AI-based preeclampsia detection and prediction with electrocardiogram data.

Introduction: More than 76,000 women die yearly from preeclampsia and hypertensive disorders of pregnancy. Early diagnosis and management of preeclampsia can improve outcomes for both mother and baby. In this study, we developed artificial intelligence models to detect and predict preeclampsia from electrocardiograms (ECGs) in point-of-care settings. Methods: Ten-second 12-lead ECG data was obtained from two large health care settings: University of Tennessee Health Science Center (UTHSC) and Atrium Health Wake Forest Baptist (AHWFB). UTHSC data was split into 80% training and 20% holdout data. The model used a modified ResNet convolutional neural network, taking one-dimensional raw ECG signals comprising 12 channels as an input, to predict risk of preeclampsia. Sub-analyses were performed to assess the predictive accuracy for preeclampsia prediction within 30, 60, or 90 days before diagnosis. Results: The UTHSC cohort included 904 ECGs from 759 females (78.8% African American) with a mean ± sd age of 27.3 ± 5.0 years. The AHWFB cohort included 817 ECGs from 141 females (45.4 African American) with a mean ± sd age of 27.4 ± 5.9 years. The cross-validated ECG-AI model yielded an AUC (95% CI) of 0.85 (0.77-0.93) on UTHSC holdout data, and an AUC (95% CI) of 0.81 (0.77-0.84) on AHWFB data. The sub-analysis of different time windows before preeclampsia prediction resulted in AUCs (95% CI) of 0.92 (0.84-1.00), 0.89 (0.81-0.98) and 0.90 (0.81-0.98) when tested on ECGs 30 days, 60 days and 90 days, respectively, before diagnosis. When assessed on early onset preeclampsia (preeclampsia diagnosed at <34 weeks of pregnancy), the model's AUC (95% CI) was 0.98 (0.89-1.00). Discussion: We conclude that preeclampsia can be identified with high accuracy via application of AI models to ECG data.

Development and validation of an electrocardiographic artificial intelligence model for detection of peripartum cardiomyopathy.

BACKGROUND: This study used electrocardiogram data in conjunction with artificial intelligence methods as a noninvasive tool for detecting peripartum cardiomyopathy. OBJECTIVE: This study aimed to assess the efficacy of an artificial intelligence-based heart failure detection model for peripartum cardiomyopathy detection. STUDY DESIGN: We first built a deep-learning model for heart failure detection using retrospective data at the University of Tennessee Health Science Center. Cases were adult and nonpregnant female patients with a heart failure diagnosis; controls were adult nonpregnant female patients without heart failure. The model was then tested on an independent cohort of pregnant women at the University of Tennessee Health Science Center with or without peripartum cardiomyopathy. We also tested the model in an external cohort of pregnant women at Atrium Health Wake Forest Baptist. Key outcomes were assessed using the area under the receiver operating characteristic curve. We also repeated our analysis using only lead I electrocardiogram as an input to assess the feasibility of remote monitoring via wearables that can capture single-lead electrocardiogram data. RESULTS: The University of Tennessee Health Science Center heart failure cohort comprised 346,339 electrocardiograms from 142,601 patients. In this cohort, 60% of participants were Black and 37% were White, with an average age (standard deviation) of 53 (19) years. The heart failure detection model achieved an area under the curve of 0.92 on the holdout set. We then tested the ability of the heart failure model to detect peripartum cardiomyopathy in an independent University of Tennessee Health Science Center cohort of pregnant women and an external Atrium Health Wake Forest Baptist cohort of pregnant women. The independent University of Tennessee Health Science Center cohort included 158 electrocardiograms from 115 patients; our deep-learning model achieved an area under the curve of 0.83 (0.77-0.89) for this data set. The external Atrium Health Wake Forest Baptist cohort involved 80 electrocardiograms from 43 patients; our deep-learning model achieved an area under the curve of 0.94 (0.91-0.98) for this data set. For identifying peripartum cardiomyopathy diagnosed ≥10 days after delivery, the model achieved an area under the curve of 0.88 (0.81-0.94) for the University of Tennessee Health Science Center cohort and of 0.96 (0.93-0.99) for the Atrium Health Wake Forest Baptist cohort. When we repeated our analysis by building a heart failure detection model using only lead-I electrocardiograms, we obtained similarly high detection accuracies, with areas under the curve of 0.73 and 0.93 for the University of Tennessee Health Science Center and Atrium Health Wake Forest Baptist cohorts, respectively. CONCLUSION: Artificial intelligence can accurately detect peripartum cardiomyopathy from electrocardiograms alone. A simple electrocardiographic artificial intelligence-based peripartum screening could result in a timelier diagnosis. Given that results with 1-lead electrocardiogram data were similar to those obtained using all 12 leads, future studies will focus on remote screening for peripartum cardiomyopathy using smartwatches that can capture single-lead electrocardiogram data.

Qualitative feedback from caregivers in a multidisciplinary pediatric neuromuscular clinic.

PURPOSE: This study explored family satisfaction and perceived quality of care in a pediatric neuromuscular care clinic to assess the value of the multidisciplinary clinic (MDC) model in delivering coordinated care to children with neuromuscular disorders, such as cerebral palsy. METHODS: Caregivers of 22 patients were administered a qualitative survey assessing their perceptions of clinic efficiency, care coordination, and communication. Surveys were audio-recorded and transcribed. Thematic analysis was completed using both deductive and inductive methods. RESULTS: All caregivers reported that providers adequately communicated next steps in the patient's care, and most reported high confidence in caring for the patient as a result of the clinic. Four major themes were identified from thematic analysis: Care Delivery, Communication, Care Quality, and Family-Centeredness. Caregivers emphasized that the MDC model promoted access to care, enhanced efficiency, promoted provider teamwork, and encouraged shared care planning. Caregivers also valued a physical environment that was suitable for patients with complex needs. CONCLUSION: This study demonstrated that caregivers believed the MDC model was both efficient and convenient for pediatric patients with neuromuscular disorders. This model has the potential to streamline medical care and can be applied more broadly to improve care coordination for children with medical complexity.

Quantitative Prediction of Right Ventricular Size and Function From the ECG.

BACKGROUND: Right ventricular ejection fraction (RVEF) and end-diastolic volume (RVEDV) are not readily assessed through traditional modalities. Deep learning-enabled ECG analysis for estimation of right ventricular (RV) size or function is unexplored. METHODS AND RESULTS: We trained a deep learning-ECG model to predict RV dilation (RVEDV >120 mL/m2), RV dysfunction (RVEF ≤40%), and numerical RVEDV and RVEF from a 12-lead ECG paired with reference-standard cardiac magnetic resonance imaging volumetric measurements in UK Biobank (UKBB; n=42 938). We fine-tuned in a multicenter health system (MSHoriginal [Mount Sinai Hospital]; n=3019) with prospective validation over 4 months (MSHvalidation; n=115). We evaluated performance with area under the receiver operating characteristic curve for categorical and mean absolute error for continuous measures overall and in key subgroups. We assessed the association of RVEF prediction with transplant-free survival with Cox proportional hazards models. The prevalence of RV dysfunction for UKBB/MSHoriginal/MSHvalidation cohorts was 1.0%/18.0%/15.7%, respectively. RV dysfunction model area under the receiver operating characteristic curve for UKBB/MSHoriginal/MSHvalidation cohorts was 0.86/0.81/0.77, respectively. The prevalence of RV dilation for UKBB/MSHoriginal/MSHvalidation cohorts was 1.6%/10.6%/4.3%. RV dilation model area under the receiver operating characteristic curve for UKBB/MSHoriginal/MSHvalidation cohorts was 0.91/0.81/0.92, respectively. MSHoriginal mean absolute error was RVEF=7.8% and RVEDV=17.6 mL/m2. The performance of the RVEF model was similar in key subgroups including with and without left ventricular dysfunction. Over a median follow-up of 2.3 years, predicted RVEF was associated with adjusted transplant-free survival (hazard ratio, 1.40 for each 10% decrease; P=0.031). CONCLUSIONS: Deep learning-ECG analysis can identify significant cardiac magnetic resonance imaging RV dysfunction and dilation with good performance. Predicted RVEF is associated with clinical outcome.

Epidemiology of pediatric trauma and fractures during and beyond the COVID-19 pandemic.

Purpose: Previous literature has shown decreases in pediatric trauma during the COVID-19 outbreak, but few have analyzed beyond the peak of the pandemic. This study assesses the epidemiology of pediatric trauma cases in a high-volume teaching hospital in New York City before, during, and after the height of the COVID-19 pandemic. Methods: Institutional data on pediatric trauma orthopedic cases from January 1, 2018 to November 30, 2021 were extracted. The following time frames were studied: (1) April 1-June 22 in 2018 and 2019 (pre-pandemic), (2) April 1-June 22, 2020 (peak pandemic), and (3) April 1-June 22, 2021 (post-peak pandemic). Inferential statistics were used to compare patient and trauma characteristics. Results: Compared to the pre-pandemic cohort (n = 6770), the peak pandemic cohort (n = 828) had a greater proportion of fractures (p < 0.01) and had a significantly decreased overall traumas per week rate (p < 0.01) and fractures per week rate (p < 0.01). These decreased trauma (p < 0.01) and fracture rates (p < 0.01) persisted for the post-peak pandemic cohort (n = 2509). Spatial analysis identified zip code clusters throughout New York City with higher rates of emergency department presentation during the peak pandemic compared to pre-pandemic, and these areas aligned with lower-income neighborhoods. Conclusion: During the peak of the pandemic, overall trauma and fracture volumes decreased, the types of prevalent injuries changed, and neighborhoods of different economic resources were variably impacted. These trends have mostly persisted for 12 months post-peak pandemic. This longitudinal analysis helps inform and improve long-term critical care and public health resource allocation for the future. Level of evidence: Level III.

The Impact of COVID-19 on the Timing of Rotator Cuff Repair and Method of Postoperative Follow-up.

Objective Rotator cuff repair (RCR) is one of the most common arthroscopic procedures. Our investigation aims to quantify the impact that the COVID-19 pandemic had on RCR, specifically on patients with acute, traumatic injuries. Methods Institutional records were queried to identify patients who underwent arthroscopic RCR between March 1 st to October 31 st of both 2019 and 2020. Patient demographic, preoperative, perioperative, and postoperative data were collected from electronic medical records. Inferential statistics were used to analyze data. Results Totals of 72 and of 60 patients were identified in 2019 and in 2020, respectively. Patients in 2019 experienced shorter lengths of time from MRI to surgery (62.7 ± 70.5 days versus 115.7 ± 151.0 days; p = 0.01). Magnetic resonance imaging (MRI) scans showed a smaller average degree of retraction in 2019 (2.1 ± 1.3 cm versus 2.6 ± 1.2 cm; p = 0.05) but no difference in anterior to posterior tear size between years (1.6 ± 1.0 cm versus 1.8 ± 1.0 cm; p = 0.17). Less patients in 2019 had a telehealth postoperative consultation with their operating surgeon compared with 2020 (0.0% versus 10.0%; p = 0.009). No significant changes in complications (0.0% versus 0.0%; p > 0.999), readmission (0.0% versus 0.0%; p > 0.999), or revision rates (5.6% versus 0.0%; p = 0.13) were observed. Conclusion From 2019 to 2020, there were no significant differences in patient demographics or major comorbidities. Our data suggests that even though the time from MRI to surgery was delayed in 2020 and telemedicine appointments were necessary, RCR was still performed in a timely manner and with no significant changes in early complications. Level of Evidence III.

Externally validated deep learning model to identify prodromal Parkinson's disease from electrocardiogram.

Little is known about electrocardiogram (ECG) markers of Parkinson's disease (PD) during the prodromal stage. The aim of the study was to build a generalizable ECG-based fully automatic artificial intelligence (AI) model to predict PD risk during the prodromal stage, up to 5 years before disease diagnosis. This case-control study included samples from Loyola University Chicago (LUC) and University of Tennessee-Methodist Le Bonheur Healthcare (MLH). Cases and controls were matched according to specific characteristics (date, age, sex and race). Clinical data were available from May, 2014 onward at LUC and from January, 2015 onward at MLH, while the ECG data were available as early as 1990 in both institutes. PD was denoted by at least two primary diagnostic codes (ICD9 332.0; ICD10 G20) at least 30 days apart. PD incidence date was defined as the earliest of first PD diagnostic code or PD-related medication prescription. ECGs obtained at least 6 months before PD incidence date were modeled to predict a subsequent diagnosis of PD within three time windows: 6 months-1 year, 6 months-3 years, and 6 months-5 years. We applied a novel deep neural network using standard 10-s 12-lead ECGs to predict PD risk at the prodromal phase. This model was compared to multiple feature engineering-based models. Subgroup analyses for sex, race and age were also performed. Our primary prediction model was a one-dimensional convolutional neural network (1D-CNN) that was built using 131 cases and 1058 controls from MLH, and externally validated on 29 cases and 165 controls from LUC. The model was trained on 90% of the MLH data, internally validated on the remaining 10% and externally validated on LUC data. The best performing model resulted in an external validation AUC of 0.67 when predicting future PD at any time between 6 months and 5 years after the ECG. Accuracy increased when restricted to ECGs obtained within 6 months to 3 years before PD diagnosis (AUC 0.69) and was highest when predicting future PD within 6 months to 1 year (AUC 0.74). The 1D-CNN model based on raw ECG data outperformed multiple models built using more standard ECG feature engineering approaches. These results demonstrate that a predictive model developed in one cohort using only raw 10-s ECGs can effectively classify individuals with prodromal PD in an independent cohort, particularly closer to disease diagnosis. Standard ECGs may help identify individuals with prodromal PD for cost-effective population-level early detection and inclusion in disease-modifying therapeutic trials.

Quantitative prediction of right ventricular and size and function from the electrocardiogram.

Background: Right ventricular ejection fraction (RVEF) and end-diastolic volume (RVEDV) are not readily assessed through traditional modalities. Deep-learning enabled 12-lead electrocardiogram analysis (DL-ECG) for estimation of RV size or function is unexplored. Methods: We trained a DL-ECG model to predict RV dilation (RVEDV>120 mL/m2), RV dysfunction (RVEF≤40%), and numerical RVEDV/RVEF from 12-lead ECG paired with reference-standard cardiac MRI (cMRI) volumetric measurements in UK biobank (UKBB; n=42,938). We fine-tuned in a multi-center health system (MSHoriginal; n=3,019) with prospective validation over 4 months (MSHvalidation; n=115). We evaluated performance using area under the receiver operating curve (AUROC) for categorical and mean absolute error (MAE) for continuous measures overall and in key subgroups. We assessed association of RVEF prediction with transplant-free survival with Cox proportional hazards models. Results: Prevalence of RV dysfunction for UKBB/MSHoriginal/MSHvalidation cohorts was 1.0%/18.0%/15.7%, respectively. RV dysfunction model AUROC for UKBB/MSHoriginal/MSHvalidation cohorts was 0.86/0.81/0.77, respectively. Prevalence of RV dilation for UKBB/MSHoriginal/MSHvalidation cohorts was 1.6%/10.6%/4.3%. RV dilation model AUROC for UKBB/MSHoriginal/MSHvalidation cohorts 0.91/0.81/0.92, respectively. MSHoriginal MAE was RVEF=7.8% and RVEDV=17.6 ml/m2. Performance was similar in key subgroups including with and without left ventricular dysfunction. Over median follow-up of 2.3 years, predicted RVEF was independently associated with composite outcome (HR 1.37 for each 10% decrease, p=0.046). Conclusions: DL-ECG analysis can accurately identify significant RV dysfunction and dilation both overall and in key subgroups. Predicted RVEF is independently associated with clinical outcome.

The Role of Experience: How Case Volume and Endourology-Fellowship Training Impact Surgical Outcomes for Ureteroscopy.

Introduction: Surgical experience is associated with superior outcomes in complex urologic cases, such as prostatectomy, nephrectomy, and cystectomy. The question remains whether experience is predictive of outcomes for less complex procedures, such as ureteroscopy (URS). Our study examined how case volume and endourology-fellowship training impacts URS outcomes. Methods: We retrospectively reviewed URS cases from 2017 to 2019 by high ureteroscopy volume urologists (HV), low ureteroscopy volume urologists (LV), endourology-fellowship trained (FT), and non-endourology FT (NFT) urologists. Surgical outcomes including stone-free rate (SFR), complication and reoperation rates, and postoperative imaging follow-up were analyzed between groups. Results: One thousand fifty-seven cases were reviewed across 23 urologists: 6 HV, 17 LV, 3 FT, and 20 NFT. Both FT and HV operated on more complex cases with lower rates of pre-stented patients. HV also operated on patients with higher rates of renal stones, lower pole involvement, and prior failed procedures. Despite this, FT and HV showed between 11.7% and 14.4% higher SFR, representing 2.7- to 3.6-fold greater odds of stone-free outcomes for primary and secondary stones. Additionally, HV and FT had a 4.9% to 7.8% lower rate of postoperative complications and a 3.3% to 4.3% lower rate of reoperations, representing 1.9- to 4.0-fold lower odds of complications. Finally, their patients had a 1.6- to 2.1-fold higher odds of postoperative imaging follow-up with a greater proportion receiving postoperative imaging within the recommended 3-month postoperative period. Conclusions: More experienced urologists, as defined by higher case volume and endourology-fellowship training, had higher SFR, lower complication and reoperation rates, and better postoperative imaging follow-up compared with less experienced urologists. Although less experienced urologists had outcomes in-line with clinical and literature standards, continued training and experience may be a predictor of better outcomes across multiple URS modalities.

The Impact of COVID-19 on the Timing of Rotator Cuff Repair and Method of Postoperative Follow-up / Impacto da COVID-19 no momento de reparo do manguito rotador e método de acompanhamento pós-operatório

Abstract Objective Rotator cuff repair (RCR) is one of the most common arthroscopic procedures. Our investigation aims to quantify the impact that the COVID-19 pandemic had on RCR, specifically on patients with acute, traumatic injuries. Methods Institutional records were queried to identify patients who underwent arthroscopic RCR between March 1st to October 31st of both 2019 and 2020. Patient demographic, preoperative, perioperative, and postoperative data were collected from electronic medical records. Inferential statistics were used to analyze data. Results Totals of 72 and of 60 patients were identified in 2019 and in 2020, respectively. Patients in 2019 experienced shorter lengths of time from MRI to surgery (62.7 ± 70.5 days versus 115.7 ± 151.0 days; p = 0.01). Magnetic resonance imaging (MRI) scans showed a smaller average degree of retraction in 2019 (2.1 ± 1.3 cm versus 2.6 ± 1.2 cm; p = 0.05) butnodifference in anterior toposterior tear size between years (1.6 ± 1.0 cm versus 1.8 ± 1.0 cm; p = 0.17). Less patients in 2019 had a tele-health postoperative consultation with their operating surgeon compared with 2020 (0.0% versus 10.0%; p = 0.009). No significant changes in complications (0.0% versus 0.0%; p > 0.999), readmission (0.0% versus 0.0%; p > 0.999), or revision rates (5.6% versus 0.0%; p = 0.13) were observed. Conclusion From 2019 to 2020, there were no significant differences in patient demographics or major comorbidities. Our data suggests that even though the time from MRI to surgery was delayed in 2020 and telemedicine appointments were necessary, RCR was still performed in a time in early complications. Level of Evidence III.

Resumo Objetivo Oreparodomanguitorotador (RMR) é um dos procedimentos artroscópi-cos maiscomuns. Nossapesquisavisaquantificar o impacto da pandemia de COVID-19 sobre o RMR, especificamente em pacientes com lesões agudas e traumáticas. Métodos Os prontuários institucionais foram consultados para identificação de pacientes submetidos ao RMR artroscópico entre 1° de março e 31 de outubro de 2019 e de 2020. Dados demográficos, pré-operatórios, perioperatórios e pós-operatórios dos pacientes foram coletados de prontuários eletrônicos. Os dados foram analisados por estatística inferencial. Resultados Totais de 72 ede60pacientes foramidentificados em 2019 e 2020, respectivamente. Os pacientes de 2019 apresentaram menor intervalo entre a ressonância magnética (RM) e a cirurgia (62,7 ± 70,5 dias versus 115,7 ± 151,0 dias; p = 0,01). Os exames de RM mostraram menor grau médio de retração em 2019 (2,1 ± 1,3 cm versus 2,6 ± 1,2 cm; p = 0,05), mas nenhuma diferença foi observada na extensão anteroposterior da laceração entre os anos (1,6 ± 1,0 cm versus 1,8 ± 1,0 cm; p = 0,17).Em 2019,o número de pacientes atendidos por seus cirurgiões em consultas pós-operatórias por telemedicina foi menor em comparação com 2020 (0,0% versus 10,0%; p = 0,009). Não foram observadas alterações significativas nas taxas de complicação (0,0% versus 0,0%; p > 0,999), de readmissão (0,0% versus 0,0%; p > 0,999) ou de revisão (5,6% versus 0,0%; p = 0,13). Conclusão Não houve diferenças significativas nos dados demográficos dos pacientes ou nas principais comorbidades entre 2019 e 2020. Nossos dados sugerem que, embora o intervalo entre a RM e a cirurgia tenha sido maior em 2020 e tenha havido necessidade de consultas por telemedicina, o RMR ainda foi realizado em tempo hábil e sem alterações significativas nas complicações precoces. Nível de Evidência III.

Risk factors for 90-day readmission and prolonged length of stay after hip surgery in children with cerebral palsy.

Objective: Spastic hip dysplasia is a common complication of cerebral palsy in children, and surgical intervention is usually warranted. While current literature has primarily analyzed single institution outcomes, this study utilized a national database to describe readmission rates and factors correlated with readmission for children with cerebral palsy undergoing hip surgery in order to treat this population more effectively. Methods: This study queried the Nationwide Readmissions Database (2014-2018) for pediatric patients with cerebral palsy who underwent hip surgery. Patient demographics, pre-operative comorbidities, length of stay (LOS), treatment complications, and readmission data were collected for each patient and analyzed with inferential statistics. Results: Of the 1225 patients included, the average age was 9.3 ± 3.8 years and 42.8% were female. Approximately 26.3% patients had a prolonged LOS (≥5 days) and 14.2% patients required readmission within 90-days of surgery. Medical complications, cardiac arrhythmias, and iron deficiency anemia were all significantly associated with elongated LOS as well as 90-day readmission. Patients with Medicaid were more frequently associated with an inpatient medical complication and the overall complication rate was 5.5%. Conclusions: While current literature has analyzed common risk factors and complications associated with hip surgery in the pediatric cerebral palsy patient, this study identifies a national readmission rate (14.2%) as well as preoperative comorbidities associated with readmission within 90-days and/or elongated LOS. Notably, complications are more frequently associated with patients using Medicaid. These results further exemplify the importance of equitable access to care and thorough selection of pediatric cerebral palsy patients appropriate for hip surgery.

S-acylation of Sprouty and SPRED proteins by the S-acyltransferase zDHHC17 involves a novel mode of enzyme-substrate interaction.

S-acylation is an essential post-translational modification, which is mediated by a family of 23 zDHHC enzymes in humans. Several thousand proteins are modified by S-acylation; however, we lack a detailed understanding of how enzyme-substrate recognition and specificity is achieved. Previous work showed that the ankyrin repeat domain of zDHHC17 (ANK17) recognizes a short linear motif, known as the zDHHC ANK binding motif (zDABM) in substrate protein SNAP25, as a mechanism of substrate recruitment prior to S-acylation. Here, we investigated the S-acylation of the Sprouty and SPRED family of proteins by zDHHC17. Interestingly, although Sprouty-2 (Spry2) contains a zDABM that interacts with ANK17, this mode of binding is dispensable for S-acylation, and indeed removal of the zDABM does not completely ablate binding to zDHHC17. Furthermore, the related SPRED3 protein interacts with and is efficiently S-acylated by zDHHC17, despite lacking a zDABM. We undertook mutational analysis of SPRED3 to better understand the basis of its zDABM-independent interaction with zDHHC17. This analysis found that the cysteine-rich SPR domain of SPRED3, which is the defining feature of all Sprouty and SPRED proteins, interacts with zDHHC17. Surprisingly, the interaction with SPRED3 was independent of ANK17. Our mutational analysis of Spry2 was consistent with the SPR domain of this protein containing a zDHHC17-binding site, and Spry2 also showed detectable binding to a zDHHC17 mutant lacking the ANK domain. Thus, zDHHC17 can recognize its substrates through zDABM-dependent and/or zDABM-independent mechanisms, and some substrates display more than one mode of binding to this enzyme.

A national analysis of the effect alcohol use disorder has on short-term complications and readmissions following total shoulder arthroplasty.

Background: Alcohol use disorder has been associated with broad health consequences that may interfere with healing after total shoulder arthroplasty. The aim of this study was to explore the impact of alcohol use disorder on readmissions and complications following total shoulder arthroplasty. Methods: We used data from the Healthcare Cost and Utilization Project National Readmissions Database (NRD) from 2016 to 2018. Patients were included based on International Classification of Diseases, 10th Revision (ICD-10) procedure codes for anatomic total shoulder arthroplasty (aTSA) and reverse total shoulder arthroplasty (rTSA). Patients with an alcohol use disorder (AUD) were identified using the ICD-10 diagnosis code F10.20. Demographics, complications, and 30-day and 90-day readmission were collected for all patients. A univariate logistic regression was performed to investigate AUD as a factor affecting readmission and complication rates. A multivariate logistic regression model was created to assess the impact of alcohol use disorder on complications and readmission while controlling for demographic factors. Results: In total, 164,527 patients were included, and 503 (0.3%) patients had a prior diagnosis of AUD. Revision surgery was more common in patients with an alcohol use disorder (8.8% vs. 6.2%; p = 0.022). Postoperative infection (p = 0.026), dislocation (p = 0.025), liver complications (p < 0.01), and 90-day readmission (p < 0.01) were more common in patients with a diagnosed AUD. On multivariate analysis, patients with an AUD were found to be at increased odds for liver complications (OR: 46.8; 95% CI: [32.8, 66.8]; p < 0.01). Comparatively, mean age, length of stay, and over healthcare costs were also higher for patients with an AUD. Conclusion: Patients with a diagnosis of AUD were more likely to suffer from shoulder dislocation, liver complications, and 90-day readmission, while also being younger and having longer hospital stays. Therefore, surgeons should take caution to anticipate and prevent complications and readmissions following total shoulder arthroplasty in patients with an AUD.

A generalizable electrocardiogram-based artificial intelligence model for 10-year heart failure risk prediction.

Background: Heart failure (HF) is a progressive condition with high global incidence. HF has two main subtypes: HF with preserved ejection fraction (HFpEF) and HF with reduced ejection fraction (HFrEF). There is an inherent need for simple yet effective electrocardiogram (ECG)-based artificial intelligence (AI; ECG-AI) models that can predict HF risk early to allow for risk modification. Objective: The main objectives were to validate HF risk prediction models using Multi-Ethnic Study of Atherosclerosis (MESA) data and assess performance on HFpEF and HFrEF classification. Methods: There were six models in comparision derived using ARIC data. 1) The ECG-AI model predicting HF risk was developed using raw 12-lead ECGs with a convolutional neural network. The clinical models from 2) ARIC (ARIC-HF) and 3) Framingham Heart Study (FHS-HF) used 9 and 8 variables, respectively. 4) Cox proportional hazards (CPH) model developed using the clinical risk factors in ARIC-HF or FHS-HF. 5) CPH model using the outcome of ECG-AI and the clinical risk factors used in CPH model (ECG-AI-Cox) and 6) A Light Gradient Boosting Machine model using 288 ECG Characteristics (ECG-Chars). All the models were validated on MESA. The performances of these models were evaluated using the area under the receiver operating characteristic curve (AUC) and compared using the DeLong test. Results: ECG-AI, ECG-Chars, and ECG-AI-Cox resulted in validation AUCs of 0.77, 0.73, and 0.84, respectively. ARIC-HF and FHS-HF yielded AUCs of 0.76 and 0.74, respectively, and CPH resulted in AUC = 0.78. ECG-AI-Cox outperformed all other models. ECG-AI-Cox provided an AUC of 0.85 for HFrEF and 0.83 for HFpEF. Conclusion: ECG-AI using ECGs provides better-validated predictions when compared to HF risk calculators, and the ECG feature model and also works well with HFpEF and HFrEF classification.

Building better pediatric surgeons: A sentiment analysis of online physician review websites.

Purpose: Physician review websites are a heavily utilized patient tool for finding, rating, and reviewing surgeons. Natural language processing such as sentiment analysis provides a comprehensive approach to better understand the nuances of patient perception. This study utilizes sentiment analysis to examine how specific patient sentiments correspond to positive and negative experiences in online reviews of pediatric orthopedic surgeons. Methods: The online written reviews and star ratings of pediatric surgeons belonging to the Pediatric Orthopaedic Society of North America were obtained from healthgrades.com. A sentiment analysis package obtained compound scores of each surgeon's reviews. Inferential statistics analyzed relationships between demographic variables and star/sentiment scores. Word frequency analyses and multiple logistic regression analyses were performed on key terms. Results: A total of 749 pediatric surgeons (3830 total online reviews) were included. 80.8% were males and 33.8% were below 50 years of age. Male surgeons and younger surgeons had higher mean star ratings. Surgeon attributes including "confident" (p < 0.01) and "comfortable" (p < 0.01) improved the odds of positive reviews, while "rude" (p < 0.01) and "unprofessional" (p < 0.01) decreased these odds. Comments regarding "pain" lowered the odds of positive reviews (p < 0.01), whereas "pain-free" increased these odds (p < 0.01). Conclusion: Pediatric surgeons who were younger, communicated effectively, eased pain, and curated a welcoming office setting were more likely to receive positively written online reviews. This suggests that a spectrum of interpersonal and ancillary factors impact patient experience and perceptions beyond surgical skill. These outcomes can advise pediatric surgeons on behavioral and office qualities that patients and families prioritize when rating/recommending surgeons online. Level of evidence: IV.

Predicting Parkinson's Disease and Its Pathology via Simple Clinical Variables.

BACKGROUND: Parkinson's disease (PD) is a chronic, disabling neurodegenerative disorder. OBJECTIVE: To predict a future diagnosis of PD using questionnaires and simple non-invasive clinical tests. METHODS: Participants in the prospective Kuakini Honolulu-Asia Aging Study (HAAS) were evaluated biannually between 1995-2017 by PD experts using standard diagnostic criteria. Autopsies were sought on all deaths. We input simple clinical and risk factor variables into an ensemble-tree based machine learning algorithm and derived models to predict the probability of developing PD. We also investigated relationships of predictive models and neuropathologic features such as nigral neuron density. RESULTS: The study sample included 292 subjects, 25 of whom developed PD within 3 years and 41 by 5 years. 116 (46%) of 251 subjects not diagnosed with PD underwent autopsy. Light Gradient Boosting Machine modeling of 12 predictors correctly classified a high proportion of individuals who developed PD within 3 years (area under the curve (AUC) 0.82, 95%CI 0.76-0.89) or 5 years (AUC 0.77, 95%CI 0.71-0.84). A large proportion of controls who were misclassified as PD had Lewy pathology at autopsy, including 79%of those who died within 3 years. PD probability estimates correlated inversely with nigral neuron density and were strongest in autopsies conducted within 3 years of index date (r = -0.57, p < 0.01). CONCLUSION: Machine learning can identify persons likely to develop PD during the prodromal period using questionnaires and simple non-invasive tests. Correlation with neuropathology suggests that true model accuracy may be considerably higher than estimates based solely on clinical diagnosis.

Comparison of Patient Preference, Understanding, and Sentiment for Minimally Invasive Versus Open Spine Surgery.

STUDY DESIGN: Retrospective questionnaire analysis. OBJECTIVE: The goal of this study was to analyze patients' understanding and preferences for minimally invasive spine (MIS) versus open spine surgery. SUMMARY OF BACKGROUND DATA: MIS surgery is increasing in prevalence. However, there is insufficient literature to evaluate how the availability of MIS surgery influences the patients' decision-making process and perceptions of spine procedures. METHODS: A survey was administered to patients who received a microdiscectomy or transforaminal lumbar interbody fusion between 2016 and 2020. All eligible patients were stratified into two cohorts based on the use of minimally invasive techniques. Each cohort was administered a survey that evaluated patient preferences, perceptions, and understanding of their surgery. RESULTS: One hundred fifty two patients completed surveys (MIS: 88, Open: 64). There was no difference in time from surgery to survey (MIS: 2.1 ±â€Š1.4 yrs, Open: 1.9 ±â€Š1.4 yrs; P = 0.36) or sex (MIS: 56.8% male, Open: 53.1% male; P = 0.65). The MIS group was younger (MIS: 53.0 ±â€Š16.9 yrs, Open: 58.2 ±â€Š14.6 yrs; P = 0.05). More MIS patients reported that their technique influenced their surgeon choice (MIS: 64.0%, Open: 37.5%; P  < 0.00001) and increased their preoperative confidence (MIS: 77.9%, Open: 38.1%; P  < 0.00001). There was a trend towards the MIS group being less informed about the intraoperative specifics of their technique (MIS: 35.2%, Open: 23.4%; P = 0.12). More of the MIS cohort reported perceived advantages to their surgical technique (MIS: 98.8%, Open: 69.4%; P < 0.00001) and less reported disadvantages (MIS: 12.9%, Open: 68.8%; P < 0.00001). 98.9% and 87.1% of the MIS and open surgery cohorts reported a preference for MIS surgery in the future. CONCLUSION: Patients who received a MIS approach more frequently sought out their surgeons, were more confident in their procedure, and reported less perceived disadvantages following their surgery compared with the open surgery cohort. Both cohorts would prefer MIS surgery in the future. Overall, patients have positive perceptions of MIS surgery.Level of Evidence: 3.

Image and structured data analysis for prognostication of health outcomes in patients presenting to the ED during the COVID-19 pandemic.

BACKGROUND: Patients admitted to the emergency department (ED) with COVID-19 symptoms are routinely required to have chest radiographs and computed tomography (CT) scans. COVID-19 infection has been directly related to the development of acute respiratory distress syndrome (ARDS) and severe infections could lead to admission to intensive care and increased risk of death. The use of clinical data in machine learning models available at time of admission to ED can be used to assess possible risk of ARDS, the need for intensive care (admission to the Intensive Care Unit; ICU) as well as risk of mortality. In addition, chest radiographs can be inputted into a deep learning model to further assess these risks. PURPOSE: This research aimed to develop machine and deep learning models using both structured clinical data and image data from the electronic health record (EHR) to predict adverse outcomes following ED admission. MATERIALS AND METHODS: Light Gradient Boosting Machine (LightGBM) was used as the main machine learning algorithm using all clinical data including 42 variables. Compact models were also developed using the 15 most important variables to increase applicability of the models in clinical settings. To predict risk (or early stratified risk) of the aforementioned health outcome events, transfer learning from the CheXNet model was also implemented on the available data. This research utilized clinical data and chest radiographs of 3,571 patients, 18 years and older, admitted to the emergency department between 9th March 2020 and 29th October 2020 at Loyola University Medical Center. MAIN FINDINGS: The research results show that we can detect COVID-19 infection (AUC = 0.790 (0.746-0.835)), predict the risk of developing ARDS (AUC = 0.781 (0.690-0.872), risk stratification of the need for ICU admission (AUC = 0.675 (0.620-0.713)) and mortality (AUC = 0.759 (0.678-0.840)) at moderate accuracy from both chest X-ray images and clinical data. PRINCIPAL CONCLUSIONS: The results can help in clinical decision making, especially when addressing ARDS and mortality, during the assessment of patients admitted to the ED with or without COVID-19 symptoms.

Machine learning models improve prediction of large vessel occlusion and mechanical thrombectomy candidacy in acute ischemic stroke.

BACKGROUND AND PURPOSE: Early identification of large vessel occlusions (LVO) and timely recanalization are paramount to improved clinical outcomes in acute ischemic stroke. A stroke assessment that maximizes sensitivity and specificity for LVOs is needed to identify these cases and not overburden the health system with unnecessary transfers. Machine learning techniques are being used for predictive modeling in many aspects of stroke care and may have potential in predicting LVO presence and mechanical thrombectomy (MT) candidacy. METHODS: Ischemic stroke patients treated at Loyola University Medical Center from July 2018 to June 2019 (N = 286) were included. Thirty-five clinical and demographic variables were analyzed using machine learning algorithms, including logistic regression, extreme gradient boosting, random forest (RF), and decision trees to build models predictive of LVO presence and MT candidacy by area of the curve (AUC) analysis. The best performing model was compared with prior stroke scales. RESULTS: When using all 35 variables, RF best predicted LVO presence (AUC = 0.907 ± 0.856-0.957) while logistic regression best predicted MT candidacy (AUC = 0.930 ± 0.886-0.974). When compact models were evaluated, a 10-feature RF model best predicted LVO (AUC = 0.841 ± 0.778-0.904) and an 8-feature RF model best predicted MT candidacy (AUC = 0.862 ± 0.782-0.942). The compact RF models had sensitivity, specificity, negative predictive value and positive predictive value of 0.81, 0.87, 0.92, 0.72 for LVO and 0.87, 0.97, 0.97, 0.86 for MT, respectively. The 10-feature RF model was superior at predicting LVO to all previous stroke scales (AUC 0.944 vs 0.759-0.878) and the 8-feature RF model was superior at predicting MT (AUC 0.970 vs 0.746-0.834). CONCLUSION: Random forest machine learning models utilizing clinical and demographic variables predicts LVO presence and MT candidacy with a high degree of accuracy in an ischemic stroke cohort. Further validation of this strategy for triage of stroke patients requires prospective and external validation.

ECG-AI: electrocardiographic artificial intelligence model for prediction of heart failure.

AIMS: Heart failure (HF) is a leading cause of death. Early intervention is the key to reduce HF-related morbidity and mortality. This study assesses the utility of electrocardiograms (ECGs) in HF risk prediction. METHODS AND RESULTS: Data from the baseline visits (1987-89) of the Atherosclerosis Risk in Communities (ARIC) study was used. Incident hospitalized HF events were ascertained by ICD codes. Participants with good quality baseline ECGs were included. Participants with prevalent HF were excluded. ECG-artificial intelligence (AI) model to predict HF was created as a deep residual convolutional neural network (CNN) utilizing standard 12-lead ECG. The area under the receiver operating characteristic curve (AUC) was used to evaluate prediction models including (CNN), light gradient boosting machines (LGBM), and Cox proportional hazards regression. A total of 14 613 (45% male, 73% of white, mean age ± standard deviation of 54 ± 5) participants were eligible. A total of 803 (5.5%) participants developed HF within 10 years from baseline. Convolutional neural network utilizing solely ECG achieved an AUC of 0.756 (0.717-0.795) on the hold-out test data. ARIC and Framingham Heart Study (FHS) HF risk calculators yielded AUC of 0.802 (0.750-0.850) and 0.780 (0.740-0.830). The highest AUC of 0.818 (0.778-0.859) was obtained when ECG-AI model output, age, gender, race, body mass index, smoking status, prevalent coronary heart disease, diabetes mellitus, systolic blood pressure, and heart rate were used as predictors of HF within LGBM. The ECG-AI model output was the most important predictor of HF. CONCLUSIONS: ECG-AI model based solely on information extracted from ECG independently predicts HF with accuracy comparable to existing FHS and ARIC risk calculators.