Refine
Document Type
- Article (5)
Has Fulltext
- yes (5)
Is part of the Bibliography
- no (5)
Keywords
- prediction (5) (remove)
Institute
Background: The aim of this study was to identify pre-operative parameters able to predict length of stay (LoS) based on clinical data and patient-reported outcome measures (PROMs) from a scorecard database in patients with significant aortic stenosis who underwent TAVI (transfemoral aortic valve implantation). Methods: 302 participants (51.7% males, age range 78.2–84.2 years.) were prospectively recruited. After computing the median LoS value (=6 days, range = 5–8 days), we implemented a decision tree algorithm by setting dichotomized values at median LoS as the dependent variable and assessed baseline clinical variables and PROMs (Clinical Frailty Scale (CFS), EuroQol-5 Dimension-5 Levels (EQ-5D) and Kansas City Cardiomyopathy Questionnaire (KCCQ)) as potential predictors. Results: Among clinical parameters, only peripheral arterial disease (p = 0.029, HR = 1.826) and glomerular filtration rate (GFR, cut-off < 33 mL/min/1.73 m2, p = 0.003, HR = 2.252) were predictive of LoS. Additionally, two PROMs (CFS; cut-off = 3, p < 0.001, HR = 1.324 and KCCQ; cut-off = 30, p = 0.003, HR = 2.274) were strong predictors. Further, a risk score for LoS (RS_LoS) was calculated based on these predictors. Patients with RS_LoS = 0 had a median LoS of 5 days; patients RS_LoS ≥ 3 had a median LoS of 8 days. Conclusions: based on the pre-operative values of the above four predictors, a personalized prediction of LoS after TAVI can be achieved.
Background and Objectives: We tested if a novel combination of predictors could improve the accuracy of outcome prediction after transfemoral transcatheter aortic valve implantation (TAVI). Materials and Methods: This prospective study recruited 169 participants (49% female; median age 81 years). The primary endpoint was midterm mortality; secondary endpoints were acute Valve Academic Research Consortium (VARC)-3 complication rate and post-TAVI in-hospital length of stay (LoS). EuroSCORE II (ESII), comorbidities (e.g., coronary artery disease), eGFR (estimated glomerular filtration rate; based on cystatin C), hemoglobin, creatinine, N-Terminal pro-Brain Natriuretic Peptide (NTproBNP) levels and patient-reported outcome measures (PROMs, namely EuroQol-5-Dimension-5-Levels, EQ5D5L; Kansas City Cardiomyopathy Questionnaire, KCCQ; clinical frailty scale, CFS) at baseline were tested as predictors. Regression (uni- and multi-variate Cox; linear; binary logistic) and receiver operating characteristic (ROC)-curve analysis were applied. Results: Within a median follow-up of 439 (318–585) days, 12 participants died (7.1%). Independent predictors of mortality using multivariate Cox regression were baseline eGFR (p = 0.001) and KCCQ (p = 0.037). Based on these predictors, a Linear Prediction Score (LPS1) was calculated. The LPS1-area under the curve (AUC)-value (0.761) was significantly higher than the ESII-AUC value (0.597; p = 0.035). Independent predictors for LoS > 6 days (the median LoS) were eGFR (p = 0.028), NTproBNP (p = 0.034), and EQ5D5L values (p = 0.002); a respective calculated LPS2 provided an AUC value of 0.677 (p < 0.001). Eighty participants (47.3%) experienced complications. Male sex predicted complications only in the univariate analysis. Conclusions: The combination of KCCQ and eGFR can better predict midterm mortality than ES II alone. Combining eGFR, NTproBNP, and EQ5D5L can reliably predict LoS after TAVI. This novel method improves personalized TAVI risk stratification and hence may help reduce post-TAVI risk.
Optimal investment decisions by institutional investors require accurate predictions with respect to the development of stock markets. Motivated by previous research that revealed the unsatisfactory performance of existing stock market prediction models, this study proposes a novel prediction approach. Our proposed system combines Artificial Intelligence (AI) with data from Virtual Investment Communities (VICs) and leverages VICs’ ability to support the process of predicting stock markets. An empirical study with two different models using real data shows the potential of the AI-based system with VICs information as an instrument for stock market predictions. VICs can be a valuable addition but our results indicate that this type of data is only helpful in certain market phases.
Most human actions produce concomitant sounds. Action sounds can be either part of the action goal (GAS, goal-related action sounds), as for instance in tap dancing, or a mere by-product of the action (BAS, by-product action sounds), as for instance in hurdling. It is currently unclear whether these two types of action sounds—incidental or intentional—differ in their neural representation and whether the impact on the performance evaluation of an action diverges between the two. We here examined whether during the observation of tap dancing compared to hurdling, auditory information is a more important factor for positive action quality ratings. Moreover, we tested whether observation of tap dancing vs. hurdling led to stronger attenuation in primary auditory cortex, and a stronger mismatch signal when sounds do not match our expectations. We recorded individual point-light videos of newly trained participants performing tap dancing and hurdling. In the subsequent functional magnetic resonance imaging (fMRI) session, participants were presented with the videos that displayed their own actions, including corresponding action sounds, and were asked to rate the quality of their performance. Videos were either in their original form or scrambled regarding the visual modality, the auditory modality, or both. As hypothesized, behavioral results showed significantly lower rating scores in the GAS condition compared to the BAS condition when the auditory modality was scrambled. Functional MRI contrasts between BAS and GAS actions revealed higher activation of primary auditory cortex in the BAS condition, speaking in favor of stronger attenuation in GAS, as well as stronger activation of posterior superior temporal gyri and the supplementary motor area in GAS. Results suggest that the processing of self-generated action sounds depends on whether we have the intention to produce a sound with our action or not, and action sounds may be more prone to be used as sensory feedback when they are part of the explicit action goal. Our findings contribute to a better understanding of the function of action sounds for learning and controlling sound-producing actions.