Egill Axfjord Fridgeirsson
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Egill Axfjord Fridgeirsson has a total of 72 co-authors. Here are the top co-authors by number of shared publications:
- Jenna Marie Reps
Janssen Research and Development, Raritan, USA
5 shared publications Damiaan Denys
University of Amsterdam
4 shared publications- Peter R. Rijnbeek
Department of Medical Informatics, Erasmus University Medical Center
4 shared publications - Guido Alexander van Wingen
Amsterdam UMC
3 shared publications - Jan A. Kors
Department of Medical Informatics Erasmus University Medical Center Rotterdam The Netherlands
3 shared publications - Martijn Figee
Nash Family Center for Advanced Circuit Therapeutics, Icahn School of Medicine at Mount Sinai, New York, USA
3 shared publications
Latest Publications
- Open Access
- 5 November 2026
- Published by Cambridge University Press (CUP) in Psychological Medicine
- Vol. 56, e354
Network-based virtual perturbation for modeling DBS-related network effects in obsessive-compulsive and major depressive disorders
Background: Deep brain stimulation (DBS) efficacy for psychiatric disorders remains heterogeneous due to incomplete mechanistic insights and insufficient individualized treatment planning. A biologically grounded computational framework is urgently needed to simulate DBS-related differences for treatment-resistant patients. Methods: Resting-state fMRI data from 11 patients with obsessive-compulsive disorder (OCD) and 11 with major depressive disorder (MDD), scanned in DBS-ON and DBS-OFF states, were used to derive functional connectivity (FC) via network deconvolution. Whole-brain dynamic modeling incorporated stimulation targets, amplitudes, and frequencies to simulate DBS effects in silico. Results: The model recapitulated syndrome-specific DBS-related network patterns in exploratory analyses. OCD stimulation primarily altered frontal-subcortical connectivity, whereas MDD stimulation predominantly affected default mode network hubs ( p < 0.05). Under leave-one-out cross-validation, correlations between predicted and observed clinical outcomes were r = 0.625 in OCD ( p = 0.040) and r = 0.659 in MDD ( p = 0.028). Applying the same FC signatures to simulated FC yielded r = 0.465 in OCD ( p = 0.149) and r = 0.627 in MDD ( p = 0.039). In a proof-of-concept application to a separate OCD cohort ( N = 92), three individuals showed high model-predicted NAc-response profiles and predicted improvement exceeding 35%. Conclusions: This hypothesis-generating framework retrospectively models DBS-related network effects using individualized resting-state FC and virtual perturbations. Linking simulated FC states with empirical DBS-ON/OFF patterns may help characterize how baseline functional architecture shapes DBS-related network responses and inform future individualized DBS modeling.
- 5 November 2026
- Published by Cambridge University Press (CUP) in Psychological Medicine
- Vol. 56, e354
Open AccessNetwork-based virtual perturbation for modeling DBS-related network effects in obsessive-compulsive and major depressive disorders
Background: Deep brain stimulation (DBS) efficacy for psychiatric disorders remains heterogeneous due to incomplete mechanistic insights and insufficient individualized treatment planning. A biologically grounded computational framework is urgently needed to simulate DBS-related differences for treatment-resistant patients. Methods: Resting-state fMRI data from 11 patients with obsessive-compulsive disorder (OCD) and 11 with major depressive disorder (MDD), scanned in DBS-ON and DBS-OFF states, were used to derive functional connectivity (FC) via network deconvolution. Whole-brain dynamic modeling incorporated stimulation targets, amplitudes, and frequencies to simulate DBS effects in silico. Results: The model recapitulated syndrome-specific DBS-related network patterns in exploratory analyses. OCD stimulation primarily altered frontal-subcortical connectivity, whereas MDD stimulation predominantly affected default mode network hubs ( p < 0.05). Under leave-one-out cross-validation, correlations between predicted and observed clinical outcomes were r = 0.625 in OCD ( p = 0.040) and r = 0.659 in MDD ( p = 0.028). Applying the same FC signatures to simulated FC yielded r = 0.465 in OCD ( p = 0.149) and r = 0.627 in MDD ( p = 0.039). In a proof-of-concept application to a separate OCD cohort ( N = 92), three individuals showed high model-predicted NAc-response profiles and predicted improvement exceeding 35%. Conclusions: This hypothesis-generating framework retrospectively models DBS-related network effects using individualized resting-state FC and virtual perturbations. Linking simulated FC states with empirical DBS-ON/OFF patterns may help characterize how baseline functional architecture shapes DBS-related network responses and inform future individualized DBS modeling.
- Open Access
- 25 June 2026
- Published by Association for Computing Machinery (ACM)
- p. 3302-3321
Estimating the Sample Size Required to Minimize Model Uncertainty Across Different Risk Profiles
- 25 June 2026
- Published by Association for Computing Machinery (ACM)
- p. 3302-3321
Open AccessEstimating the Sample Size Required to Minimize Model Uncertainty Across Different Risk Profiles
- Open Access
- 8 August 2025
- Published by Cambridge University Press (CUP) in Psychological Medicine
- Vol. 55, e231
Deep brain stimulation modulates directional limbic connectivity in major depressive disorder
Background: Deep brain stimulation (DBS) is being investigated as a treatment for patients with refractory major depressive disorder (MDD). However, little is known about how DBS exerts its antidepressive effects. Here, we investigated whether ventral anterior limb of the internal capsule stimulation modulates a limbic network centered around the amygdala in patients with treatment-resistant MDD. Methods: Nine patients underwent resting state functional magnetic resonance scans before DBS surgery and after 1 year of treatment. In addition, they were scanned twice within 2 weeks during the subsequent double-blind cross-over phase with active and sham treatment. Twelve matched controls underwent scans at the same time intervals to account for test–retest effects. The imaging data were investigated with functional connectivity (FC) analysis and dynamic causal modelling. Results: Results showed that 1 year of DBS treatment was associated with increased FC of the left amygdala with precentral cortex and left insula, along with decreased bilateral connectivity between nucleus accumbens and ventromedial prefrontal cortex. No changes in FC were observed during the cross-over phase. Effective connectivity analyses using dynamic causal models revealed widespread amygdala-centric changes between presurgery and 1 year follow-up, while the cross-over phase was associated with insula-centric changes between active and sham stimulation. Conclusions: These results suggest that ventral anterior limb of the internal capsule DBS results in complex rebalancing of the limbic network involved in emotion, reward, and interoceptive processing.
- 8 August 2025
- Published by Cambridge University Press (CUP) in Psychological Medicine
- Vol. 55, e231
Open AccessDeep brain stimulation modulates directional limbic connectivity in major depressive disorder
Background: Deep brain stimulation (DBS) is being investigated as a treatment for patients with refractory major depressive disorder (MDD). However, little is known about how DBS exerts its antidepressive effects. Here, we investigated whether ventral anterior limb of the internal capsule stimulation modulates a limbic network centered around the amygdala in patients with treatment-resistant MDD. Methods: Nine patients underwent resting state functional magnetic resonance scans before DBS surgery and after 1 year of treatment. In addition, they were scanned twice within 2 weeks during the subsequent double-blind cross-over phase with active and sham treatment. Twelve matched controls underwent scans at the same time intervals to account for test–retest effects. The imaging data were investigated with functional connectivity (FC) analysis and dynamic causal modelling. Results: Results showed that 1 year of DBS treatment was associated with increased FC of the left amygdala with precentral cortex and left insula, along with decreased bilateral connectivity between nucleus accumbens and ventromedial prefrontal cortex. No changes in FC were observed during the cross-over phase. Effective connectivity analyses using dynamic causal models revealed widespread amygdala-centric changes between presurgery and 1 year follow-up, while the cross-over phase was associated with insula-centric changes between active and sham stimulation. Conclusions: These results suggest that ventral anterior limb of the internal capsule DBS results in complex rebalancing of the limbic network involved in emotion, reward, and interoceptive processing.
- Open Access
Comparison of deep learning and conventional methods for disease onset prediction
Background: Conventional prediction methods such as logistic regression and gradient boosting have been widely utilized for disease onset prediction for their reliability and interpretability. Deep learning methods promise enhanced prediction performance by extracting complex patterns from clinical data, but face challenges like data sparsity and high dimensionality. Methods: This study compares conventional and deep learning approaches to predict lung cancer, dementia, and bipolar disorder using observational data from eleven databases from North America, Europe, and Asia. Models were developed using logistic regression, gradient boosting, ResNet, and Transformer, and validated both internally and externally across the data sources. Discrimination performance was assessed using AUROC, and calibration was evaluated using Eavg. Findings: Across 11 datasets, conventional methods generally outperformed deep learning methods in terms of discrimination performance, particularly during external validation, highlighting their better transportability. Learning curves suggest that deep learning models require substantially larger datasets to reach the same performance levels as conventional methods. Calibration performance was also better for conventional methods, with ResNet showing the poorest calibration. Interpretation: Despite the potential of deep learning models to capture complex patterns in structured observational healthcare data, conventional models remain highly competitive for disease onset prediction, especially in scenarios involving smaller datasets and if lengthy training times need to be avoided. The study underscores the need for future research focused on optimizing deep learning models to handle the sparsity, high dimensionality, and heterogeneity inherent in healthcare datasets, and find new strategies to exploit the full capabilities of deep learning methods.
Open AccessComparison of deep learning and conventional methods for disease onset prediction
Background: Conventional prediction methods such as logistic regression and gradient boosting have been widely utilized for disease onset prediction for their reliability and interpretability. Deep learning methods promise enhanced prediction performance by extracting complex patterns from clinical data, but face challenges like data sparsity and high dimensionality. Methods: This study compares conventional and deep learning approaches to predict lung cancer, dementia, and bipolar disorder using observational data from eleven databases from North America, Europe, and Asia. Models were developed using logistic regression, gradient boosting, ResNet, and Transformer, and validated both internally and externally across the data sources. Discrimination performance was assessed using AUROC, and calibration was evaluated using Eavg. Findings: Across 11 datasets, conventional methods generally outperformed deep learning methods in terms of discrimination performance, particularly during external validation, highlighting their better transportability. Learning curves suggest that deep learning models require substantially larger datasets to reach the same performance levels as conventional methods. Calibration performance was also better for conventional methods, with ResNet showing the poorest calibration. Interpretation: Despite the potential of deep learning models to capture complex patterns in structured observational healthcare data, conventional models remain highly competitive for disease onset prediction, especially in scenarios involving smaller datasets and if lengthy training times need to be avoided. The study underscores the need for future research focused on optimizing deep learning models to handle the sparsity, high dimensionality, and heterogeneity inherent in healthcare datasets, and find new strategies to exploit the full capabilities of deep learning methods.
- Open Access
- 20 May 2024
- Published by Oxford University Press (OUP) in Journal of the American Medical Informatics Association
- Vol. 31 (7) , 1514-1521
Comparing penalization methods for linear models on large observational health data
Objective: This study evaluates regularization variants in logistic regression (L1, L2, ElasticNet, Adaptive L1, Adaptive ElasticNet, Broken adaptive ridge [BAR], and Iterative hard thresholding [IHT]) for discrimination and calibration performance, focusing on both internal and external validation. Materials and Methods: We use data from 5 US claims and electronic health record databases and develop models for various outcomes in a major depressive disorder patient population. We externally validate all models in the other databases. We use a train-test split of 75%/25% and evaluate performance with discrimination and calibration. Statistical analysis for difference in performance uses Friedman’s test and critical difference diagrams. Results: Of the 840 models we develop, L1 and ElasticNet emerge as superior in both internal and external discrimination, with a notable AUC difference. BAR and IHT show the best internal calibration, without a clear external calibration leader. ElasticNet typically has larger model sizes than L1. Methods like IHT and BAR, while slightly less discriminative, significantly reduce model complexity. Conclusion: L1 and ElasticNet offer the best discriminative performance in logistic regression for healthcare predictions, maintaining robustness across validations. For simpler, more interpretable models, L0-based methods (IHT and BAR) are advantageous, providing greater parsimony and calibration with fewer features. This study aids in selecting suitable regularization techniques for healthcare prediction models, balancing performance, complexity, and interpretability.
- 20 May 2024
- Published by Oxford University Press (OUP) in Journal of the American Medical Informatics Association
- Vol. 31 (7) , 1514-1521
Open AccessComparing penalization methods for linear models on large observational health data
Objective: This study evaluates regularization variants in logistic regression (L1, L2, ElasticNet, Adaptive L1, Adaptive ElasticNet, Broken adaptive ridge [BAR], and Iterative hard thresholding [IHT]) for discrimination and calibration performance, focusing on both internal and external validation. Materials and Methods: We use data from 5 US claims and electronic health record databases and develop models for various outcomes in a major depressive disorder patient population. We externally validate all models in the other databases. We use a train-test split of 75%/25% and evaluate performance with discrimination and calibration. Statistical analysis for difference in performance uses Friedman’s test and critical difference diagrams. Results: Of the 840 models we develop, L1 and ElasticNet emerge as superior in both internal and external discrimination, with a notable AUC difference. BAR and IHT show the best internal calibration, without a clear external calibration leader. ElasticNet typically has larger model sizes than L1. Methods like IHT and BAR, while slightly less discriminative, significantly reduce model complexity. Conclusion: L1 and ElasticNet offer the best discriminative performance in logistic regression for healthcare predictions, maintaining robustness across validations. For simpler, more interpretable models, L0-based methods (IHT and BAR) are advantageous, providing greater parsimony and calibration with fewer features. This study aids in selecting suitable regularization techniques for healthcare prediction models, balancing performance, complexity, and interpretability.
- Open Access
Using clinical text to refine unspecific condition codes in Dutch general practitioner EHR data
Objective Observational studies using electronic health record (EHR) databases often face challenges due to unspecific clinical codes that can obscure detailed medical information, hindering precise data analysis. In this study, we aimed to assess the feasibility of refining these unspecific condition codes into more specific codes in a Dutch general practitioner (GP) EHR database by leveraging the available clinical free text. Methods We utilized three approaches for text classification—search queries, semi-supervised learning, and supervised learning—to improve the specificity of ten unspecific International Classification of Primary Care (ICPC-1) codes. Two text representations and three machine learning algorithms were evaluated for the (semi-)supervised models. Additionally, we measured the improvement achieved by the refinement process on all code occurrences in the database. Results The classification models performed well for most codes. In general, no single classification approach consistently outperformed the others. However, there were variations in the relative performance of the classification approaches within each code and in the use of different text representations and machine learning algorithms. Class imbalance and limited training data affected the performance of the (semi-)supervised models, yet the simple search queries remained particularly effective. Ultimately, the developed models improved the specificity of over half of all the unspecific code occurrences in the database. Conclusions Our findings show the feasibility of using information from clinical text to improve the specificity of unspecific condition codes in observational healthcare databases, even with a limited range of machine-learning techniques and modest annotated training sets. Future work could investigate transfer learning, integration of structured data, alternative semi-supervised methods, and validation of models across healthcare settings. The improved level of detail enriches the interpretation of medical information and can benefit observational research and patient care.
Open AccessUsing clinical text to refine unspecific condition codes in Dutch general practitioner EHR data
Objective Observational studies using electronic health record (EHR) databases often face challenges due to unspecific clinical codes that can obscure detailed medical information, hindering precise data analysis. In this study, we aimed to assess the feasibility of refining these unspecific condition codes into more specific codes in a Dutch general practitioner (GP) EHR database by leveraging the available clinical free text. Methods We utilized three approaches for text classification—search queries, semi-supervised learning, and supervised learning—to improve the specificity of ten unspecific International Classification of Primary Care (ICPC-1) codes. Two text representations and three machine learning algorithms were evaluated for the (semi-)supervised models. Additionally, we measured the improvement achieved by the refinement process on all code occurrences in the database. Results The classification models performed well for most codes. In general, no single classification approach consistently outperformed the others. However, there were variations in the relative performance of the classification approaches within each code and in the use of different text representations and machine learning algorithms. Class imbalance and limited training data affected the performance of the (semi-)supervised models, yet the simple search queries remained particularly effective. Ultimately, the developed models improved the specificity of over half of all the unspecific code occurrences in the database. Conclusions Our findings show the feasibility of using information from clinical text to improve the specificity of unspecific condition codes in observational healthcare databases, even with a limited range of machine-learning techniques and modest annotated training sets. Future work could investigate transfer learning, integration of structured data, alternative semi-supervised methods, and validation of models across healthcare settings. The improved level of detail enriches the interpretation of medical information and can benefit observational research and patient care.
- Open Access
- 7 December 2023
- Published by Springer Nature in BMC Medical Research Methodology
- Vol. 23 (1) , 1-10
Attention-based neural networks for clinical prediction modelling on electronic health records
Deep learning models have had a lot of success in various fields. However, on structured data they have struggled. Here we apply four state-of-the-art supervised deep learning models using the attention mechanism and compare against logistic regression and XGBoost using discrimination, calibration and clinical utility. We develop the models using a general practitioners database. We implement a recurrent neural network, a transformer with and without reverse distillation and a graph neural network. We measure discrimination using the area under the receiver operating characteristic curve (AUC) and the area under the precision recall curve (AUPRC). We assess smooth calibration using restricted cubic splines and clinical utility with decision curve analysis. Our results show that deep learning approaches can improve discrimination up to 2.5% points AUC and 7.4% points AUPRC. However, on average the baselines are competitive. Most models are similarly calibrated as the baselines except for the graph neural network. The transformer using reverse distillation shows the best performance in clinical utility on two out of three prediction problems over most of the prediction thresholds. In this study, we evaluated various approaches in supervised learning using neural networks and attention. Here we do a rigorous comparison, not only looking at discrimination but also calibration and clinical utility. There is value in using deep learning models on electronic health record data since it can improve discrimination and clinical utility while providing good calibration. However, good baseline methods are still competitive.
- 7 December 2023
- Published by Springer Nature in BMC Medical Research Methodology
- Vol. 23 (1) , 1-10
Open AccessAttention-based neural networks for clinical prediction modelling on electronic health records
Deep learning models have had a lot of success in various fields. However, on structured data they have struggled. Here we apply four state-of-the-art supervised deep learning models using the attention mechanism and compare against logistic regression and XGBoost using discrimination, calibration and clinical utility. We develop the models using a general practitioners database. We implement a recurrent neural network, a transformer with and without reverse distillation and a graph neural network. We measure discrimination using the area under the receiver operating characteristic curve (AUC) and the area under the precision recall curve (AUPRC). We assess smooth calibration using restricted cubic splines and clinical utility with decision curve analysis. Our results show that deep learning approaches can improve discrimination up to 2.5% points AUC and 7.4% points AUPRC. However, on average the baselines are competitive. Most models are similarly calibrated as the baselines except for the graph neural network. The transformer using reverse distillation shows the best performance in clinical utility on two out of three prediction problems over most of the prediction thresholds. In this study, we evaluated various approaches in supervised learning using neural networks and attention. Here we do a rigorous comparison, not only looking at discrimination but also calibration and clinical utility. There is value in using deep learning models on electronic health record data since it can improve discrimination and clinical utility while providing good calibration. However, good baseline methods are still competitive.
- Open Access
- 20 July 2023
- Published by JMIR Publications Inc. in Journal of Medical Internet Research
- Vol. 25, e46165
Privacy-Preserving Federated Model Predicting Bipolar Transition in Patients With Depression: Prediction Model Development Study
Background: Mood disorder has emerged as a serious concern for public health; in particular, bipolar disorder has a less favorable prognosis than depression. Although prompt recognition of depression conversion to bipolar disorder is needed, early prediction is challenging due to overlapping symptoms. Recently, there have been attempts to develop a prediction model by using federated learning. Federated learning in medical fields is a method for training multi-institutional machine learning models without patient-level data sharing. Objective: This study aims to develop and validate a federated, differentially private multi-institutional bipolar transition prediction model. Methods: This retrospective study enrolled patients diagnosed with the first depressive episode at 5 tertiary hospitals in South Korea. We developed models for predicting bipolar transition by using data from 17,631 patients in 4 institutions. Further, we used data from 4541 patients for external validation from 1 institution. We created standardized pipelines to extract large-scale clinical features from the 4 institutions without any code modification. Moreover, we performed feature selection in a federated environment for computational efficiency and applied differential privacy to gradient updates. Finally, we compared the federated and the 4 local models developed with each hospital's data on internal and external validation data sets. Results: In the internal data set, 279 out of 17,631 patients showed bipolar disorder transition. In the external data set, 39 out of 4541 patients showed bipolar disorder transition. The average performance of the federated model in the internal test (area under the curve [AUC] 0.726) and external validation (AUC 0.719) data sets was higher than that of the other locally developed models (AUC 0.642-0.707 and AUC 0.642-0.699, respectively). In the federated model, classifications were driven by several predictors such as the Charlson index (low scores were associated with bipolar transition, which may be due to younger age), severe depression, anxiolytics, young age, and visiting months (the bipolar transition was associated with seasonality, especially during the spring and summer months). Conclusions: We developed and validated a differentially private federated model by using distributed multi-institutional psychiatric data with standardized pipelines in a real-world environment. The federated model performed better than models using local data only.
- 20 July 2023
- Published by JMIR Publications Inc. in Journal of Medical Internet Research
- Vol. 25, e46165
Open AccessPrivacy-Preserving Federated Model Predicting Bipolar Transition in Patients With Depression: Prediction Model Development Study
Background: Mood disorder has emerged as a serious concern for public health; in particular, bipolar disorder has a less favorable prognosis than depression. Although prompt recognition of depression conversion to bipolar disorder is needed, early prediction is challenging due to overlapping symptoms. Recently, there have been attempts to develop a prediction model by using federated learning. Federated learning in medical fields is a method for training multi-institutional machine learning models without patient-level data sharing. Objective: This study aims to develop and validate a federated, differentially private multi-institutional bipolar transition prediction model. Methods: This retrospective study enrolled patients diagnosed with the first depressive episode at 5 tertiary hospitals in South Korea. We developed models for predicting bipolar transition by using data from 17,631 patients in 4 institutions. Further, we used data from 4541 patients for external validation from 1 institution. We created standardized pipelines to extract large-scale clinical features from the 4 institutions without any code modification. Moreover, we performed feature selection in a federated environment for computational efficiency and applied differential privacy to gradient updates. Finally, we compared the federated and the 4 local models developed with each hospital's data on internal and external validation data sets. Results: In the internal data set, 279 out of 17,631 patients showed bipolar disorder transition. In the external data set, 39 out of 4541 patients showed bipolar disorder transition. The average performance of the federated model in the internal test (area under the curve [AUC] 0.726) and external validation (AUC 0.719) data sets was higher than that of the other locally developed models (AUC 0.642-0.707 and AUC 0.642-0.699, respectively). In the federated model, classifications were driven by several predictors such as the Charlson index (low scores were associated with bipolar transition, which may be due to younger age), severe depression, anxiolytics, young age, and visiting months (the bipolar transition was associated with seasonality, especially during the spring and summer months). Conclusions: We developed and validated a differentially private federated model by using distributed multi-institutional psychiatric data with standardized pipelines in a real-world environment. The federated model performed better than models using local data only.
- Open Access
- 10 March 2023
- Published by IOP Publishing in Journal of Neural Engineering
- Vol. 20 (2) , 026008
Patient specific intracranial neural signatures of obsessions and compulsions in the ventral striatum
Objective. Deep brain stimulation is a treatment option for patients with refractory obsessive-compulsive disorder. A new generation of stimulators hold promise for closed loop stimulation, with adaptive stimulation in response to biologic signals. Here we aimed to discover a suitable biomarker in the ventral striatum in patients with obsessive compulsive disorder using local field potentials. Approach. We induced obsessions and compulsions in 11 patients undergoing deep brain stimulation treatment using a symptom provocation task. Then we trained machine learning models to predict symptoms using the recorded intracranial signal from the deep brain stimulation electrodes. Main results. Average areas under the receiver operating characteristics curve were 62.1% for obsessions and 78.2% for compulsions for patient specific models. For obsessions it reached over 85% in one patient, whereas performance was near chance level when the model was trained across patients. Optimal performances for obsessions and compulsions was obtained at different recording sites. Significance. The results from this study suggest that closed loop stimulation may be a viable option for obsessive-compulsive disorder, but that intracranial biomarkers are patient and not disorder specific. Clinical Trial: Netherlands trial registry NL7486.
- 10 March 2023
- Published by IOP Publishing in Journal of Neural Engineering
- Vol. 20 (2) , 026008
Open AccessPatient specific intracranial neural signatures of obsessions and compulsions in the ventral striatum
Objective. Deep brain stimulation is a treatment option for patients with refractory obsessive-compulsive disorder. A new generation of stimulators hold promise for closed loop stimulation, with adaptive stimulation in response to biologic signals. Here we aimed to discover a suitable biomarker in the ventral striatum in patients with obsessive compulsive disorder using local field potentials. Approach. We induced obsessions and compulsions in 11 patients undergoing deep brain stimulation treatment using a symptom provocation task. Then we trained machine learning models to predict symptoms using the recorded intracranial signal from the deep brain stimulation electrodes. Main results. Average areas under the receiver operating characteristics curve were 62.1% for obsessions and 78.2% for compulsions for patient specific models. For obsessions it reached over 85% in one patient, whereas performance was near chance level when the model was trained across patients. Optimal performances for obsessions and compulsions was obtained at different recording sites. Significance. The results from this study suggest that closed loop stimulation may be a viable option for obsessive-compulsive disorder, but that intracranial biomarkers are patient and not disorder specific. Clinical Trial: Netherlands trial registry NL7486.
- Open Access
- 1 February 2023
- Published by JMIR Publications Inc. in JMIR Preprints
Privacy-Preserving Federated Model Predicting Bipolar Transition in Patients With Depression: Prediction Model Development Study (Preprint)
BACKGROUND Mood disorder has emerged as a serious concern for public health; in particular, bipolar disorder has a less favorable prognosis than depression. Although prompt recognition of depression conversion to bipolar disorder is needed, early prediction is challenging due to overlapping symptoms. Recently, there have been attempts to develop a prediction model by using federated learning. Federated learning in medical fields is a method for training multi-institutional machine learning models without patient-level data sharing. OBJECTIVE This study aims to develop and validate a federated, differentially private multi-institutional bipolar transition prediction model. METHODS This retrospective study enrolled patients diagnosed with the first depressive episode at 5 tertiary hospitals in South Korea. We developed models for predicting bipolar transition by using data from 17,631 patients in 4 institutions. Further, we used data from 4541 patients for external validation from 1 institution. We created standardized pipelines to extract large-scale clinical features from the 4 institutions without any code modification. Moreover, we performed feature selection in a federated environment for computational efficiency and applied differential privacy to gradient updates. Finally, we compared the federated and the 4 local models developed with each hospital's data on internal and external validation data sets. RESULTS In the internal data set, 279 out of 17,631 patients showed bipolar disorder transition. In the external data set, 39 out of 4541 patients showed bipolar disorder transition. The average performance of the federated model in the internal test (area under the curve [AUC] 0.726) and external validation (AUC 0.719) data sets was higher than that of the other locally developed models (AUC 0.642-0.707 and AUC 0.642-0.699, respectively). In the federated model, classifications were driven by several predictors such as the Charlson index (low scores were associated with bipolar transition, which may be due to younger age), severe depression, anxiolytics, young age, and visiting months (the bipolar transition was associated with seasonality, especially during the spring and summer months). CONCLUSIONS We developed and validated a differentially private federated model by using distributed multi-institutional psychiatric data with standardized pipelines in a real-world environment. The federated model performed better than models using local data only.
- 1 February 2023
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessPrivacy-Preserving Federated Model Predicting Bipolar Transition in Patients With Depression: Prediction Model Development Study (Preprint)
BACKGROUND Mood disorder has emerged as a serious concern for public health; in particular, bipolar disorder has a less favorable prognosis than depression. Although prompt recognition of depression conversion to bipolar disorder is needed, early prediction is challenging due to overlapping symptoms. Recently, there have been attempts to develop a prediction model by using federated learning. Federated learning in medical fields is a method for training multi-institutional machine learning models without patient-level data sharing. OBJECTIVE This study aims to develop and validate a federated, differentially private multi-institutional bipolar transition prediction model. METHODS This retrospective study enrolled patients diagnosed with the first depressive episode at 5 tertiary hospitals in South Korea. We developed models for predicting bipolar transition by using data from 17,631 patients in 4 institutions. Further, we used data from 4541 patients for external validation from 1 institution. We created standardized pipelines to extract large-scale clinical features from the 4 institutions without any code modification. Moreover, we performed feature selection in a federated environment for computational efficiency and applied differential privacy to gradient updates. Finally, we compared the federated and the 4 local models developed with each hospital's data on internal and external validation data sets. RESULTS In the internal data set, 279 out of 17,631 patients showed bipolar disorder transition. In the external data set, 39 out of 4541 patients showed bipolar disorder transition. The average performance of the federated model in the internal test (area under the curve [AUC] 0.726) and external validation (AUC 0.719) data sets was higher than that of the other locally developed models (AUC 0.642-0.707 and AUC 0.642-0.699, respectively). In the federated model, classifications were driven by several predictors such as the Charlson index (low scores were associated with bipolar transition, which may be due to younger age), severe depression, anxiolytics, young age, and visiting months (the bipolar transition was associated with seasonality, especially during the spring and summer months). CONCLUSIONS We developed and validated a differentially private federated model by using distributed multi-institutional psychiatric data with standardized pipelines in a real-world environment. The federated model performed better than models using local data only.