Development and external validation of temporal fusion transformer models for continuous intraoperative blood pressure forecasting.

Background: During surgery, intraoperative hypotension is associated with postoperative morbidity and should therefore be avoided. Predicting the occurrence of hypotension in advance may allow timely interventions to prevent hypotension. Previous prediction models mostly use high-resolution waveform data, which is often not available. Methods: We utilised a novel temporal fusion transformer (TFT) algorithm to predict intraoperative blood pressure trajectories 7 min in advance. We trained the model with low-resolution data (sampled every 15 s) from 73,009 patients who were undergoing general anaesthesia for non-cardiothoracic surgery between January 1, 2017, and December 30, 2020, at the General Hospital of Vienna, Austria. The data set contained information on patient demographics, vital signs, medication, and ventilation. The model was evaluated using an internal (n = 8113) and external test set (n = 5065) obtained from the openly accessible Vital Signs Database. Findings: In the internal test set, the mean absolute error for predicting mean arterial blood pressure was 0.376 standard deviations-or 4 mmHg-and 0.622 standard deviations-or 7 mmHg-in the external test set. We also adapted the TFT model to binarily predict the occurrence of hypotension as defined by mean arterial blood pressure < 65 mmHg in the next one, three, five, and 7 min. Here, model discrimination was excellent, with a mean area under the receiver operating characteristic curve (AUROC) of 0.933 in the internal test set and 0.919 in the external test set. Interpretation: Our TFT model is capable of accurately forecasting intraoperative arterial blood pressure using only low-resolution data showing a low prediction error. When used for binary prediction of hypotension, we obtained excellent performance. Funding: No external funding.

Severity Stratification of Coronary Artery Disease Using Novel Inner Ellipse-Based Foveal Avascular Zone Biomarkers.

Purpose: Given the similarities between the retinal and coronary microvasculature, the retina holds promising potential to serve as a non-invasive screening tool for coronary artery disease (CAD). We aimed to develop novel inner ellipse-based metrics and discern whether foveal avascular zone (FAZ) alterations can serve as indicators for CAD presence and severity. Methods: Patients admitted to the Department of Cardiology who underwent coronary angiography were included. This resulted in an inclusion of 212 patients, of which 73 had no CAD. During the same visit, 6 × 6-mm (nominal size) fovea-centered optical coherence tomography angiography images of both eyes were acquired. The Gensini score (GS) was utilized to quantify CAD severity. Six known FAZ shape metrics were assessed and three novel biomarkers based on the inner ellipse were defined: absolute inner ellipse difference, Hausdorff distance, and Chamfer distance. Results: Eight out of nine metrics showed significant associations with the GS in the left eye. However, significant differences across three CAD severity groups were only demonstrated by the novel metrics. Utilizing the Chamfer distance, age, and sex, patients with and without CAD could be distinguished with an average area under the curve (AUC) of 0.89 (95% confidence interval [CI], 0.84-0.95). Moreover, three CAD severity groups could be discerned with a macro average AUC of 0.77 (95% CI, 0.72-0.84). Conclusions: A comprehensive assessment of FAZ shape descriptors was performed, and a strong association with CAD was found. The inner ellipse-based biomarkers especially demonstrated high predictive abilities for CAD presence and severity.

Relationship between GPS-based community mobility data and orthopedic trauma admissions during the COVID-19 pandemic in Austria: a multicenter analysis.

OBJECTIVE: The main objective of this study was to examine the relationship between mobility patterns during the coronavirus disease 2019 (COVID-19) pandemic and orthopedic trauma patients in Austria. Utilizing global positioning system (GPS)-based mobility data, the attempt was to assess both the impact of COVID-19 lockdowns on reducing orthopedic trauma patients and the degree of compliance to the imposed movement restrictions. METHODS: This retrospective analysis included all patients (283,501) treated at 3 major level I trauma centers in Austria. Analyzed time periods were 1 January 2019 to 8 February 2021. Freely available GPS-based mobility data from Google and Apple Inc. was gathered. RESULTS: A moderate to strong correlation between the cumulative average outpatients and the assessed mobility index was observed for all cities (Google: r = 0.70 p < 0.001, 95% confidence interval, CI: 0.67-0.73; Apple: r = 0.64 p < 0.001, 95% CI: 0.61-0.67). A significant linear regression equation was found for Vienna (adjusted r2 = 0.48; F(1, 350) = 328,05; p < 0.01). During the first lockdown there was a drastic decline in mobility (up to -75.36%) and in numbers of orthopedic trauma outpatients (up to -64%, from 153 patients/day 2019 to 55 patients/day 2020) in comparison to the prepandemic era. The decline diminished as time passed. CONCLUSION: Analyses of GPS-based mobility patterns show a correlation with trauma patient numbers. These findings can be used to develop prediction models, leading to better resource planning and public health policy, enhancing patient care and cost-effectiveness, especially in the event of future pandemics. Furthermore, the results suggest that compliance to mobility restrictions decreased over time during the COVID-19 pandemic, resulting in increased mobility and trauma patients.

Feasibility of aluminum phantom radiography for osteoporosis detection in postmenopausal women with a fragility fracture of the distal radius compared to DXA and HR-pQCT.

Recently, promising results have been reported for detection of osteoporosis with use of an aluminum phantom. Therefore, the aim of this study was to evaluate the feasibility of radiography-based bone mineral density (BMD) measurement using a graded aluminum phantom. This study included 27 postmenopausal women with a distal radius fracture. Aluminum phantom radiography of the healthy radius was conducted as well as high-resolution peripheral quantitative computed tomography (HR-pQCT) measurement of the ultradistal radius and dual energy X-ray absorptiometry (DXA) of the radius, spine, and hip. A strong correlation was observed between aluminum phantom radiography-based mean gray value (mGV) and DXA-derived BMD, especially for the ultradistal radius (ρ = 0.75; p < 0.001). A moderate correlation for the femoral neck (ρ = 0.61 and p < 0.001) between modalities was found. Radius mGV and HR-pQCT-derived BMD only showed a moderate correlation (ρ = 0.48; p < 0.09). Aluminum phantom radiography might serve as a cost efficient, highly available, low-radiation dose screening, and diagnostic method for osteoporosis additively to DXA measurements. Especially, an application in areas with constrained DXA availability and such as preoperative trauma settings would be beneficial. However, further investigation and assessment of specificity and sensitivity is needed.

Predicting 30-Day and 180-Day Mortality in Elderly Proximal Hip Fracture Patients: Evaluation of 4 Risk Prediction Scores at a Level I Trauma Center.

This study evaluated the use of risk prediction models in estimating short- and mid-term mortality following proximal hip fracture in an elderly Austrian population. Data from 1101 patients who sustained a proximal hip fracture were retrospectively analyzed and applied to four models of interest: Physiological and Operative Severity Score for the enUmeration of Mortality and Morbidity (POSSUM), Charlson Comorbidity Index, Portsmouth-POSSUM and the American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP®) Risk Score. The performance of these models according to the risk prediction of short- and mid-term mortality was assessed with a receiver operating characteristic curve (ROC). The median age of participants was 83 years, and 69% were women. Six point one percent of patients were deceased by 30 days and 15.2% by 180 days postoperatively. There was no significant difference between the models; the ACS-NSQIP had the largest area under the receiver operating characteristic curve for within 30-day and 180-day mortality. Age, male gender, and hemoglobin (Hb) levels at admission <12.0 g/dL were identified as significant risk factors associated with a shorter time to death at 30 and 180 days postoperative (p < 0.001). Among the four scores, the ACS-NSQIP score could be best-suited clinically and showed the highest discriminative performance, although it was not specifically designed for the hip fracture population.