Refine
Year of publication
Document Type
- Article (30)
- Preprint (2)
- Conference Proceeding (1)
Has Fulltext
- yes (33)
Is part of the Bibliography
- no (33)
Keywords
- Antimicrobial treatment (2)
- Biofilm (2)
- Complications (2)
- Deep brain stimulation (2)
- Low-virulent infection (2)
- Sonication (2)
- ARDS (1)
- Affective disorders (1)
- Biomarker (1)
- Bipolar disorder (1)
- Brain tumors (1)
- COVID-19 (1)
- Cancer (1)
- Children and adolescents (1)
- Comparative analysis (1)
- Critical care (1)
- Deep neural network (1)
- Diagnostic test (1)
- Differential diagnosis (1)
- Electron-pion identification (1)
- Experimental models of disease (1)
- Fibre/foam sandwich radiator (1)
- Gaussian Process (1)
- Glioblastoma (1)
- Gradient boosting (1)
- Ionisation energy loss (1)
- Leukemia (1)
- Lithium (1)
- MGMT (1)
- Multi-syndrome classification (1)
- Multi-wire proportional drift chamber (1)
- Multiple Kernel (1)
- Multiplex immunoassay (1)
- Neural network (1)
- Neurodegenerative syndromes (1)
- Obesity (1)
- Outcome (1)
- Patient reported outcomes (1)
- Prognostic models (1)
- Random forest (1)
- Randomized controlled trial (1)
- Sarcomas (1)
- Social participation (1)
- Suicidal behavior (1)
- Suicidal thoughts (1)
- Support Vector Regression (1)
- Support Vector Regression Model (1)
- Support vector machine (1)
- TR (1)
- Temozolomide (1)
- Thrombosis (1)
- Tracking (1)
- Transition radiation detector (1)
- Trigger (1)
- Virtual Screening (1)
- Xenon-based gas mixture (1)
- anaesthesia in orthopaedics (1)
- anaesthetics (1)
- cancer (1)
- dE/dx (1)
- geriatric medicine (1)
- pancreatic cancer (1)
Highlights
• A panel of 20 biomarkers was identified capable of differentiating BD patients from controls.
• Excellent discrimination between established BD patients and controls.
• Good to excellent discrimination between misdiagnosed BD patients and first onset MDD patients.
• Fair to good discrimination between pre-diagnostic BD patients and controls.
• Study demonstrates the potential utility of a protein biomarker panel as a diagnostic test for BD.
Abstract
Background: Bipolar disorder (BD) is a costly, devastating and life shortening mental disorder that is often misdiagnosed, especially on initial presentation. Misdiagnosis frequently results in ineffective treatment. We investigated the utility of a biomarker panel as a diagnostic test for BD.
Methods and findings: We performed a meta-analysis of eight case-control studies to define a diagnostic biomarker panel for BD. After validating the panel on established BD patients, we applied it to undiagnosed BD patients. We analysed 249 BD, 122 pre-diagnostic BD, 75 pre-diagnostic schizophrenia and 90 first onset major depression disorder (MDD) patients and 371 controls. The biomarker panel was identified using ten-fold cross-validation with lasso regression applied to the 87 analytes available across the meta-analysis studies.
We identified 20 protein analytes with excellent predictive performance [area under the curve (AUC) ⩾ 0.90]. Importantly, the panel had a good predictive performance (AUC 0.84) to differentiate 12 misdiagnosed BD patients from 90 first onset MDD patients, and a fair to good predictive performance (AUC 0.79) to differentiate between 110 pre-diagnostic BD patients and 184 controls. We also demonstrated the disease specificity of the panel.
Conclusions: An early and accurate diagnosis has the potential to delay or even prevent the onset of BD. This study demonstrates the potential utility of a biomarker panel as a diagnostic test for BD.
Background: Intensive Care Resources are heavily utilized during the COVID-19 pandemic. However, risk stratification and prediction of SARS-CoV-2 patient clinical outcomes upon ICU admission remain inadequate. This study aimed to develop a machine learning model, based on retrospective & prospective clinical data, to stratify patient risk and predict ICU survival and outcomes. Methods: A Germany-wide electronic registry was established to pseudonymously collect admission, therapeutic and discharge information of SARS-CoV-2 ICU patients retrospectively and prospectively. Machine learning approaches were evaluated for the accuracy and interpretability of predictions. The Explainable Boosting Machine approach was selected as the most suitable method. Individual, non-linear shape functions for predictive parameters and parameter interactions are reported. Results: 1039 patients were included in the Explainable Boosting Machine model, 596 patients retrospectively collected, and 443 patients prospectively collected. The model for prediction of general ICU outcome was shown to be more reliable to predict “survival”. Age, inflammatory and thrombotic activity, and severity of ARDS at ICU admission were shown to be predictive of ICU survival. Patients’ age, pulmonary dysfunction and transfer from an external institution were predictors for ECMO therapy. The interaction of patient age with D-dimer levels on admission and creatinine levels with SOFA score without GCS were predictors for renal replacement therapy. Conclusions: Using Explainable Boosting Machine analysis, we confirmed and weighed previously reported and identified novel predictors for outcome in critically ill COVID-19 patients. Using this strategy, predictive modeling of COVID-19 ICU patient outcomes can be performed overcoming the limitations of linear regression models. Trial registration “ClinicalTrials” (clinicaltrials.gov) under NCT04455451.
Importance: The entry of artificial intelligence into medicine is pending. Several methods have been used for the predictions of structured neuroimaging data, yet nobody compared them in this context.
Objective: Multi-class prediction is key for building computational aid systems for differential diagnosis. We compared support vector machine, random forest, gradient boosting, and deep feed-forward neural networks for the classification of different neurodegenerative syndromes based on structural magnetic resonance imaging.
Design, setting, and participants: Atlas-based volumetry was performed on multi-centric T1-weighted MRI data from 940 subjects, i.e., 124 healthy controls and 816 patients with ten different neurodegenerative diseases, leading to a multi-diagnostic multi-class classification task with eleven different classes.
Interventions: N.A.
Main outcomes and measures: Cohen’s kappa, accuracy, and F1-score to assess model performance.
Results: Overall, the neural network produced both the best performance measures and the most robust results. The smaller classes however were better classified by either the ensemble learning methods or the support vector machine, while performance measures for small classes were comparatively low, as expected. Diseases with regionally specific and pronounced atrophy patterns were generally better classified than diseases with widespread and rather weak atrophy.
Conclusions and relevance: Our study furthermore underlines the necessity of larger data sets but also calls for a careful consideration of different machine learning methods that can handle the type of data and the classification task best.