Jakob Heinzle

Last Name

Heinzle

First Name

Jakob

ORCID

0000-0001-5228-041X

Show all metadata

Search Results

Publications 1 - 10 of 25

A Hilbert-based method for processing respiratory timeseries
Harrison, Samuel J.; Bianchi, Samuel; Heinzle, Jakob; et al. (2021)
NeuroImage
In this technical note, we introduce a new method for estimating changes in respiratory volume per unit time (RVT) from respiratory bellows recordings. By using techniques from the electrophysiological literature, in particular the Hilbert transform, we show how we can better characterise breathing rhythms, with the goal of improving physiological noise correction in functional magnetic resonance imaging (fMRI). Specifically, our approach leads to a representation with higher time resolution and better captures atypical breathing events than current peak-based RVT estimators. Finally, we demonstrate that this leads to an increase in the amount of respiration-related variance removed from fMRI data when used as part of a typical preprocessing pipeline. Our implementation is publicly available as part of the PhysIO package, which is distributed as part of the open-source TAPAS toolbox (https://translationalneuromodeling.org/tapas).
Machine Learning Model for Response to Internet-Delivered CBT vs Antidepressant Medication
Lee, Chi Tak; Richards, Derek; Heinzle, Jakob; et al. (2025)
JAMA Network Open
Importance: Many treatments exist for depression, yet none are universally effective. Multivariable predictive models support personalized treatment selection. Objective: To develop a model predicting response to internet-delivered cognitive behavioral therapy (iCBT) and test its treatment specificity against antidepressant medications. Design, Setting, and Participants: The Precision in Psychiatry prognostic study was a 4-week study collecting extensive baseline self-report and cognitive data online to predict early iCBT response, from February 2019 to May 2022. Patients in the iCBT group were recruited via an Irish mental health charity and a UK NHS Talking Therapies clinic. A separate antidepressant group was recruited globally online and via print advertisements. Participants were aged 18 to 70 years, fluent in English, had computer access, started treatment within 2 days of enrollment, and scored at least 10 on the Work and Social Adjustment Scale. Analysis was completed in December 2024. Exposures: Low-intensity, clinician-guided iCBT with multimedia psychoeducation. Patients receiving antidepressants primarily received selective serotonin-reuptake inhibitors or serotonin-norepinephrine reuptake inhibitors. Main Outcomes and Measures: Machine learning models were trained using the iCBT sample to predict change in depression severity (16-item Quick Inventory of Depressive Symptomatology-Self Report) at week 4. The best model was tested on holdout iCBT and antidepressant data. A separate model was trained on patients receiving iCBT only to assess treatment-specificity. Results: Of 2674 patients screened, 883 completed baseline and final assessments, with 776 patients receiving iCBT (mean [SD] age, 31.8 [11.0] years; 600 [77.5%] female) and 107 patients receiving antidepressant medication (mean [SD] age, 30.1 [10.4] years; 78 [72.9%] female). Both samples had some treatment overlap (24% and 34%, respectively). Elastic net regression with 27 predictors best explained the variance in depression change (R² = 14%; SD, 0.8%; 95% CI, 13.8%-14.2%). Key predictors included baseline depression, treatment expectation, transdiagnostic symptoms, and, less strongly, cognitive variables. The model performed well on holdout iCBT (R² = 18.8%; root mean square error [RMSE], 0.88) and antidepressant (R² = 17.9%; RMSE, 1.10) data. Retraining on 181 patients who received iCBT only increased treatment specificity in predictions (R² = 19.3%; RMSE, 0.89) vs 71 patients who received antidepressants only (R² = 10.8%; RMSE, 1.17). Conclusions and Relevance: This prognostic study in a naturalistic setting found that self-reported data predicted iCBT response better than cognitive data. Model predictions generalized to patients receiving antidepressants, some of whom also received psychotherapy. Training models on single-treatment cohorts may yield more treatment-specific predictions.
Hemodynamic modeling of long‐term aspirin effects on blood oxygenated level dependent responses at 7 Tesla in patients at cardiovascular risk
Do, Cao-Tri; Manjaly, Zina-Mary; Heinzle, Jakob; et al. (2021)
European Journal of Neuroscience
Aspirin is considered a potential confound for functional magnetic resonance imaging (fMRI) studies. This is because aspirin affects the synthesis of prostaglandin, a vasoactive mediator centrally involved in neurovascular coupling, a process underlying blood oxygenated level dependent (BOLD) responses. Aspirin‐induced changes in BOLD signal are a potential confound for fMRI studies of at‐risk individuals or patients (e.g. with cardiovascular conditions or stroke) who receive low‐dose aspirin prophylactically and are compared to healthy controls without aspirin. To examine the severity of this potential confound, we combined high field (7 Tesla) MRI during a simple hand movement task with a biophysically informed hemodynamic model. We compared elderly individuals receiving aspirin for primary or secondary prophylactic purposes versus age‐matched volunteers without aspirin medication, testing for putative differences in BOLD responses. Specifically, we fitted hemodynamic models to BOLD responses from 14 regions activated by the task and examined whether model parameter estimates were significantly altered by aspirin. While our analyses indicate that hemodynamics differed across regions, consistent with the known regional variability of BOLD responses, we neither found a significant main effect of aspirin (i.e., an average effect across brain regions) nor an expected drug × region interaction. While our sample size is not sufficiently large to rule out small‐to‐medium global effects of aspirin, we had adequate statistical power for detecting the expected interaction. Altogether, our analysis suggests that patients with cardiovascular risk receiving low‐dose aspirin for primary or secondary prophylactic purposes do not show strongly altered BOLD signals when compared to healthy controls without aspirin. © 2020 Federation of European Neuroscience Societies and John Wiley & Sons Ltd.
Advances in spiral fMRI: A high-resolution study with single-shot acquisition
Kasper, Lars; Engel, Maria; Heinzle, Jakob; et al. (2022)
NeuroImage
Spiral fMRI has been put forward as a viable alternative to rectilinear echo-planar imaging, in particular due to its enhanced average k-space speed and thus high acquisition efficiency. This renders spirals attractive for contemporary fMRI applications that require high spatiotemporal resolution, such as laminar or columnar fMRI. However, in practice, spiral fMRI is typically hampered by its reduced robustness and ensuing blurring artifacts, which arise from imperfections in both static and dynamic magnetic fields. Recently, these limitations have been overcome by the concerted application of an expanded signal model that accounts for such field imperfections, and its inversion by iterative image reconstruction. In the challenging ultra-high field environment of 7 Tesla, where field inhomogeneity effects are aggravated, both multi-shot and single-shot 2D spiral imaging at sub-millimeter resolution was demonstrated with high depiction quality and anatomical congruency. In this work, we further these advances towards a time series application of spiral readouts, namely, single-shot spiral BOLD fMRI at 0.8 mm in-plane resolution. We demonstrate that high-resolution spiral fMRI at 7 T is not only feasible, but delivers both excellent image quality, BOLD sensitivity, and spatial specificity of the activation maps, with little artifactual blurring. Furthermore, we show the versatility of the approach with a combined in/out spiral readout at a more typical resolution (1.5 mm), where the high acquisition efficiency allows to acquire two images per shot for improved sensitivity by echo combination.
Individual treatment expectations predict clinical outcome after lumbar injections against low back pain
Müller-Schrader, Matthias; Heinzle, Jakob; Müller, Alfred; et al. (2023)
Pain
Subjective expectations are known to be associated with clinical outcomes. However, expectations exist about different aspects of recovery, and few studies have focused on expectations about specific treatments. Here, we present results from a prospective observational study of patients receiving lumbar steroid injections against low back pain (N = 252). Patients completed questionnaires directly before (T1), directly after (T2), and 2 weeks after (T3) the injection. In addition to pain intensity, we assessed expectations (and certainty therein) about treatment effects, using both numerical rating scale (NRS) and the Expectation for Treatment Scale (ETS). Regression models were used to explain (within-sample) treatment outcome (pain intensity at T3) based on pain levels, expectations, and certainty at T1 and T2. Using cross-validation, we examined the models' ability to predict (out-of-sample) treatment outcome. Pain intensity significantly decreased (P < 10-15) 2 weeks after injections, with a reduction of the median NRS score from 6 to 3. Numerical Rating Scale measures of pain, expectation, and certainty from T1 jointly explained treatment outcome (P < 10-15, R2= 0.31). Expectations at T1 explained outcome on its own (P < 10-10,f2=0.19) and enabled out-of-sample predictions about outcome (P < 10-4), with a median error of 1.36 on a 0 to 10 NRS. Including measures from T2 did not significantly improve models. Using the ETS as an alternative measurement of treatment expectations (sensitivity analysis) gave consistent results. Our results demonstrate that treatment expectations play an important role for clinical outcome after lumbar injections and may represent targets for concomitant cognitive interventions. Predicting outcomes based on simple questionnaires might be useful to support treatment selection.
Predictive modelling of clinically significant depressive symptoms after coronary artery bypass graft surgery: protocol for a multicentre observational study in two Swiss hospitals (the PsyCor study)
Lazaridou, Asimina; Sivakumar, Sinthujan; Biefer, Hector Rodriguez Cetina; et al. (2025)
BMJ Open
Introduction: Coronary artery bypass grafting (CABG) remains one of the most commonly performed cardiac surgeries worldwide. Despite surgical advancements, a significant proportion of patients experience psychological distress following surgery, with depression being particularly common. Current evidence regarding the effectiveness of preoperative psychological interventions in improving postoperative mental health outcomes remains inconclusive. There is a critical need for predictive models that can identify patients at risk of developing clinically significant depressive symptoms (CSDSs) and related psychological conditions after CABG. This multicentre observational study aims to develop and validate prognostic models for predicting CSDSs and other psychological outcomes, including anxiety, post-traumatic stress symptoms and quality of life, 6 weeks after elective CABG surgery. Methods and analysis: The study will recruit 300 adult patients undergoing elective CABG (with or without valve intervention) across two Swiss hospitals. Data collected will include demographic, clinical, psychometric, inflammation-related and interoceptive variables. A training set (n=200) will be used to develop predictive models using machine learning, while a held-out test set (n=100) will be used for model validation. The primary outcome prediction will focus on CSDSs, assessed using the Patient Health Questionnaire-9 (PHQ-9), with analyses conducted both categorically (PHQ-9 total score ≥10) and continuously as complementary approaches. Secondary models will address anxiety, using the General Anxiety Disorder Scale-7, post-traumatic stress, using the post-traumatic stress disorder checklist for Diagnostic and Statistical Manual of Mental Disorders-5 and health-related quality of life, using the 12-item Short Form Survey. A simplified ‘light solution’ model with fewer predictors will also be developed for broader applicability. This study will address an important gap in perioperative mental healthcare by identifying key predictors of psychological morbidity following CABG, particularly CSDSs. The resulting models may inform future screening and preventive strategies and improve postsurgical outcomes through early identification and intervention in high-risk individuals. Ethics and dissemination: The responsible ethics committee has reviewed and approved this project (Kantonale Ethikkommission Zürich, BASEC number: 2023-02040). The study minimises participant burden by integrating brief validated instruments and limiting psychiatric interviews to relevant outcomes, while ensuring ethical safeguards and respect for participant rights (including written consent). Results will be shared through peer-reviewed publications, conference presentations and stakeholder meetings involving clinicians and mental health professionals. Findings will also be communicated to participating centres and patient communities in accessible formats.
Predictive brain signals best predict upcoming and not previous choices
Soon, Chun S.; Allefeld, Carsten; Bogler, Carsten; et al. (2014)
Frontiers in Psychology
Dynamic Causal Modeling and Its Application to Psychiatric Disorders
Heinzle, Jakob; Stephan, Klaas (2018)
Computational Psychiatry: Mathematical Modeling of Mental Illness
Timing of repetition suppression of event-related potentials to unattended objects
Stefanics, Gabor; Heinzle, Jakob; Czigler, István; et al. (2020)
European Journal of Neuroscience
Feature-specific prediction errors for visual mismatch
Stefanics, Gabor; Stephan, Klaas; Heinzle, Jakob (2019)
NeuroImage

Publications 1 - 10 of 25

Jakob Heinzle

Last Name

First Name

ORCID

Organisational unit

Filters

Search Results