination era is being questioned. The aim of this narrative review was to highlight the latest evidence from the world of HPV vaccination and the future role of colposcopy in the HPV vaccination era. PubMed, MEDLINE, Scopus, Web of Science were searched for relevant articles up to June 2024. HPV vaccination may influence colposcopic practice by reducing HR-HPV vaccine type, colposcopy referrals, colposcopic positive predictive value for CIN 2+, and by changing referral patterns, colposcopic performance, and procedures. The relative incidence of HPV-negative adenocarcinoma and its precursors may be increasing, presenting a new challenge that may increase the difficulty of colposcopic assessment. The role of colposcopy in the management of cervical abnormalities will vary according to vaccination status, vaccine type and timing of vaccination. As this evolves, colposcopy will need to adapt to provide optimal care tailored to individual risk profiles and to maintain the key role in cervical cancer prevention that it has acquired over the past decades. There will certainly be a need to improve the quality and performance of colposcopy by reducing inter- and intra-observer variability in colposcopic practice, including more objective patient selection, greater knowledge, expertise and skill, with the aim of achieving reproducible, sensitive and less biased assessment of cervical appearance.

Breast cancer survival prediction using an automated mitosis detection pipeline.

Mitotic count (MC) is the most common measure to assess tumor proliferation in breast cancer patients and is highly predictive of patient outcomes. It is, however, subject to inter- and intraobserver variation and reproducibility challenges that may hamper its clinical utility. In past studies, artificial intelligence (AI)-supported MC has been shown to correlate well with traditional MC on glass slides. Considering the potential of AI to improve reproducibility of MC between pathologists, we undertook the next validation step by evaluating the prognostic value of a fully automatic method to detect and count mitoses on whole slide images using a deep learning model. The model was developed in the context of the Mitosis Domain Generalization Challenge 2021 (MIDOG21) grand challenge and was expanded by a novel automatic area selector method to find the optimal mitotic hotspot and calculate the MC per 2 mm2. We employed this method on a breast cancer cohort with long-term follow-up from the University Medical Centre Utrecht (N = 912) and compared predictive values for overall survival of AI-based MC and light-microscopic MC, previously assessed during routine diagnostics. The MIDOG21 model was prognostically comparable to the original MC from the pathology report in uni- and multivariate survival analysis. In conclusion, a fully automated MC AI algorithm was validated in a large cohort of breast cancer with regard to retained prognostic value compared with traditional light-microscopic MC.

Systematic Review of Nevus Counting and Reporting Methodologies in Contemporary Studies of the General Population

Introduction: Having many melanocytic nevi on the skin is a risk factor for melanoma. However, the reproducibility of nevus counts in previous studies is limited due to high inter- and intraobserver variation. Despite the introduction of a protocol for counting and reporting of nevi in 1990 by the International Agency for Research on Cancer (IARC), significant variations in nevus counting methods persist across studies. Objectives: We sought to review the variations in nevus counting and reporting methods, adherence and deviations from the IARC protocol, and the reproducibility of nevus counting studies. Methods: A systematic search of Embase, PubMed and Web of Science was conducted. The review was limited to nevus (>2 mm) counting studies of general population adults conducted between 2000 and 2022, and studies using skilled examiners. Results: Out of the 8 studies which were eligible for inclusion, none followed the IARC protocol. Three studies used a predefined criterion to count nevi. Five studies provided training for their observers. Three studies assessed the inter- or intraobserver variation using the correlation coefficient (>0.75), and 3 studies attempted to verify the validity and the reproducibility of the counts. There was little to no agreement in nevus counting and reporting procedures in the reviewed studies, and most studies did not report their procedures adequately. Conclusion: This review highlights the need for an easily accessible and feasible protocol for identification, counting and reporting of nevi, which also considers nevus counting from total-body imaging and automated nevus counts since these technologies are expected to become widely available for future studies.

Ejection fraction assessment and left ventricular contractile function: How accurate is this parameter?

Evaluation of the left ventricular (LV) ejection fraction (EF) is a key component for diagnosis and treatment of heart failure (HF). Echocardiography is commonly used to assess LV systolic function and LV EF due to its simplicity and availability. However, it has some limitations, including interand intra-observer variability, dependence on the quality of visualization, etc. HF trials showed that there are some differences in the HF therapy effectiveness depending on the value of LV EF. These results suggest that basic therapy for HF and reduced LVEF is also effective in HF and preserved LVEF if patients have impairment of not only diastolic but also systolic function. Thus, LVEF may not be a reliable parameter to guide treatment decisions andit is necessary to introduce additional markers into clinical practice to assess LV systolic function. The purpose of this review is to discuss the difficulties of assessing LV systolic function using echocardiography and the possibilities of using alternative diagnostic methods, in particular myocardial deformation (speckle tracking echocardiography).

ClinicAl verSus algorIthmic predictioN of Obstructive Coronary Artery Disease - the CASINO study

Abstract Background The estimation of pretest probability (PTP) is a key step when evaluating patients with suspected coronary artery disease (CAD). Current European guidelines recommend using an algorithm based on age, sex, and symptom typicality. However, many cardiologists do not regularly use this method, relying instead on overall clinical impression. The aim of this study was to compare clinical and algorithmic prediction of obstructive CAD in a set of symptomatic patients with suspected chronic coronary syndrome (CCS). Methods In this survey study, 10 cardiologists from were asked to estimate the probability of obstructive CAD (on a scale of 1-99% and without the aid of any score or algorithm) for 120 anonymized clinical vignettes of outpatients who underwent diagnostic workup for suspected CCS. The provided information included age, sex, symptom description, cardiovascular risk factors, and know comorbidities. Whenever non-invasive tests had been performed, physicians were also asked to estimate the probability of obstructive CAD after knowing that results. Obstructive CAD was defined as diameter stenosis ≥50% confirmed by invasive coronary angiography. Results Among the 120 included patients (62 women, mean age 61±10 years), symptoms consisted of chest pain in 104 and dyspnea in the remainder. The median PTP of obstructive CAD was 15% (IQR 10-25%) according to the ESC algorithm, and 30% (IQR 15-50%) according to cardiologists’ estimates (p<0.001). The observed prevalence of obstructive CAD was 20% (n=24), yielding suitable calibration for the ESC model (predicted vs. observed p=0.631), and significant overestimation for clinicians’ assessments (predicted vs. observed p=0.014) – Figure 1A. Despite using the same data, cardiologists’ PTP estimates varied widely within the same patients (mean inter-observer coefficient of variation 56%±22%). In a subset of 29 patients whose PTP was appraised twice by the same physicians, the mean absolute difference between assessments (intra-observer variability) was 22%±8%. The ESC algorithm showed fair discriminative power to identify patients with obstructive CAD (C-statistic 0.68, 95% CI 0.58-0.79, p=0.005), whereas the C-statistic for clinicians ranged from 0.53 to 0.72 (average 0.62±0.05, p values for comparisons 0.001-0.677) – Fig 1B. In the subset of 70 patients with prior non-invasive tests, clinicians significantly changed their predictions after knowing test results (mean absolute difference 22%±12%, p<0.001), improving their discriminative power significantly (mean change in C-statistic of 0.08±0.09, p=0.020). Conclusion Clinicians’ assessments of PTP in symptomatic patients display high inter- and intra-observer variability, and tend to overestimate the likelihood of obstructive CAD. Despite using more clinical data, cardiologists fail to outperform the ESC algorithm. The systematic use of this tool should be promoted in order to improve disease prediction and guide non-invasive testing.

Quantifying late gadolinium enhancement improved sudden cardiac death risk prediction in non-ischemic dilated cardiomyopathy

Abstract Background While previous studies have proved the late gadolinium enhancement (LGE) was associated with poor prognosis in non-ischemic dilated cardiomyopathy (DCM), the optimal method for LGE quantification and the ability of LGE extent in identifying sudden cardiac death (SCD) remain less clarified. Purpose This study sought to compare the reproducibility of LGE quantification methods and explore the asssociation between LGE extent and SCD endpoint in DCM patients. Methods In this prospective study, LGE was quantified by the 2-6 standard deviations (SD) above the remote myocardium and full-wide half-maximum (FWHM) methods. The primary endpoint was SCD and arrythmia endpoint included SCD and aborted SCD. Reproducibility of LGE quantification was measured by Bland-Altman analysis and intra-class correlation coefficients (ICC). Competing risk regression analysis, C-index and area under the curve (AUC) were performed to identify the prognostic value. Results Among the 770 patients, 336 (44%) patients had LGE. FWHM demonstrated the best reproducibility compared with other quantification methods (intraobserver variability: ICC = 0.94, mean bias: -0.43 ± 4.89%; interobserver variability: ICC = 0.90, mean bias: -2.65 ± 6.40%). After a median follow-up of 49 months, SCD occurred in 61 (8%) patients and arrythmia endpoint occurred in 87 (11%) patients. Among all quantification method, LGE extent measured by FWHM showed the highest predictive value of SCD (C index: 0.72) and arrythmia events (C index: 0.71) than other methods. In the competing risk analysis, LGE extent was independently association with SCD (Hazard ratio [HR] 1.05, 95% confidence interval [CI] 1.03 - 1.07, P < 0.001) and arrythmia endpoint (HR 1.05, 95% CI 1.03 - 1.06, P < 0.001). LGE extent >8% showed significantly higher risk of SCD and arrythmia endpoint than those with LGE extent ≤8% (P < 0.001). Incoporating LGE exent and 35% left ventricular ejection fraction (LVEF) significantly improved the SCD (C index: 0.74 vs 0.61, P < 0.001) and arrythmia endpoint (C index: 0.73 vs 0.59, P < 0.001) prediction compared with LVEF. Conclusions FWHM demonstrated best reproducibility and highest SCD prediction ability among LGE quantification methods. Adding LGE extent to LVEF significantly improved the predictive value of SCD and arrythmia endpoint.Figure 1

Fully automated artificial intelligence-based cine cardiac magnetic resonance image quantification

Abstract Background Analyzing cine cardiac magnetic resonance (CMR) images is a task that demands substantial experience and time, and it is subject to both inter- and intra-observer variability. Existing automated tools in the clinics for image analysis often necessitate substantial editing to enable accurate quantification. Objective To tackle this issue, in this study, we developed and validated a fully automated artificial intelligence (AI)-based system for the analysis of cine CMR images, aiming to address these challenges. Method In this study, 1,038 patients with CMR images were enrolled. Out of these, 350 CMR images, along with corresponding segmentations for the left ventricle myocardium (LVM) and the left- and right ventricles (LV and RV, respectively), sourced from open access data from 5 centers and databases (with different cardiac diseases), were utilized for model development. The remaining images from two local registries were used to evaluate the model. A 2D U-Net network with deep supervision, enhanced with various image augmentation techniques, was developed for automatic segmentation. The developed model was subsequently utilized to segment the test dataset of 4D (3D+time) short-axis cine CMR images. End-diastolic volumes (EDV) and end-systolic volumes (ESV) were automatically detected from these segmentations and evaluated against the ground truth calculations performed by two experienced physicians using commercially validated software. Results In the test set, Dice scores of 0.84, 0.90, and 0.93 were achieved for the RV, LV, and LVM, respectively. The mean percentage error for EDV was 0.6% for LV and -6.32% for RV. Regarding ESV, the mean percentage error was 2.95% for LV and -6.32% for RV. A -6.95% error was achieved for LV mass calculation. The stroke volumes for LV and RV were calculated with a mean percent error of -2.52% and -8.43%, respectively. We achieved a mean percent error of -3.22% for LV and -1.64% for RV ejection fraction. Conclusion The fully automated AI-based quantitative analysis of cine cardiac MR images developed in this study demonstrated high agreement with the ground truth across large datasets, including those from external centers. This method can significantly reduce the time required for image analysis and enhance the accuracy and consistency of cardiac quantitative assessments in clinical settings.

Repeatability and reproducibility of artificial intelligence-acquired fetal brain measurements (SonoCNS) in the second and third trimesters of pregnancy

Artificial Intelligence (AI)-based algorithms are increasingly entering clinical practice, aiding in the assessment of fetal anatomy and biometry. One such tool for evaluating the fetal head and central nervous system structures is SonoCNS™, which delineates appropriate planes for measuring head circumference (HC), biparietal diameter (BPD), occipitofrontal diameter (OFD), transcerebellar diameter (TCD), width of the posterior horn of the lateral ventricle (Vp), and cisterna magna (CM) based on a 3D volume acquired at the level of the fetal head’s thalamic plane. This study aimed to evaluate the intra- and interobserver variability of measurements obtained using this software. The study included 381 patients, 270 in their second trimester of pregnancy (70%) and 111 in the third trimester. Each patient underwent manual biometric measurements of the aforementioned structures and twice using the SonoCNS software. We calculated the intraobserver variability between the manual measurements and the average of the automated measurements, as well as the interobserver variability for automated measurements. We also compared the median examination time for manual and automated measurements. The interclass correlation coefficients (ICC) for interobserver and intraobserver variability for parameters BPD, HC, and OFD ranged from good to excellent reproducibility in the general population and subgroups (> 0.75). CM and Vp measurements, both in the general population and subgroups, fell into the category of moderate (0.5–0.75) and poor reproducibility (< 0.5). TCD measurements showed moderate (> 0.5) to good reproducibility (0.75–0.9), and OFD showed good and excellent reproducibility. The assessment of the biometry of fetal head structures using SonoCNS took an average of 63 s compared to 14 s for manual measurement (p < 0.001). The SonoCNS™ software is characterized by good to excellent reproducibility and repeatability in the measurement of fetal skull biometry (BPD, HC, and OFD), with poorer performance in measurements of intracranial structures (CM, Vp, TCD). Apart from biometric parameters, the software is useful in clinical practice for delineating appropriate planes from the acquired volume of the fetal head and shortening examination time.

On the accuracy of the segmentation process and transcatheter heart valve dimensions in TAVI patients

Accurate segmentation of medical images is critical for generating patient-specific models suitable for computational analyses, particularly in the context of transcatheter aortic valve implantation (TAVI). This study aimed to quantify the accuracy of the segmentation process from medical images of TAVI patients to understand the uncertainty in patient-specific geometries. We also quantified discrepancies between actual and CT-related diameter measurements due to artifacts and intra-observer variability. Segmentation accuracy was assessed using both synthetic phantom models and patient-specific data. The impact of voxelization and CT scanner resolution on segmentation accuracy was evaluated, while the intersection over union (IoU) metric was used to compare the consistency of different segmentation methodologies. The voxelization process introduced a marginal error (<1%) in phantom models relative to CAD models. CT scanner resolution impacted segmented model accuracy only after a 7.5-fold increase in voxel size compared to the baseline medical image. IoU analysis revealed higher segmentation accuracy for calcification (93.4 ± 3.1 %) compared to the aortic wall (85.4 ± 8.4 %) and native valve leaflets (75.5 ± 6.3 %). Discrepancies in THV diameter measurements highlighted a ∼5 % error due to metallic artifacts, with variability among observers and at different THV heights. Errors due to voxel size, segmentation methodologies and CT-related artifacts can impact the reliability of patient-specific geometries and ultimately computational predictions used to asses clinical outcomes and enhance decision-making. This study underscores the importance of accurate segmentation and its standardization for patient-specific modeling of TAVI simulations.

Efficient application of deep learning-based elective lymph node regions delineation for pelvic malignancies.

While there are established international consensuses on the delineation of pelvic lymph node regions (LNRs), significant inter- and intra-observer variabilities persist. Contouring these clinical target volumes for irradiation in pelvic malignancies is both time-consuming and labor-intensive. The purpose of this study was to develop a deep learning model of pelvic LNRs delineation for patients with pelvic cancers. Planning computed tomography (CT) studies of 160 patients with pelvic primary malignancies (including rectal, prostate, and cervical cancer) were retrospectively collected and divided into training set (n=120) and testing set (n=40). Six pelvic LNRs, including abdominal presacral, pelvic presacral, internal iliac nodes, external iliac nodes, obturator nodes, and inguinal nodes were delineated by two radiation oncologists as ground truth (Gt) contours. The cascaded multi-heads U-net (CMU-net) was constructed based on the Gt contours from training cohort, which was subsequently verified in the testing cohort. The automatic delineation of six LNRs (Auto) was evaluated using dice similarity coefficient (DSC), average surface distance (ASD), 95th percentile Hausdorff distance (HD95), and a 7-point scale score. In the testing set, the DSC of six pelvic LNRs by CMU-net model varied from 0.851 to 0.942, ASD from 0.381 to 1.037mm, and HD95 from 2.025 to 3.697mm. No significant differences were founded in these three parameters between postoperative and preoperative cases. 95.9% and 96.2% of auto delineations by CMU-net model got a score of 1-3 by two expert radiation oncologists, respectively, meaning only minor edits needed. The CMU-net was successfully developed for automated delineation of pelvic LNRs for pelvic malignancies radiotherapy with improved contouring efficiency and highly consistent, which might justify its implementation in radiotherapy work flow.

Precision HER2: a comprehensive AI system for accurate and consistent evaluation of HER2 expression in invasive breast Cancer

BackgroundWith the development of novel anti-HER2 targeted drugs, such as ADCs, it has become increasingly important to accurately interpret HER2 expression in breast cancer. Previous studies have demonstrated high intra-observer and inter-observer variabilities in evaluating HER2 staining by human eyes. There exists a strong requirement to develop artificial intelligence (AI) systems to achieve high-precision HER2 expression scoring for better clinical therapy.MethodsIn the present study, we collected breast cancer tissue samples and stained consecutive sections with anti-Calponin and anti-HER2 antibodies. High-quality digital images were selected from immunohistochemical slides and interpreted as HER2 3+, 2+, 1+, and 0. AI models were trained and assessed using annotated training and testing sets. The AI model was trained to automatically identify ductal carcinoma in situ (DCIS) by Calponin staining and myoepithelial annotation and filter out DCIS components in HER2-stained slides using image-overlapping techniques. Furthermore, we organized two-phase validation studies. In phase one, pathologists interpreted 112 HER2 whole-slide images (WSIs) without AI assistance, whereas in phase two, pathologists read the same slides using the AI system after a washing period of 2 weeks.ResultsOur AI model greatly improved the accuracy of reading (0.902 vs. 0.710). The number of HER2 1 + patients misdiagnosed as HER2 0 was significantly reduced (32/279 vs. 65/279), and they benefitted from ADC drugs. In addition, the AI algorithm improved the intra-group consistency of HER2 readings by pathologists with different years of experience (intra-class correlation coefficient [ICC]: 0.872–0.926 vs. 0.818–0.908), with the improvement most pronounced among junior pathologists (0.885 vs. 0.818).ConclusionsWe proposed a high-precision AI system to identify and filter out DCIS components and automatically evaluate HER2 expression in invasive breast cancer.

Quantification of Mitral Valve Regurgitation in Cavalier King Charles Spaniels and Chihuahuas Using Radius of Proximal Isovelocity Surface Area

Mitral regurgitation (MR) resulting from myxomatous mitral valve disease (MMVD) is a prevalent condition in dogs, particularly smaller breeds like Cavalier King Charles Spaniels (CKCSs) and Chihuahuas (CHHs). An accurate assessment of MR severity is essential for effective treatment and disease monitoring, yet a standardized method has yet to be established. In this retrospective study, we evaluated 124 client-owned dogs diagnosed with MMVD, including 64 CKCSs and 60 CHHs. Dogs were categorized into three stages: asymptomatic (B1), remodeled (B2), and congestive heart failure (CHF, C). The MR severity was quantified using the proximal isovelocity surface area (PISA) method, specifically focusing on the PISA radius (PISA-r). The PISA-r measurements exhibited significant increases across disease stages and demonstrated strong correlations with echocardiographic parameters (ranging from 0.83 to 0.94), including the left atrial size and left ventricular internal diameter. The receiver operating characteristic (ROC) curve analysis revealed the high sensitivity and specificity of PISA-r in classifying disease stages, establishing optimal cut-off values. The method displayed excellent repeatability (interobserver variability: 0.95) and reproducibility (intraobserver variability: 0.97). In conclusion, the PISA method, specifically PISA-r, was reliable for assessing MR severity in dogs with MMVD. By simplifying the diagnosis and management of MR, this research can potentially improve the life and management of MMVD-affected dogs.

Observed and expected reliability of echocardiographic volumetric methods and critical change values for quantification of mitral regurgitant fraction in dogs.

Reliability of echocardiographic calculations for stroke volume and mitral regurgitant fraction (RFMR) are affected by observer variability and lack of a gold standard. Variability is used to calculate critical change values (CCVs) that are thresholds representing real change in a measure not associated with observer variability. Observed intra- and interobserver accuracy and variability in healthy dogs help model CCV for RFMR. Reliability cohort of 34 healthy dogs; allometric scaling cohort of 99 dogs with heart disease and 25 healthy dogs. Accuracy, variability, and CCV of 2 observers using geometric and flow-based echocardiography were prospectively compared against a standard of RFMR = 0% and extrapolated across a range of expected RFMR values in the reliability cohort partly derived from cardiac dimensions predicted by the allometric cohort. Accuracy of methods to determine RFMR in descending order was 4-chamber bullet (Bullet4CH), mitral inflow, cube formula, and Simpson's method of disks. Intraobserver variability was relatively high. The CCV for RFMR ranged from 28% to 88% and was inversely related to RFMR when extrapolated for use in affected dogs. For both observers, the Bullet4CH method had the lowest intraobserver CCV (Operator 1:28%, Operator 2:41%). Interobserver strength of agreement was low with intraclass correlation coefficients ranging from 0.210 to 0.413. Echocardiographic volumetric methods used to calculate stroke volume and RFMR have low accuracy and high variability in healthy dogs. Extrapolation of observed CCV to a range of expected RFMR suggests observers and methods are not interchangeable and variability might hinder routine clinical usage. Individual observers should be aware of their own variability and CCV.

Comparison of Visual and Quantra Software Mammographic Density Assessment According to BI-RADS® in 2D and 3D Images.

Mammographic density (MD) assessment is subject to inter- and intra-observer variability. An automated method, such as Quantra software, could be a useful tool for an objective and reproducible MD assessment. Our purpose was to evaluate the performance of Quantra software in assessing MD, according to BI-RADS® Atlas Fifth Edition recommendations, verifying the degree of agreement with the gold standard, given by the consensus of two breast radiologists. A total of 5009 screening examinations were evaluated by two radiologists and analysed by Quantra software to assess MD. The agreement between the three assigned values was expressed as intraclass correlation coefficients (ICCs). The agreement between the software and the two readers (R1 and R2) was moderate with ICC values of 0.725 and 0.713, respectively. A better agreement was demonstrated between the software's assessment and the average score of the values assigned by the two radiologists, with an index of 0.793, which reflects a good correlation. Quantra software appears a promising tool in supporting radiologists in the MD assessment and could be part of a personalised screening protocol soon. However, some fine-tuning is needed to improve its accuracy, reduce its tendency to overestimate, and ensure it excludes high-density structures from its assessment.

Impact of motion management strategies on abdominal organ at risk delineation for magnetic resonance-guided radiotherapy

Background and purposeThe impact of respiratory motion management strategies for abdominal radiotherapy, such as abdominal compression (AC) and breath hold (BH), on abdominal organ at risk (OAR) delineation on magnetic resonance imaging (MRI) is unknown. This feasibility study compared the inter- and intra- observer delineation variation on MRI acquired with AC, BH for three critical abdominal OAR. Materials and methodsT2-weighted (W) 3D MRI in free-breathing (FB) and with AC, and T1W 3D mDixon exhale BH were acquired. Four observers blinded to motion management strategy used, delineated stomach, liver, and duodenum on all MRI. One case per strategy was repeated over 6 weeks later to quantify intra-observer variation. Simultaneous truth and performance level estimation (STAPLE) contours for each OAR were generated, median and IQR mean distance to agreement (mDTA) and maximum Hausdorff distance (HD) between observer and STAPLE contours were calculated. Observers scored organ visibility on each MRI using a four-point Likert scale. ResultsA total of 27 scans including repeats were delineated. Pooled mDTA for all OARs was 1.3 mm (0.5 mm) with AC, 1.4 mm (1.0 mm) with BH, and 1.3 mm (0.5 mm) in FB. Intra-observer mDTA was highest for all organs in FB with 10.8 mm for duodenum, 1.8 mm for liver, and 2.7 mm for stomach. The pooled mean perceptual quality score value was highest for AC across organs. ConclusionsNo motion management strategy demonstrated superior similarity across OAR, emphasizing the need for personalised approaches based on individual clinical and patient factors.

Assessment of adult renal cell carcinoma types presenting to a tertiary care hospital in Sri Lanka and the intraobserver variability of two nuclear grading systems for clear cell renal cell carcinoma

Assessment of adult renal cell carcinoma types presenting to a tertiary care hospital in Sri Lanka and the intraobserver variability of two nuclear grading systems for clear cell renal cell carcinoma