Limits Of Agreement Research Articles

Lymph node metastasis prediction from in-situ lung squamous cell carcinoma histopathology images using deep learning.

Lung squamous cell carcinoma (LUSC), a subtype of non-small cell lung cancer, represents a significant portion of lung cancer cases with distinct histologic patterns impacting prognosis and treatment. The current pathological assessment methods face limitations such as inter-observer variability, necessitating more reliable techniques. This study seeks to predict lymph node metastasis in LUSC using deep learning models applied to histopathology images of primary tumors, offering a more accurate and objective method for diagnosis and prognosis.Whole slide images (WSIs) from the Outdo-LUSC and TCGA-LUSC cohorts were used to train and validate deep-learning models. Multi-instance learning was applied, with patch-level predictions aggregated into WSI-level outcomes. The study employed the ResNet-18 network, transfer learning, and rigorous data preprocessing. To represent WSI features, innovative techniques like patch likelihood histogram (PLH) and bag of words (BoW) were used, followed by training of machine learning classifiers, including the ExtraTrees algorithm.The diagnostic model for lymph node metastasis showed strong performance, particularly using the ExtraTrees algorithm, as demonstrated by receiver operating characteristic (ROC) curves and Grad-CAM visualizations. The signature generated by the ExtraTrees algorithm, named LN_ISLUSCH (lymph node status-related in-situ lung squamous cell carcinoma histopathology), achieved an area under the curve (AUC) of 0.941 (95% CI: 0.926-0.955) in the training set and 0.788 (95% CI: 0.748-0.827) in the test set. Kaplan–Meier analyses confirmed that the LN_ISLUSCH model was a significant prognostic factor (p = 0.02).This study underscores the potential of artificial intelligence in enhancing diagnostic precision in pathology. The LN_ISLUSCH model stands out as a promising tool for predicting lymph node metastasis and prognosis in LUSC. Future studies should focus on larger and more diverse cohorts and explore the integration of additional omics data to further refine predictive accuracy and clinical utility.

Automatic motion correction for myocardial blood flow estimation improves diagnostic performance for coronary artery disease in 18F-flurpiridaz PET-MPI

BackgroundMotion correction (MC) is critical for accurate quantification of myocardial blood flow (MBF) and flow reserve (MFR) from 18F-flurpiridaz PET myocardial perfusion imaging (MPI). However, manual correction is time consuming and introduces inter-observer variability. We aimed to validate an automatic MC algorithm for 18F-flurpiridaz PET-MPI in terms of diagnostic performance for predicting coronary artery disease (CAD). MethodsIn total, 231 patients who underwent invasive coronary angiography and rest/pharmacologic stress 18F-flurpiridaz PET-MPI from phase III Flurpiridaz trial (NCT01347710) were enrolled. For manual MC, two operators (Reader 1 and Reader 2) shifted each frame's images in three directions. The automatic MC algorithm, initially developed for 82Rb-chloride PET-MPI, was optimized for 18F-flurpiridaz. Diagnostic performance was compared using minimal segmental MBF/MFR with and without MC to predict CAD ≥70% stenosis by angiography. ResultsManual MC took 10 minutes per case (both stress and rest) on average, while automatic MC required <17 seconds. The area under the receiver operating characteristic curves (AUCs) for significant CAD using minimal segmental MBF were comparable between automatic and manual MC (AUC=0.877 automatic, AUC=0.888 Reader 1 and AUC=0.892 Reader 2; all p>0.05). AUCs of minimal segmental MBF with manual and automatic MC were significantly higher than without MC (p<0.05 for both). Similar findings were observed with minimal segmental MFR. ConclusionsAutomatic MC can be performed rapidly, with diagnostic performance for predicting obstructive CAD comparable to manual MC. This method could be utilized for analysis of MBF/MFR in patients undergoing 18F-flurpiridaz PET-MPI.

Harnessing Natural Language Processing and High-Dimensional Clinical Notes to Detect Goals-of-Care and Surrogate-Designation Conversations.

Advance care planning, involving goals-of-care and surrogate-designation conversations, is crucial for patient-centered care. However, determining the optimal timing and participants for these conversations remains challenging. This study explored the frequency, timing, and predictors of documenting two advance care planning elements, goals-of-care and surrogate-designation conversations, in clinical notes for patients with advanced illness. In this retrospective observational study, we leveraged high-dimensional data and natural language processing (NLP) to analyze clinical notes and predict the presence or absence of advance care planning conversations. We included notes for patients treated at a Midwestern United States hospital who had advanced chronic conditions and eventually passed away. We manually labeled a gold-standard dataset (n = 913 notes) for the presence or absence of advance care planning conversations at the note level, achieving excellent inter-annotator agreement (90.5%). Training and testing four NLP models to detect goals-of-care and surrogate-designation conversations revealed that a transformer-based model (Bidirectional Encoder Representations from Transformers [BERT]) achieved the highest accuracy, with an F1 score of 93.6. We then deployed the BERT model to a high-dimensional corpus of 247,241 notes for 4,341 patients and detected goals-of-care and surrogate-designation conversations in the records of 85% and 60% of patients, respectively. Temporal analysis revealed that goals-of-care and surrogate-designation conversations were first documented at medians 28 and 8 days before death, respectively. Patient characteristics and referral to specialty palliative care emerged as significant factors associated with documenting these conversations. Our findings demonstrate the potential of NLP, particularly Transformer-based models like BERT, to accurately detect goals-of-care and surrogate-designation conversations in clinical narratives. This study identified significant temporal patterns, including late documentation, and patient characteristics associated with these conversations. It highlights the value of high-dimensional data in enhancing our understanding of advance care planning and offers insights for improving patient-centered care in clinical settings. Future research should explore the integration of these models into clinical workflows to facilitate timely and effective advance care planning discussions.

Cardiovascular magnetic resonance semi-automated threshold-based post-processing of right ventricular volumes in repaired tetralogy of Fallot.

Cardiovascular magnetic resonance (CMR) is the gold-standard to estimate right ventricular (RV) volumes, which are key for clinical management of patients with repaired tetralogy of Fallot (rTOF). Semi-automated threshold-based methods (SAT) have been proposed for CMR post-processing as alternatives to fully manual standard tracing. We investigated the impact of SAT on RV analysis using different thresholds in rTOF patients. RV volumes and mass were estimated using SAT and standard fully manual tracing methods in rTOF patients. Two threshold levels were set for SAT, i.e., default 50 (SAT-50) and 30 (SAT-30). RV stroke volumes (SV) were compared to main pulmonary artery forward flow (MPA-FF). Post-processing time, intra- and interobserver variabilities were compared across methods. Sixty-two CMRs of rTOF patients were analyzed. Compared to the standard fully manual tracing, no significant differences in RV mass, volumes and ejection fraction were observed using SAT-30, whereas SAT-50 significantly underestimated RV end-diastolic-volume index (EDVi) by 10.4% (mean difference of - 11.8 ± 6.2ml/m2, p 0.03) and overestimated RV mass index by 21.8% (mean difference of 14.2 ± 11.9g/m2, p 0.002). Compared to MPA-FF, RVSV by standardfully manual method and SAT-30 showed minor biases, respectively, 0.03ml/m2 and 0.7ml/m2, while SAT-50 underestimated RVSV by 6.86ml/m2 (p < 0.001). In six patients, the degree of RV EDVi underestimation by SAT-50 determined a change of category from dilated to non-dilated RV. Intra- and interobserver variabilities were good to excellent for all methods. Post-processing duration was shorter for SAT compared to standardmanual segmentation (5.5 ± 1.7min vs. 19.5 ± 4.4min, p < 0.001). CMR SAT-30 post-processing is a precise, accurate and time-saving method for biventricular assessment of volumes, ejection fraction and mass in rTOF.

Inter-observer variability in assessing image-defined risk factors: implications for risk stratification in locoregional abdominopelvic neuroblastoma.

Risk stratification for locoregional neuroblastoma partially relies on image-defined risk factors (IDRFs). This study aimed to evaluate how inter-observer variability in assessing IDRFs impacts risk stratification in locoregional abdominopelvic neuroblastoma. A retrospective analysis was conducted on 123 patients who underwent upfront contrast-enhanced CT scans. Two radiologists independently assessed the presence of IDRFs. Patients were staged as either L1 (IDRF-negative) or L2 (IDRF-positive) according to the International Neuroblastoma Risk Group Staging System. Based on the radiologists' evaluations, 97 cases with sufficient clinical data were classified into risk groups using the revised Children's Oncology Group neuroblastoma risk classifier. The kappa values and 95% confidence intervals (CIs) were calculated to assess inter-radiologist agreement on IDRF evaluation and risk stratification. There was low agreement between radiologists in assessing L1/L2 status with a kappa value of 0.28 (95% CI: 0.14-0.42). However, agreement for evaluating the number of IDRFs was good, with an intraclass correlation coefficient of 0.73 (95% CI: 0.64-0.80). Based on the first radiologist's evaluation, 13 patients were classified as low-risk, 52 as intermediate-risk, and 32 as high-risk. Based on the second radiologist's evaluation, 37 patients were classified as low-risk, 37 as intermediate-risk, and 23 as high-risk. The kappa value for risk stratification between the two radiologists was 0.47 (95% CI: 0.33-0.62). Inter-observer variability in assessing IDRF presence may affect risk stratification in locoregional abdominopelvic neuroblastoma.

Long-Term Recurrence Risk, Metastatic Potential, and Length of Cystoscopic Surveillance of Low-Grade Non-Muscle-Invasive Bladder Cancer.

Patients with Ta low-grade (LG) NMIBC rarely develop metastases or die of it. Long-term data are scant and length of follow-up poorly defined. This retrospective study included 521 patients diagnosed with primary TaLG NMIBC (n = 491) or papillary urothelial neoplasm of low malignant potential (n = 30) from 1989 to 2019 at an academic center. Patient data were acquired using patient records chart review and a bladder cancer informatics registry at the center. Risk of recurrence and progression in stage, to muscle invasion, metastases, and death due to BC were analyzed. RNAseq assessed the transcriptomic profiles of 4 TaLG NMIBCs that metastasized. Interobserver variability in pathological grading (WHO 2004/2022 and 1973, n = 80) was blindly assessed by 3 expert pathologists. The median follow-up was 9.6 (95% CI: 8.6-10.2) years. Among 521 patients (73% men, median age 67.0 years), 350 recurred, 57 progressed in stage, 20 developed metastases, and 15 died of BC (median of 9.6 years after diagnosis). Cancer-specific survival probabilities were 0.99, 0.98, and 0.96 at 5, 10, and 15 years, respectively. Fifty patients who were recurrence free for the first 5 years developed late recurrences and 2 of them died of BC. Metastatic TaLG NMIBC had more adverse transcriptomic findings in keeping with higher grade tumors despite being phenotypically similar to indolent tumors. Grading concordance for the 2004/2022 system and WHO 1973 was 0.78 (95% CI: 0.65-0.90) and 0.41 (95% CI: 0.32-0.50), respectively. This study with long-term data challenges the assumption that primary TaLG NMIBC nearly never progresses to lethal disease if followed long enough. However, the risk of BC-related mortality is extremely low in patients who are recurrence-free for the first 5 years. Minimizing variability in pathological grading remains an unmet need.

Multi-vendor validation of automated left ventricular analysis from 3D echocardiography using artificial intelligence

Abstract Background Accurate echocardiographic assessment of left ventricular (LV) function is dependent on both image acquisition and analysis. Although 3D echocardiography (3DE) is increasingly recommended for volume quantification, manually derived volumes are limited by operator subjectivity, variable image quality, and the lengthy duration of analysis. Consequently, several automated approaches based on artificial intelligence (AI) have been proposed to segment the LV cavity from 3DE and have demonstrated better accuracy and reproducibility compared to human experts. While the utility of AI models in the clinic depends on their ability to generalise to different populations and ultrasound vendors, the external validation of developed models remains limited. Purpose This study evaluates the ability of an existing AI model trained using data from a single vendor to accurately estimate routine 3DE LV indices in an independent external dataset acquired using equipment from a different vendor. Methods A previously published and internally validated deep learning model for segmentation of the LV cavity and myocardium from 3DE was used to automatically derive LV indices including end-diastolic volume (EDV), end-systolic volume (ESV), LV mass, and ejection fraction (EF) [1]. A second independent dataset consisting of paired 3DE and CMR images from 65 marathon runners from a different centre (using equipment from a different vendor) was subsequently used for external validation. Images from both datasets were subjectively scored from 1 (poor) to 5 (excellent) as a measure of image quality. Agreement with CMR as the reference was assessed using a two-way mixed effects intraclass correlation coefficient (ICC), supplemented by Bland-Altman analysis, to determine the model bias (between 3DE and CMR) and 95% limits of agreement (LOA). To assess whether performance could be improved by inclusion of multi-vendor data during training, the initial model was fine-tuned using a small subset (n=9) of images from the second centre. Results The mean image quality score from the second dataset was 2.5 (± 1.1), which was significantly lower than the mean of 3.4 (± 1.0) from the first dataset. Despite vendor and image quality differences, the initial model produced good agreement with CMR in EDV, ESV, and LV mass but only poor agreement in EF (Fig 1). Fine tuning greatly improved agreement in EF with reduced bias and narrower 95% LOA (Fig 2). Conclusion With the addition of a small number of cases for fine-tuning, AI models for automated LV analysis can achieve good performance analysing 3DE images acquired from a different vendor and population. Advancing model generalisability through more sophisticated data augmentation strategies may enable truly vendor-agnostic automated LV analysis from 3DE.

Interscan reproducibility of computed tomography derived coronary plaque volume measurements using a semi-automated analysis software

Abstract Background Coronary computed tomography angiography (CCTA) enables detailed quantification and characterization of coronary atherosclerotic plaque, offering diagnostic and prognostic value. Despite the growing demand for accurate plaque analysis, studies on interscan reproducibility of automated plaque quantification are limited in number and size. Moreover, impact of clinical and technical factors on the reproducibility of plaque assessment tools are scarce. Purpose The objective of this study was to assess the interscan reproducibility of CCTA-derived plaque volume measurements and the implications of clinical and technical characteristics on interscan reproducibility. Methods A total of 101 patients with known coronary artery disease underwent two CCTA scans within hour interval using identical CT acquisition protocol. Coronary plaque volume measurements were quantified on a per-patient level using a contemporary semi-automated plaque analysis software. Results There was no statistically significant difference in median plaque volumes between the first and second scan. The direct correlation was excellent for volumes of non-calcified plaque (NCP), calcified plaque (CP), and total plaque (TP) across all analyses (Pearson’s coefficient 0.96 to 0.99), while moderate for low-density non-calcified plaque (LD-NCP) volumes (Pearson’s coefficient 0.77). Bland Altman analyses demonstrated high interscan agreement with narrow absolute and relative limits of agreement (LoA) across plaque subtypes, except for LD-NCP volumes (Table 1). Interscan reproducibility on CP volumes was affected by CT image quality with narrower LoA in scans with the highest image quality score (p = 0.003), or lowest image reconstructive iteration level (p &amp;lt;0.001). Limits of agreement were significantly narrower for NCP, CP, and TP volumes in lesions located in LAD compared to non-LAD lesions (p &amp;lt;0.001). However, no significant difference in LoA was present when comparing patients according to BMI (&amp;lt;27 and ≥27 kg/m2), heart rate (&amp;lt;65 and ≥65 beats per minute), or Agatston score (&amp;lt;193 and ≥193). Conclusions This comprehensive interscan reproducibility study revealed good agreement in the measurement of plaque subtypes. Adding new insight to existing literature, we have demonstrated that factors such as image quality, image reconstructive algorithms, and lesion location affect the reproducibility of plaque volume measurements. CT-derived plaque quantification using a semi-automated plaque analysis software holds potential for monitoring disease progression or regression in patients with CAD, provided high CCTA image quality.

Good agreement between cardiac magnetic imaging protocols using free-breathing and breath-holding techniques for flow quantification in valvular heart disease

Abstract Background Cardiac magnetic resonance (CMR) evaluation of valvular heart disease is an important diagnostic tool when echocardiography is inconclusive. Flow quantification is usually performed during breath hold, which can be challenging in valvular heart disease patients often suffering from dyspnea and heart failure. Purpose The purpose of the present study is to evaluate a free-breathing protocol of flow measurement in the aortic, pulmonary and tricuspid valves in healthy subjects (HS) and patients with tricuspid regurgitation (TR). Methods Phase contrast flow volume measurements were performed in 20 HS and 25 patients with TR using standard breath-holding (BH) and a free-breathing (FB) technique. Two venc-adjusted volume acquisition were acquired in the aortic, pulmonary and tricuspid valve during BH and one during FB. Mean, standard deviation (SD) and limits of agreement (LoA) for aortic forward flow volume (AoFF), pulmonary forward flow volume (PuFF) and tricuspid inflow volume (TrIF) were evaluated. Results Figure 1 presents flow volume (ml) of AoFF, PuFF and TrIF during BH (mean of 2 measurements) and FB. Conclusions The present study demonstrated good agreements between phase contrast flow measurements using FB and BH techniques. These parameters are used in the evaluation of aortic, mitral, pulmonary and tricuspid regurgitations using CMR imaging. Thus, free-breathing CMR acquisition can be an important complement in the assessment of valvular heart disease and may be applied in patients with difficulty to hold their breath.

Global longitudinal strain measurement in the modern era - is artificial intelligence the answer ?

Abstract Background Global longitudinal strain (GLS) enhances the assessment of left ventricular (LV) function beyond ejection fraction (EF) in 2D echocardiography (Echo). However, manual GLS measurement, the most common technique, is operator dependent. A novel vendor-independent artificial intelligence (AI)-based software streamlines automated 2D-Echo exam evaluation and GLS calculation. Purpose We sought to evaluate the ability of the AI to automatically identify LV 4-, 2-, and 3-chamber views (CV) from 2D-Echo exams and to validate GLS against the conventional manual method. Methods 540 patients underwent standard 2D-Echo exams. The server-based AI recognized the optimal LV CVs from 2D-Echo exams and automatically calculated triplane GLS by endocardial delineation (n=489, 91%). Manual LV GLS was performed independently by accredited cardiologists. Pearson’s correlation (R) and Bland-Altman analysis were performed to assess bias and limits of agreement (LOA) for GLS. Multivariate between-method regression was performed using ordinary least squares to disclose possible sources of possible bias. Significance was defined as a 2-tailed P value &amp;lt; 0.05. Results Patients’ median age was 70 years [18-93], 59% were males. The 2D-Echo indications were coronary artery disease (36%), valvular heart disease (15%), non-ischemic heart failure (15%) and others (34%). The feasibility of the AI to calculate GLS was 91% (n=489) due to endocardial border delineation issues. AI correctly classified all CVs. An excellent correlation (r=0.92, p&amp;lt;0.001) was observed between AI and manual GLS, with low bias (0.5%) and narrow LOAs (-3.1 and +4.2%) (Figure 1, top row). We identified LV end-diastolic volume and EF as statistically significant confounders (p &amp;lt; 0.05) with only small effect as depicted in Figure 1 (bottom row). Neither sex, age, acoustic window, body mass index, nor heart rate, along with the interaction with sex, end-diastolic volume, or EF, were significant (p &amp;gt; 0.23). Conclusion AI-based LV GLS calculation in 2D-Echo shows promising feasibility and accuracy. Strong agreement between AI-derived GLS and manual measurements supports its reliability. The AI's ability to identify optimal LV views underscores its robustness. Further validation and refinement are warranted, highlighting the transformative potential of AI in advancing cardiovascular diagnostics.

Automated non-invasive quantification of myocardial work using 3D echocardiography and validation with invasive catheter measurements

Abstract Background Myocardial work is an emerging technique for the assessment of left ventricular (LV) performance. Using blood pressure as an afterload determinate offers further insight into LV mechanics than can be detected with load-sensitive parameters of ejection fraction or global longitudinal strain (GLS). Observational studies have signalled benefit in identifying vulnerable ventricles in cardiomyopathy, valvular, and coronary disease. While the reference standard for calculating myocardial work is through catheterisation, this is impractical for routine assessment. Currently, a vendor specific non-invasive index of myocardial work is commercially available, but requires subjective manual analysis and identification of valvular events. With increasing availability of 3D echocardiography (3DE) and automated analysis techniques, there is potential to derive more reproducible estimates of myocardial work. Purpose We sought to test whether global myocardial work could be accurately, non-invasively, and automatically estimated using 3DE, in combination with standard non-invasive clinical measurements. Methods Patients undergoing left heart catheterisation as part of routine care were prospectively recruited, and 3DE was performed within one hour of invasive LV pressure measurement. A validated deep learning model was used to automatically analyse the 3DE image volumes to produce GLS curves, from which a smaller number of features (principal components) were extracted. The invasive pressure traces were separated into cycles, which were each paired with the strain trace that had the minimal difference in R-R interval (and excluded if the relative R-R interval difference exceeded 25%). For each patient, an invasively-derived estimate of myocardial work was calculated as the area enclosed by the median pressure-strain (PS) loop. A multivariate regression model was used to relate a set of routine clinical parameters to the invasively-derived myocardial work, and subsequently assessed using leave-one-out cross-validation. Results A total of 131 out of 153 patients were included for analysis (2 excluded due to R-R interval mismatch, 20 excluded due to poor image or pressure trace quality), with basic demographics presented in Table I. The final parameter set consisted of 6 variables: systolic and diastolic brachial cuff pressures, peak GLS, R-R interval, and two strain curve principal components. When fit into a leave-one-out model, this resulted in a Pearson r-value of 0.86, demonstrating a very high correlation between predicted and invasively-derived myocardial work estimates, with a bias of 0 mmHg% and 95% limits of agreement (LOA) of ± 423 mmHg% (Figure 1). Conclusion This study demonstrates that fully automated and reliable estimates of myocardial work can be derived using 3DE and deep learning combined with routine clinical parameters. Automated and non-invasive approaches may facilitate increased clinical uptake of myocardial work assessment.

A 2D echocardiographic method for accurate right ventricular quantification using deep learning

Abstract Background/Introduction Transthoracic two-dimensional echocardiography (2DE) remains the backbone for non-invasive assessment of the right ventricle (RV) size and function. However, calculation of RV end-diastolic (RVEDV), end-systolic volume (RVESV) and ejection fraction (RVEF) cannot be performed using 2DE due to the complex RV shape (1), whereas currently used, respective 2DE indices lack diagnostic accuracy. Deep learning (DL) may be able to assist in developing new 2DE methods for RV quantification. Purpose To develop a novel DL-based 2DE method for quantitative evaluation of RV size and function without geometric assumptions. Methods 50 adult patients underwent 2DE and CMR within 30 days as part of routine cardiac care between 6/2020 – 9/2021 at a medical center. We excluded patients with prior cardiac surgery of the RV, tricuspid or pulmonic valve repair or replacement, complex congenital heart disease, more than small pericardial effusion, cardiac tamponade, atrial fibrillation at the time of either study, severe RV dysfunction or inadequate imaging quality. End-diastolic and end-systolic areas derived from planimetry of 8 2DE RV views, along with patient age and gender, were to used to train and test a DL algorithm, namely the Feature-Tokenizer and Transformer (Figure 1) in order to predict the corresponding, CMR-derived, RVEDV or RVESV, and subsequently calculate RVEF. We used 5-fold cross-validation for training and evaluation. We assessed the relative importance of each 2DE view using gain metric-powered explainability analysis and subsequently performed a sensitivity analysis to identify optimal combinations of views. Variability analysis was performed for a random subcohort of 10 patients. Results Median age was 51 (interquartile range 32-62) and 42% were women. Mean RVEDV, RVESV and RVEF) by CMR were 163±70ml, 82±42ml and 51±8% respectively. Six patients (12%) had RV dilatation and 9 (18%) had RV dysfunction when using published CMR criteria. Explainability analysis showed that the 3 most important views were parasternal long axis (PLAX) (relative feature importance [RFI] = 0.197), 4-chamber (RFI=0.145) and parasternal short axis at the base (PSAX-base) (RFI=0.112). The 3-view combination of PLAX, 4-chamber and PSAX-base achieved high accuracy when compared to CMR (R2=0.964, absolute percentage error [APE]=8.09±12.07% for RV volumes and APE=9.86±10.68% for RVEF). Good (intraclass correlation coefficient [ICC]&amp;gt;0.75) to excellent (ICC&amp;gt;0.90) intra- and interobserver variability was found for these 3 views. Conclusions We propose an innovative, clinically applicable DL-based 2DE method for quantification of RV size and function that requires manual RV planimetry of 3 views, namely 4-chamber, PLAX and PSAX-base (Figure 2). This method may revolutionize 2DE RV assessment. Further studies with largest number of patients and wider range of RV size and function are needed to validate this method.Deep learning architectureGraphical abstract

Left atrial strain parameters are able to predict presence of atrial fibrillation

Abstract Background Three-dimensional speckle-tracking echocardiography allows evaluation of left atrial function, left atrial function has been assessed in various diseases for prognostication [1]. Atrial fibrillation (AF) is most common arrhythmia, and one of its manifestations is increased left atrial filling pressure and left atrial dilatation. Methods Echocardiography loops were collected between December 2019 and December 2021, only valid atrial loops, and sinus rhythm during examination. Spearman test was used for correlation matrices of left atrial parameters; uni- , multivariate and binary logistic regression to predict presence or absence of atrial fibrillation. Results A total of 116 patients (50% females) aged 76.9±11.2 were included in the analysis. Males and females differed in anthropometric and other clinical parameters. Intra- and interobserver variability was close to 1% since most of the measurements are semi-automatized. 2D and 3D left atrial volumes correlated significantly (r=0.8, P&amp;lt;0.0001, Fig.1). Almost all left atrial strain measurement uni- and multivariately predicted the presence of atrial fibrillation, binary logistic regression identified two most significant parameters (left atrial longitudinal strain reservoir and conduit phase, LASr, LAScd, P=0.042, P=0.044, respectively, Fig.2). LASr is significantly greater in patients without history of AF vs patient with AF history, (16±16, 10.5±7.1, respectively, P=0.012). The probability of sinus rhythm increases linearly with the increasing value of LASr. Conclusion Left atrial longitudinal strain was able to differentiate patients with paroxysmal atrial fibrillation as compared to patients without history of this arrhythmia.

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 &amp;lt; 0.001) and arrythmia endpoint (HR 1.05, 95% CI 1.03 - 1.06, P &amp;lt; 0.001). LGE extent &amp;gt;8% showed significantly higher risk of SCD and arrythmia endpoint than those with LGE extent ≤8% (P &amp;lt; 0.001). Incoporating LGE exent and 35% left ventricular ejection fraction (LVEF) significantly improved the SCD (C index: 0.74 vs 0.61, P &amp;lt; 0.001) and arrythmia endpoint (C index: 0.73 vs 0.59, P &amp;lt; 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

Right chambers deformation after tricuspid edge-to-edge percutaneous repair

Abstract Background The development of new analysis software has allowed to broaden myocardial deformation study to right chambers. There are already reference values for right ventricle (RV) and more recently for right atrium (RA). Tricuspid edge-to-edge repair (TEER) potentially causes changes in right chambers mechanics. Purpose Our aim is to evaluate the impact of TEER in the myocardial deformation of RV and RA. Methods Observational retrospective study including all the consecutive patients undergoing TEER in our tertiary centre. Additionally, a control group of healthy volunteers was included. Baseline and 3 months postprocedural echocardiograms were analyzed with a dedicated software to obtain RV, free wall (FW-) and 4 chambers (4C-) global longitudinal strain (GLS), and RA, reservoir (SrRA), conduit (ScdRA) and centrality (SctRA) strains , deformation values. Interobserver variability was evaluated in the control group with the intraclass correlation coefficient (ICC). Differences in baseline values between cases and control group were evaluated with Mann-Whitney U test. Evolutive changes in deformation parameters were studied using paired Wilcoxon W test. Results 15 healthy controls and 29 patients undergoing TEER (79.3% female, 79 [75-82] years-old) were included. Baseline tricuspid regurgitation (TR) was severe in 21 (72.4%), whereas the others were massive or torrential, and the most frequent mechanism was functional (22, 75,9%). Interobserver agreement in the control group was excellent for FW-GLS (0.88; p &amp;lt;0.001), SrRA (0.99; p &amp;lt;0.001) y ScdRA (0.93; p &amp;lt;0.001), good for SctRA (0.68; p= 0.019) and insufficient for 4C-SGL (0.36; p= 0.20). RV deformation values were significantly worse in patients than in controls (Table). However, solely FW-GLS showed improvement 3 months after the procedure (-15.7 vs -20.5 %; p= 0.022). Conclusions Right chamber deformation analysis is feasible and significantly reduced in patients undergoing TEER. Nevertheless, percutaneous treatment of TR has scarce effect in mechanics at early phases. Longer follow up may be required to evaluate the impact in right strain parameters. Table. Right chambers myocardial deformation values in healthy controls and patients undergoing TEER.Figure

Donor-derived cell-free DNA versus left ventricular global longitudinal strain in noninvasive detection of heart allograft injury

Abstract Introduction Invasive endomyocardial biopsy (EMB) is still considered the gold standard method for monitoring heart transplant (HTx) rejection. However, it is associated with potentially serious complications, and its histopathologic interpretation has high interobserver variability. Thus, noninvasive methods for surveillance of the transplanted organ remains of great interest. The noninvasive donor-derived cell-free DNA is a marker of graft injury which can indicate acute rejection, even before the histopathological diagnosis of rejection detected by EMB. The assay has a high negative predictive value for acute rejection. Left ventricular global longitudinal strain (LV GLS) is an emerging marker of subclinical myocardial injury which plays a crucial role in the evaluation of several cardiac diseases; however, its role in HTx injury has so far been poorly investigated. Purpose We aimed to investigate the efficacy of a local laboratory-run dd-cfDNA assay-based heart transplant rejection surveillance programme, and the prognostic and discriminatory performance of LV GLS for subsequent heart allograft injury. Methods HTx recipients without a history of allograft rejection requiring intensification of immunosuppressive therapy were transitioned from our EMB-based surveillance protocol to dd-cfDNA-based rejection surveillance. We assessed the percentage of dd-cfDNA to total cell-free DNA. Injury cut-off was defined by a dd-cfDNA level ≥0.20%, while the severe injury threshold was at ≥0.35%. LV ejection fraction was calculated using 2D Simpson’s method. To derive LV GLS, we analyzed apical 4-chamber, 2-chamber, and long-axis images by 2D speckle tracking echocardiography. Results A total of 244 dd-cfDNA samples were analyzed from 46 patients in fifteen runs performed monthly between October 2022 and December 2023. The dd-cfDNA fraction remained below the injury cut-off in most patients (81%). There were no missed rejection episodes. 88% of routine EMBs that would have otherwise been performed were safely avoided. Eighteen for-cause EMBs were performed based on dd-cfDNA levels suggesting injury. One EMB verified moderate cellular rejection (ISHLT grade 2R) that required steroid pulse therapy. Two EMBs proved mild cellular rejection, one EMB confirmed mild antibody-mediated rejection, while one EMB proved mixed rejection. Thirteen EMBs found no rejection. Mean LVEF was preserved (57.6%±7.6%). Mean LV GLS was -15.1%±2.5% in the whole cohort, while -15.6%±4.7% in recipients with HTx rejection on indication EMB. There was no significant correlation between dd-cfDNA and LV GLS (n=159). Conclusion The noninvasive dd-cfDNA assay is effective in ruling out acute rejection and safely decreases the necessity of invasive surveillance EMBs after HTx. LV GLS is less sensitive in comparison with the dd-cfDNA assay for detection of myocardial injury in the early stages of graft rejection in HTx recipients at low risk for rejection.

Interobserver variability of clinical target volume delineation in patients undergoing breast-conserving surgery without surgical clips: a pilot study on preoperative magnetic resonance simulation

BackgroundIn patients undergoing breast-conserving therapy without surgical clip implantation, the accuracy of tumor bed identification and the consistency of clinical target volume (CTV) delineation under computed tomography (CT) simulation remain suboptimal. This study aimed to investigate the feasibility of implementing preoperative magnetic resonance (MR) simulation on delineations by assessing interobserver variability (IOV).MethodsPreoperative MR and postoperative CT simulations were performed in patients who underwent breast-conserving surgery with no surgical clips implanted. Custom immobilization pads were used to ensure the same supine position. Three radiation oncologists independently delineated the CTV of tumor bed on the images acquired from MR and CT simulation registration and CT simulation alone. Cavity visualization score (CVS) was assigned to each patient based on the clarity of the tumor bed on CT simulation images. IOV was indicated by generalized conformity index (CIgen), denoted as CIgen−CT and CIgen−MR/CT, and the distance between the centroid of mass (dCOM), denoted as dCOMCT and dCOMMR/CT. The variation of IOV in different CVS subgroups was analyzed.ResultsA total of 10 patients were enrolled in this study. The median and interquartile range (IQR) of maximum pathological diameter of the tumors in all patients were 1.55 (0.80–1.92) cm. No statistical significance was found between the volumes of CTVs on CT simulation and on MR/CT simulation registration images (p = 0.387). CIgen−MR/CT was significantly larger than CIgen−CT (p = 0.005). dCOMMR/CT was significantly smaller than dCOMCT (p = 0.037). The median and IQR of CVS in all patients were 2.34 (2.00–3.08). The difference of CIgen between CIgen−MR/CT and CIgen−CT was larger in the low CVS group (p = 0.016). The difference of dCOM showed a decreasing trend when CVS was lower, although it did not reach statistical significance (p = 0.095).ConclusionsFor patients who underwent breast-conserving surgery without surgical clip implantation, the use of preoperative MR simulation in delineating the CTV of tumor bed decreased the IOV among observers. The consistency of tumor bed identification was improved especially in cases where the margins of tumor bed were challenging to visualize on CT simulation images. The study findings offer potential benefits in reducing local recurrence and minimizing tissue irritation in the surrounding areas. Future investigation in a larger patient cohort to validate our results is warranted.

Serous Ovarian Carcinoma: Detailed Analysis of Clinico-Pathological Characteristics as Prognostic Factors.

Serous ovarian carcinoma (SOC) is the most common subtype of epithelial ovarian cancer, with high-grade (HGSOC) and low-grade (LGSOC) subtypes presenting distinct clinical behaviours. This study aimed to evaluate histopathologic features in SOC, correlating these with prognostic outcomes, and explore the potential clinical implications. We analysed 51 SOC cases for lymphovascular space invasion (LVSI), tumour border configuration (TBC), microvessel density (MVD), tumour budding (TB), the tumour-stroma ratio (TSR), the stromal type, tumour-infiltrating lymphocytes (TILs), and tertiary lymphoid structures (TLSs). A validation cohort of 54 SOC cases from The Cancer Genome Atlas (TCGA) was used for comparison. In the discovery set, significant predictors of aggressive behaviour included LVSI, high MVD, high TB, and low TILs. These findings were validated in the validation set where the absence of TLSs, lower peritumoural TILs, immature stromal type, and low TSR were associated with worse survival outcomes. The stromal type was identified as an independent prognostic predictor in SOC across both datasets. Inter-observer variability analysis demonstrated substantial to almost perfect agreement for these features, ensuring the reproducibility of the findings. The histopathological evaluation of immune and stromal features, such as TILs, TLSs, TB, TSR, and stromal type, provides critical prognostic information for SOC. Incorporating these markers into routine pathological assessments could enhance risk stratification and guide treatment, offering practical utility, particularly in low-resource settings when molecular testing is not feasible.

Mitigating Interobserver Variability in Radiomics with ComBat: A Feasibility Study

This study investigates Intraobserver Features Variability (IFV) in radiomics studies and assesses the effectiveness of the ComBat harmonization method in mitigating these effects. Methods: This study utilizes data from the NSCLC-Radiomics-Interobserver1 dataset, comprising CT scans of 22 Non-Small Cell Lung Cancer (NSCLC) patients, with multiple Gross Tumor Volume (GTV) delineations performed by five radiation oncologists. Segmentation was completed manually (“vis”) or by autosegmentation with manual editing (“auto”). A total of 1229 radiomic features were extracted from each GTV, segmentation method, and oncologist. Features extracted included first order, shape, GLCM, GLRLM, GLSZM, and GLDM from original, wavelet-filtered, and LoG-filtered images. Results: Before implementing ComBat harmonization, 83% of features exhibited p-values below 0.05 in the “vis” approach; this percentage decreased to 34% post-harmonization. Similarly, for the “auto” approach, 75% of features demonstrated statistical significance prior to ComBat, but this figure declined to 33% after its application. Among a subset of three expert radiation oncologists, percentages changed from 77% to 25% for “vis” contouring and from 64% to 23% for “auto” contouring. This study demonstrates that ComBat harmonization could effectively reduce IFV, enhancing the feasibility of multicenter radiomics studies. It also highlights the significant impact of physician experience on radiomics analysis outcomes.

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.