Faculty Bibliography

BACKGROUND:Prolongation of the QTc interval is associated with the risk of torsades de pointes. Determination of the QTc interval is therefore of critical importance. There is no reliable method for measuring or correcting the QT interval in atrial fibrillation (AF). OBJECTIVES:The authors sought to evaluate the use of a convolutional neural network (CNN) applied to AF electrocardiograms (ECGs) for accurately estimating the QTc interval and ruling out prolongation of the QTc interval. METHODS:The authors identified patients with a 12-lead ECG in AF within 10 days of a sinus ECG, with similar (±10 ms) QRS durations, between October 23, 2001, and November 5, 2021. A multilayered deep CNN was implemented in TensorFlow 2.5 (Google) to predict the MUSE (GE Healthcare) software-generated sinus QTc value from an AF ECG waveform, demographic characteristics, and software-generated features. RESULTS:The study identified 6,432 patients (44% female) with an average age of 71 years. The CNN predicted sinus QTc values with a mean absolute error of 22.2 ms and root mean squared error of 30.6 ms, similar to the intrinsic variability of the sinus QTc interval. Approximately 84% and 97% of the model's predictions were contained within 1 SD (±30.6 ms) and 2 SD (±61.2 ms) from the sinus QTc interval. The model outperformed the AFQTc method, exhibiting narrower error ranges (mean absolute error comparison P < 0.0001). The model performed best for ruling out QTc prolongation (negative predictive value 0.82 male, 0.92 female; specificity 0.92 male, 0.97 female). CONCLUSIONS:A CNN model applied to AF ECGs accurately predicted the sinus QTc interval, outperforming current alternatives and exhibiting a high negative predictive value.

Cardiac amyloidosis (CA) often goes unrecognized as a cause of heart failure with preserved ejection fraction (HFpEF). There is paucity of contemporary data evaluating the trends of CA diagnosis and associated sex differences. Adult heart failure hospitalizations were identified from the National Inpatient Sample between 2016 and 2019. Hospitalizations with heart failure other than HFpEF were excluded. Hospitalizations with a diagnosis of CA were identified. A Linear regression was utilized to calculate the trend of CA diagnosis over time. A multivariate logistic regressions analysis was performed to analyze sex differences. There was an increasing trend of CA from 1.2 to 2.3 per 1000 HFpEF admission in the first quarter of 2016 to the fourth quarter of 2019 (P_trend <0.001). In females, as compared to males, there was an increased risk of AIS (6% vs 3%, aOR: 1.68[1.24-2.27], P=0.001) and major bleeding events (10% vs 5%, aOR: 1.97[1.53-2.52], P<0.001). No difference was observed in the in-hospital mortality outcome (8% vs 7%, aOR: 1.2[0.95-1.53], P=0.12) between both groups. Our real-world contemporary analysis showed an increase in CA diagnosis from 2016 to 2019. Despite similar in-hospital mortality, females were associated with higher AIS and major bleeding events rates. Further prospective studies are needed to validate these results.

INTRODUCTION/BACKGROUND:Contemporary efforts to predict surgical outcomes focus on the associations between traditional discrete surgical risk factors. We aimed to determine whether natural language processing (NLP) of unstructured operative notes improves the prediction of residual disease in women with advanced epithelial ovarian cancer (EOC) following cytoreductive surgery. METHODS:Electronic Health Records were queried to identify women with advanced EOC including their operative notes. The Term Frequency - Inverse Document Frequency (TF-IDF) score was used to quantify the discrimination capacity of sequences of words (n-grams) regarding the existence of residual disease. We employed the state-of-the-art RoBERTa-based classifier to process unstructured surgical notes. Discrimination was measured using standard performance metrics. An XGBoost model was then trained on the same dataset using both discrete and engineered clinical features along with the probabilities outputted by the RoBERTa classifier. RESULTS:The cohort consisted of 555 cases of EOC cytoreduction performed by eight surgeons between January 2014 and December 2019. Discrete word clouds weighted by n-gram TF-IDF score difference between R0 and non-R0 resection were identified. The words 'adherent' and 'miliary disease' best discriminated between the two groups. The RoBERTa model reached high evaluation metrics (AUROC .86; AUPRC .87, precision, recall, and F1 score of .77 and accuracy of .81). Equally, it outperformed models that used discrete clinical and engineered features and outplayed the performance of other state-of-the-art NLP tools. When the probabilities from the RoBERTa classifier were combined with commonly used predictors in the XGBoost model, a marginal improvement in the overall model's performance was observed (AUROC and AUPRC of .91, with all other metrics the same). CONCLUSION/IMPLICATIONS/CONCLUSIONS:We applied a sui generis approach to extract information from the abundant textual surgical data and demonstrated how it can be effectively used for classification prediction, outperforming models relying on conventional structured data. State-of-art NLP applications in biomedical texts can improve modern EOC care.

INTRODUCTION/BACKGROUND:Key challenges of left atrial appendage occlusion (LAAO) procedures include accurate preprocedural device sizing and proper device positioning at the left atrial appendage (LAA) ostium to ensure sufficient device anchoring and avoid peri-device leaks. To address these challenges, 3D printing (3DP) of LAA models has recently emerged in the literature, first being described in 2015. We present a review of the benefits and drawbacks of employing this technology in LAAO procedures. METHODS:A comprehensive search of Pubmed was performed from database inception to October 1, 2021. A total of 19 eligible studies were identified and evaluated based on pre-, intra-, and postprocedural outcomes. Preprocediral imaging modalities, 3D segmentation software, 3D printer, 3DD printing material, occlusion deevice, cohort size, average age, thrombotic /bleeding risk scores, and AF type were also recorded when available. RESULTS:Preprocedurally the use of 3DP can consistently and accurately determine LAAO device size over the standard of care approaches. Intraprocedurally 3DP's impact entailed a statistically significant decrease in the number of devices used per procedure, as well as in the fluoroscopic time and dose. Postprocedurally, there is some evidence that 3DP could reduce the rate of peri-device leaks, with limited data on its effect on complication rates. CONCLUSION/CONCLUSIONS:Based on existing evidence, we recommend the focused application of 3DP to cases of complex LAA anatomy and for the training of proceduralists. However, next-generation LAAO devices and emerging augmented/virtual reality (AR/VR) systems could limit even this narrow window of clinical benefit afforded by 3DP. Key challenges of left atrial appendage include accurate preprocedural device sizing and proper device positioning at the left atrial appendage (LAA) ostium to ensure sufficient device anchoring and avoid peri-device leaks. To address these challenges, 3D printing (3DP) of LAA models has recently emerged in the literature, first being described in 2015. We present a review of the benefits and drawbacks of employing this technology in LAAO procedures.

Background/UNASSIGNED:, known to colonize all parts of the cotton plant. Only one prior case report has suggested cotton fever as a potential risk factor of infective endocarditis (IE). Case summary/UNASSIGNED:IE with bilateral septic pulmonary emboli. Transthoracic echocardiography findings included new tricuspid regurgitation and two mobile echodensities on the right atrial implantable cardioverter defibrillator (ICD) lead. Despite broad antibiotic coverage and extraction of the ICD leads, the patient passed away from septic shock. Discussion/UNASSIGNED:genus were isolated in patients' blood cultures. This evidence supports the endotoxin theory as the leading pathophysiologic explanation for cotton fever and suggests cotton fever as a risk factor for Gram-negative IE. In the inpatient setting it informs proper antibiotic coverage, whereas in the outpatient setting it supports harm reduction interventions in the form of sterile cotton balls.

Complete surgical cytoreduction (R0 resection) is the single most important prognosticator in epithelial ovarian cancer (EOC). Explainable Artificial Intelligence (XAI) could clarify the influence of static and real-time features in the R0 resection prediction. We aimed to develop an AI-based predictive model for the R0 resection outcome, apply a methodology to explain the prediction, and evaluate the interpretability by analysing feature interactions. The retrospective cohort finally assessed 571 consecutive advanced-stage EOC patients who underwent cytoreductive surgery. An eXtreme Gradient Boosting (XGBoost) algorithm was employed to develop the predictive model including mostly patient- and surgery-specific variables. The Shapley Additive explanations (SHAP) framework was used to provide global and local explainability for the predictive model. The XGBoost accurately predicted R0 resection (area under curve [AUC] = 0.866; 95% confidence interval [CI] = 0.8−0.93). We identified “turning points” that increased the probability of complete cytoreduction including Intraoperative Mapping of Ovarian Cancer Score and Peritoneal Carcinomatosis Index < 4 and <5, respectively, followed by Surgical Complexity Score > 4, patient’s age < 60 years, and largest tumour bulk < 5 cm in a surgical environment of optimized infrastructural support. We demonstrated high model accuracy for the R0 resection prediction in EOC patients and provided novel global and local feature explainability that can be used for quality control and internal audit.

The ability to accurately and precisely measure the thickness of biomaterial constructs is critical for characterizing both specific dimensional features and related mechanical properties. However, in the absence of a standardized approach for thickness measurements, a variety of imaging modalities have been employed, which have been associated with varying limits of accuracy, particularly for ultrathin hydrated structures. Electron microscopy (EM), a commonly used modality, yields thickness values for extensively processed and nonhydrated constructs, potentially resulting in overestimated mechanical properties, including elastic modulus and ultimate tensile strength. Confocal laser scanning microscopy (CLSM) has often been used as a nondestructive imaging alternative. However, published CLSM-derived image analysis protocols use arbitrary signal intensity cutoffs and provide minimal information regarding thickness variability across imaged surfaces. To address the aforementioned limitations, we present a standardized, user-independent CLSM image acquisition and analysis approach developed as a custom ImageJ macro and validated with collagen-based scaffolds. In the process, we also quantify thickness discrepancies in collagen-based scaffolds between CLSM and EM techniques, further illustrating the need for improved strategies. Employing the same image acquisition protocol, we also demonstrate that this approach can be used to estimate the surface roughness of the same scaffolds without the use of specialized instrumentation.