The value of error-correcting responses for cognitive assessment in games DOI Creative Commons
Benny Markovitch, Nathan J. Evans, Max V. Birk

и другие.

Scientific Reports, Год журнала: 2024, Номер 14(1)

Опубликована: Сен. 4, 2024

Язык: Английский

Comparing cognition in parents with schizophrenia or bipolar disorder and their 7-year-old offspring DOI Creative Commons
Aja Greve, Nicoline Hemager, Erik Lykke Mortensen

и другие.

Psychiatry Research, Год журнала: 2024, Номер 340, С. 116112 - 116112

Опубликована: Июль 29, 2024

Individuals with schizophrenia (SZ) or bipolar disorder (BP) display cognitive impairments, while their first-degree relatives perform at an intermediate level between the patient groups and controls. However, environmental impact of having ill relative likely varies type kinship some studies suggest that offspring may be particularly disadvantaged. The present study aimed to investigate relationship parent child cognition in parents SZ BD 7-year-old offspring. A population-based cohort 522 children (parental SZ, n = 202; parental BP, 120; controls, 200) underwent same assessment battery covering a wide range functions. We used Bayesian statistics model performance. found performance on non-verbal tests was better than using controls as reference. for verbal tests, there little no evidence this pattern even opposite BP group: relatively findings disadvantaged abilities. Future will show whether persists throughout development.

Язык: Английский

Процитировано

0

Modeling uncertainty in individual predictions of cognitive functioning for untreated glioma patients using Bayesian regression DOI Creative Commons
Sander Martijn Boelders, Bruno Nicenboim, Eric Postma

и другие.

medRxiv (Cold Spring Harbor Laboratory), Год журнала: 2024, Номер unknown

Опубликована: Окт. 8, 2024

Introduction Cognitive impairments of patients with a glioma are increasingly considered when making treatment decisions considering personalized onco-functional balance. Predicting cognitive functioning before surgery can serve as steppingstone for the clinical goal predicting after surgery. However, in previous study, machine-learning models could not reliably predict using comprehensive set variables. The current study aims to improve predictions while uncertainty individual explicit. Method Pre-operative was predicted 340 across eight tests. This done six multivariate Bayesian regression following approach Four included interactions with- or multilevel structure over histopathological diagnosis. Point-wise were compared coefficient determination (R 2 ) and best-performing model interpreted. Results outperformed benefitted from shrinkage priors. R ranged between 0.3% 21.5% median tests 7.2%. Estimated errors prediction high. allowed parameters differ diagnoses pulling them toward population mean. Conclusion providing estimates predictions. Despite this, pre-operative variables remained Consequently, clinicians should infer these Different best treated distinct yet related. Highlights models. Predictions uncertain despite improvements. Importance serves is important two reasons. First, it demonstrates that popular Second, explicitly shows based on readily available uncertain. Last, may benefit multifaceted view treating different

Язык: Английский

Процитировано

0

Predicting cognitive function three months after surgery in patients with a glioma DOI Creative Commons
Sander Martijn Boelders, Bruno Nicenboim,

Elke Butterbrod

и другие.

medRxiv (Cold Spring Harbor Laboratory), Год журнала: 2024, Номер unknown

Опубликована: Окт. 8, 2024

Introduction Patients with a glioma often suffer from cognitive impairments both before and after anti-tumor treatment. Ideally, clinicians can rely on predictions of post-operative functioning for individual patients based information obtainable surgery. Such would facilitate selecting the optimal treatment considering patients’ onco-functional balance. Method Cognitive three months surgery was predicted 317 across eight tests. Nine multivariate Bayesian regression models were used following machine-learning approach while employing pre-operative neuropsychological test scores comprehensive set clinical predictors Model performances compared using Expected Log Pointwise Predictive Density (ELPD), pointwise assessed Coefficient Determination (R²) Mean Absolute Error. Models against only best-performing model interpreted. Moreover, an example prediction including uncertainty use provided. Results The obtained median R² 34.20%. Individual predictions, however, uncertain. Pre-operative most influential predictor. performed similarly to those (ΔELPD 14.4±10.0, ΔR² −0.53%.). Conclusion Post-operative cannot yet reliably be included predictors. relied strongly functioning. Consequently, should not infer it stresses need collect larger cross-center multimodal datasets obtain more certain patients. Importance study able that is balance could improve patient counseling. First, our shows predictors, being important Second, results demonstrate how resulting models, their estimates, may ultimately in practice. Third, show importance collecting additional stress datasets. Key points - predictor Additional are needed

Язык: Английский

Процитировано

0

The value of error-correcting responses for cognitive assessment in games DOI Creative Commons
Benny Markovitch, Nathan J. Evans, Max V. Birk

и другие.

Scientific Reports, Год журнала: 2024, Номер 14(1)

Опубликована: Сен. 4, 2024

Язык: Английский

Процитировано

0