Author name cluster

Nils D. Forkert

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

12 papers

1 author row

YNICL Journal 2026 Journal Article

Investigating causal relations between brain morphology and genetic risk variants in Parkinson’s disease

Gabrielle Dagasso
Vibujithan Vigneshwaran
Anthony J Winder
Raissa Souza
Erik Y. Ohara
Matthias Wilms
Nils D. Forkert

Imaging genomics for Parkinson’s disease (PD) research aims to integrate genetic and imaging biomarkers to explore how genetic alterations influence brain morphology and function. However, traditional methods have been largely correlative, limiting their utility. Recent advances in machine learning offer potential for exploring causal relationships, although these have not yet been applied to investigate genetic variants and brain phenotypes in PD. Thus, we employ a causal deep learning approach for genotype-phenotype analysis in PD using a novel method to assess the causal impact of genetic risk variants on brain structures. A masked causal normalizing flow model was adapted to evaluate genetic variants associated with PD and their effects on brain structures. The Parkinson’s Progression Markers Initiative (PPMI) dataset was used for development and evaluation, we included 102 controls, 214 patients with PD, and 43 patients with prodromal PD (n = 359), with 223 males (age range 31–82) An additional testing on neurologically healthy participants from the UK Biobank for validation was done as well, with 16, 861 participants (Male n = 7, 747, age range: 44–82). The causal deep learning model identified several significant causal relationships: the rs4073221 variant in SATB1 affects the right putamen volume (p-value = 6. 8x10-5) and the T408M (rs75548401) variant in GBA1 influences the right pars triangularis volume (p-value = 1x10-13), aligning with known PD pathophysiology. Complex variant analysis of LRRK2 G2019S and GBA1 E365K showed individual-level volumetric changes. Similar trends were found in the UK Biobank and PPMI datasets, demonstrating reasonable generalization. The proposed causal deep learning framework reveals promising results for investigating genetic-brain architectures in PD. It demonstrates feasibility for further imaging genomics studies in PD and other neurological disorders.

Details DOI

JBHI Journal 2024 Journal Article

Identifying Biases in a Multicenter MRI Database for Parkinson's Disease Classification: Is the Disease Classifier a Secret Site Classifier?

Raissa Souza
Anthony Winder
Emma A. M. Stanley
Vibujithan Vigneshwaran
Milton Camacho
Richard Camicioli
Oury Monchi
Matthias Wilms

Sharing multicenter imaging datasets can be advantageous to increase data diversity and size but may lead to spurious correlations between site-related biological and non-biological image features and target labels, which machine learning (ML) models may exploit as shortcuts. To date, studies analyzing how and if deep learning models may use such effects as a shortcut are scarce. Thus, the aim of this work was to investigate if site-related effects are encoded in the feature space of an established deep learning model designed for Parkinson's disease (PD) classification based on T1-weighted MRI datasets. Therefore, all layers of the PD classifier were frozen, except for the last layer of the network, which was replaced by a linear layer that was exclusively re-trained to predict three potential bias types (biological sex, scanner type, and originating site). Our findings based on a large database consisting of 1880 MRI scans collected across 41 centers show that the feature space of the established PD model (74% accuracy) can be used to classify sex (75% accuracy), scanner type (79% accuracy), and site location (71% accuracy) with high accuracies despite this information never being explicitly provided to the PD model during original training. Overall, the results of this study suggest that trained image-based classifiers may use unwanted shortcuts that are not meaningful for the actual clinical task at hand. This finding may explain why many image-based deep learning models do not perform well when applied to data from centers not contributing to the training set.

Details DOI

YNICL Journal 2023 Journal Article

Aerobic exercise increases brain vessel lumen size and blood flow in young adults with elevated blood pressure. Secondary analysis of the TEPHRA randomized clinical trial

Winok Lapidaire
Nils D. Forkert
Wilby Williamson
Odaro Huckstep
Cheryl MJ Tan
Maryam Alsharqi
Afifah Mohamed
Jamie Kitt

IMPORTANCE: Cerebrovascular changes are already evident in young adults with hypertension and exercise is recommended to reduce cardiovascular risk. To what extent exercise benefits the cerebrovasculature at an early stage of the disease remains unclear. OBJECTIVE: To investigate whether structured aerobic exercise increases brain vessel lumen diameter or cerebral blood flow (CBF) and whether lumen diameter is associated with CBF. DESIGN: Open, parallel, two-arm superiority randomized controlled (1:1) trial in the TEPHRA study on an intention-to-treat basis. The MRI sub-study was an optional part of the protocol. The outcome assessors remained blinded until the data lock. SETTING: Single-centre trial in Oxford, UK. PARTICIPANTS: and never been on prescribed hypertension medications. Out of 203 randomized participants, 135 participated in the MRI sub-study. Randomisation was stratified for sex, age ( 37 weeks). INTERVENTION: Study participants were randomised to a 16 week aerobic exercise intervention targeting 3×60 min sessions per week at 60 to 80 % peak heart rate. MAIN OUTCOMES AND MEASURES: cerebral blood flow (CBF) maps from ASL MRI scans, internal carotid artery (ICA), middle cerebral artery (MCA) M1 and M2 segments, anterior cerebral artery (ACA), basilar artery (BA), and posterior cerebral artery (PCA) diameters extracted from TOF MRI scans. RESULTS: Of the 135 randomized participants (median age 28 years, 58 % women) who had high quality baseline MRI data available, 93 participants also had high quality follow-up data available. The exercise group showed an increase in ICA (0.1 cm, 95 % CI 0.01 to 0.18, p =.03) and MCA M1 (0.05 cm, 95 % CI 0.01 to 0.10, p =.03) vessel diameter compared to the control group. Differences in the MCA M2 (0.03 cm, 95 % CI 0.0 to 0.06, p =.08), ACA (0.04 cm, 95 % CI 0.0 to 0.08, p =.06), BA (0.02 cm, 95 % CI -0.04 to 0.09, p =.48), and PCA (0.03 cm, 95 % CI -0.01 to 0.06, p =.17) diameters or CBF were not statistically significant. The increase in ICA vessel diameter in the exercise group was associated with local increases in CBF. CONCLUSIONS AND RELEVANCE: Aerobic exercise induces positive cerebrovascular remodelling in young people with early hypertension, independent of blood pressure. The long-term benefit of these changes requires further study. TRIAL REGISTRATION: Clinicaltrials.gov NCT02723552, 30 March 2016.

Details DOI

YNICL Journal 2023 Journal Article

Explainable classification of Parkinson’s disease using deep learning trained on a large multi-center database of T1-weighted MRI datasets

Milton Camacho
Matthias Wilms
Pauline Mouches
Hannes Almgren
Raissa Souza
Richard Camicioli
Zahinoor Ismail
Oury Monchi

INTRODUCTION: Parkinson's disease (PD) is a severe neurodegenerative disease that affects millions of people. Early diagnosis is important to facilitate prompt interventions to slow down disease progression. However, accurate PD diagnosis can be challenging, especially in the early disease stages. The aim of this work was to develop and evaluate a robust explainable deep learning model for PD classification trained from one of the largest collections of T1-weighted magnetic resonance imaging datasets. MATERIALS AND METHODS: A total of 2,041 T1-weighted MRI datasets from 13 different studies were collected, including 1,024 datasets from PD patients and 1,017 datasets from age- and sex-matched healthy controls (HC). The datasets were skull stripped, resampled to isotropic resolution, bias field corrected, and non-linearly registered to the MNI PD25 atlas. The Jacobian maps derived from the deformation fields together with basic clinical parameters were used to train a state-of-the-art convolutional neural network (CNN) to classify PD and HC subjects. Saliency maps were generated to display the brain regions contributing the most to the classification task as a means of explainable artificial intelligence. RESULTS: The CNN model was trained using an 85%/5%/10% train/validation/test split stratified by diagnosis, sex, and study. The model achieved an accuracy of 79.3%, precision of 80.2%, specificity of 81.3%, sensitivity of 77.7%, and AUC-ROC of 0.87 on the test set while performing similarly on an independent test set. Saliency maps computed for the test set data highlighted frontotemporal regions, the orbital-frontal cortex, and multiple deep gray matter structures as most important. CONCLUSION: The developed CNN model, trained on a large heterogenous database, was able to differentiate PD patients from HC subjects with high accuracy with clinically feasible classification explanations. Future research should aim to investigate the combination of multiple imaging modalities with deep learning and on validating these results in a prospective trial as a clinical decision support system.

Details DOI

YNICL Journal 2023 Journal Article

Motor symptoms in Parkinson’s disease are related to the interplay between cortical curvature and thickness

Hannes Almgren
Alexandru Hanganu
Milton Camacho
Mekale Kibreab
Richard Camicioli
Zahinoor Ismail
Nils D. Forkert
Oury Monchi

INTRODUCTION: Brain atrophy in Parkinson's disease occurs to varying degrees in different brain regions, even at the early stage of the disease. While cortical morphological features are often considered independently in structural brain imaging studies, research on the co-progression of different cortical morphological measurements could provide new insights regarding the progression of PD. This study's aim was to examine the interplay between cortical curvature and thickness as a function of PD diagnosis, motor symptoms, and cognitive performance. METHODS: A total of 359 de novo PD patients and 159 healthy controls (HC) from the Parkinson's Progression Markers Initiative (PPMI) database were included in this study. Additionally, an independent cohort from four databases (182 PD, 132 HC) with longer disease durations was included to assess the effects of PD diagnosis in more advanced cases. Pearson correlation was used to determine subject-specific associations between cortical curvature and thickness estimated from T1-weighted MRI images. General linear modeling (GLM) was then used to assess the effect of PD diagnosis, motor symptoms, and cognitive performance on the curvature-thickness association. Next, longitudinal changes in the curvature-thickness correlation as well as the predictive effect of the cortical curvature-thickness association on changes in motor symptoms and cognitive performance across four years were investigated. Finally, Akaike information criterion (AIC) was used to build a GLM to model PD motor symptom severity cross-sectionally. RESULTS: = 0.03, t(184.7) = 0.78, p = 0.44). No predictive effects of the CC-CT correlation on longitudinal changes in cognitive performance or motor symptoms were observed (all p-values > 0.05). The best cross-sectional model for PD motor symptoms included the curvature-thickness correlation, cognitive performance, and putamen dopamine transporter (DAT) binding, which together explained 14 % of variance. CONCLUSION: The association between cortical curvature and thickness is related to PD motor symptoms and age. This research shows the potential of modeling the curvature-thickness interplay in PD.

Details DOI

JBHI Journal 2022 Journal Article

A Deep Invertible 3-D Facial Shape Model for Interpretable Genetic Syndrome Diagnosis

Jordan J. Bannister
Matthias Wilms
J. David Aponte
David C. Katz
Ophir D. Klein
Francois P. J. Bernier
Richard A. Spritz
Benedikt Hallgrimsson

One of the primary difficulties in treating patients with genetic syndromes is diagnosing their condition. Many syndromes are associated with characteristic facial features that can be imaged and utilized by computer-assisted diagnosis systems. In this work, we develop a novel 3D facial surface modeling approach with the objective of maximizing diagnostic model interpretability within a flexible deep learning framework. Therefore, an invertible normalizing flow architecture is introduced to enable both inferential and generative tasks in a unified and efficient manner. The proposed model can be used (1) to infer syndrome diagnosis and other demographic variables given a 3D facial surface scan and (2) to explain model inferences to non-technical users via multiple interpretability mechanisms. The model was trained and evaluated on more than 4700 facial surface scans from subjects with 47 different syndromes. For the challenging task of predicting syndrome diagnosis given a new 3D facial surface scan, age, and sex of a subject, the model achieves a competitive overall top-1 accuracy of 71%, and a mean sensitivity of 43% across all syndrome classes. We believe that invertible models such as the one presented in this work can achieve competitive inferential performance while greatly increasing model interpretability in the domain of medical diagnosis.

Details DOI

AIIM Journal 2022 Journal Article

Detecting 3D syndromic faces as outliers using unsupervised normalizing flow models

Jordan J. Bannister
Matthias Wilms
J. David Aponte
David C. Katz
Ophir D. Klein
Francois P.J. Bernier
Richard A. Spritz
Benedikt Hallgrímsson

Many genetic syndromes are associated with distinctive facial features. Several computer-assisted methods have been proposed that make use of facial features for syndrome diagnosis. Training supervised classifiers, the most common approach for this purpose, requires large, comprehensive, and difficult to collect databases of syndromic facial images. In this work, we use unsupervised, normalizing flow-based manifold and density estimation models trained entirely on unaffected subjects to detect syndromic 3D faces as statistical outliers. Furthermore, we demonstrate a general, user-friendly, gradient-based interpretability mechanism that enables clinicians and patients to understand model inferences. 3D facial surface scans of 2471 unaffected subjects and 1629 syndromic subjects representing 262 different genetic syndromes were used to train and evaluate the models. The flow-based models outperformed unsupervised comparison methods, with the best model achieving an ROC-AUC of 86. 3% on a challenging, age and sex diverse data set. In addition to highlighting the viability of outlier-based syndrome screening tools, our methods generalize and extend previously proposed outlier scores for 3D face-based syndrome detection, resulting in improved performance for unsupervised syndrome detection.

Details DOI

YNICL Journal 2020 Journal Article

Cerebral volume and diffusion MRI changes in children with sensorineural hearing loss

Peter K. Moon
Jason Z. Qian
Emily McKenna
Kevin Xi
Nathan C. Rowe
Nathan N. Ng
Jimmy Zheng
Lydia T. Tam

PURPOSE: Sensorineural hearing loss (SNHL) is the most prevalent congenital sensory deficit in children. Information regarding underlying brain microstructure could offer insight into neural development in deaf children and potentially guide therapies that optimize language development. We sought to quantitatively evaluate MRI-based cerebral volume and gray matter microstructure children with SNHL. METHODS & MATERIALS: We conducted a retrospective study of children with SNHL who obtained brain MRI at 3 T. The study cohort comprised 63 children with congenital SNHL without known focal brain lesion or structural abnormality (33 males; mean age 5.3 years; age range 1 to 11.8 years) and 64 age-matched controls without neurological, developmental, or MRI-based brain macrostructure abnormality. An atlas-based analysis was used to extract quantitative volume and median diffusivity (ADC) in the following brain regions: cerebral cortex, thalamus, caudate, putamen, globus pallidus, hippocampus, amygdala, nucleus accumbens, brain stem, and cerebral white matter. SNHL patients were further stratified by severity scores and hearing loss etiology. RESULTS: Children with SNHL showed higher median ADC of the cortex (p = .019), thalamus (p < .001), caudate (p = .005), and brainstem (p = .003) and smaller brainstem volumes (p = .007) compared to controls. Patients with profound bilateral SNHL did not show any significant differences compared to patients with milder bilateral SNHL, but both cohorts independently had smaller brainstem volumes compared to controls. Children with unilateral SNHL showed greater amygdala volumes compared to controls (p = .021), but no differences were found comparing unilateral SNHL to bilateral SNHL. Based on etiology for SNHL, patients with Pendrin mutations showed higher ADC values in the brainstem (p = .029, respectively); patients with Connexin 26 showed higher ADC values in both the thalamus (p < .001) and brainstem (p < .001) compared to controls. CONCLUSION: SNHL patients showed significant differences in diffusion and volume in brain subregions, with region-specific findings for patients with Connexin 26 and Pendrin mutations. Future longitudinal studies could examine macro- and microstructure changes in children with SNHL over development and potential predictive role for MRI after interventions including cochlear implant outcome.

Details DOI

YNICL Journal 2020 Journal Article

Structural and functional connectivity of motor circuits after perinatal stroke: A machine learning study

Helen L. Carlson
Brandon T. Craig
Alicia J. Hilderley
Jacquie Hodge
Deepthi Rajashekar
Pauline Mouches
Nils D. Forkert
Adam Kirton

Developmental neuroplasticity allows young brains to adapt via experiences early in life and also to compensate after injury. Why certain individuals are more adaptable remains underexplored. Perinatal stroke is an ideal human model of neuroplasticity with focal lesions acquired near birth in a healthy brain. Machine learning can identify complex patterns in multi-dimensional datasets. We used machine learning to identify structural and functional connectivity biomarkers most predictive of motor function. Forty-nine children with perinatal stroke and 27 controls were studied. Functional connectivity was quantified by fluctuations in blood oxygen-level dependent (BOLD) signal between regions. White matter tractography of corticospinal tracts quantified structural connectivity. Motor function was assessed using validated bimanual and unimanual tests. RELIEFF feature selection and random forest regression models identified predictors of each motor outcome using neuroimaging and demographic features. Unilateral motor outcomes were predicted with highest accuracy (8/54 features r = 0.58, 11/54 features, r = 0.34) but bimanual function required more features (51/54 features, r = 0.38). Connectivity of both hemispheres had important roles as did cortical and subcortical regions. Lesion size, age at scan, and type of stroke were predictive but not highly ranked. Machine learning regression models may represent a powerful tool in identifying neuroimaging biomarkers associated with clinical motor function in perinatal stroke and may inform personalized targets for neuromodulation.

Details DOI

YNIMG Journal 2019 Journal Article

Rapid solution of the Bloch-Torrey equation in anisotropic tissue: Application to dynamic susceptibility contrast MRI of cerebral white matter

Jonathan Doucette
Luxi Wei
Enedino Hernández-Torres
Christian Kames
Nils D. Forkert
Rasmus Aamand
Torben E. Lund
Brian Hansen

Blood vessel related magnetic resonance imaging (MRI) contrast provides a window into the brain's metabolism and function. Here, we show that the spin echo dynamic susceptibility contrast (DSC) MRI signal of the brain's white matter (WM) strongly depends on the angle between WM tracts and the main magnetic field. The apparent cerebral blood flow and volume are 20% larger in fibres perpendicular to the main magnetic field compared to parallel fibres. We present a rapid numerical framework for the solution of the Bloch-Torrey equation that allows us to explore the isotropic and anisotropic components of the vascular tree. By fitting the simulated spin echo DSC signal to the measured data, we show that half of the WM vascular volume is comprised of vessels running in parallel with WM fibre tracts. The WM blood volume corresponding to the best fit to the experimental data was 2. 82%, which is close to the PET gold standard of 2. 6%.

Details DOI

YNICL Journal 2018 Journal Article

Widespread diffusion changes differentiate Parkinson's disease and progressive supranuclear palsy

Aron S. Talai
Jan Sedlacik
Kai Boelmans
Nils D. Forkert

BACKGROUND: Parkinson's disease (PD) and progressive supranuclear palsy - Richardson's syndrome (PSP-RS) are often represented by similar clinical symptoms, which may challenge diagnostic accuracy. The objective of this study was to investigate and compare regional cerebral diffusion properties in PD and PSP-RS subjects and evaluate the use of these metrics for an automatic classification framework. MATERIAL AND METHODS: Diffusion-tensor MRI datasets from 52 PD and 21 PSP-RS subjects were employed for this study. Using an atlas-based approach, regional median values of mean diffusivity (MD), fractional anisotropy (FA), radial diffusivity (RD), and axial diffusivity (AD) were measured and employed for feature selection using RELIEFF and subsequent classification using a support vector machine. RESULTS: According to RELIEFF, the top 17 diffusion values consisting of deep gray matter structures, the brainstem, and frontal cortex were found to be especially informative for an automatic classification. A MANCOVA analysis performed on these diffusion values as dependent variables revealed that PSP-RS and PD subjects differ significantly (p < .001). Generally, PSP-RS subjects exhibit reduced FA, and increased MD, RD, and AD values in nearly all brain structures analyzed compared to PD subjects. The leave-one-out cross-validation of the support vector machine classifier revealed that the classifier can differentiate PD and PSP-RS subjects with an accuracy of 87.7%. More precisely, six PD subjects were wrongly classified as PSP-RS and three PSP-RS subjects were wrongly classified as PD. CONCLUSION: The results of this study demonstrate that PSP-RS subjects exhibit widespread and more severe diffusion alterations compared to PD patients, which appears valuable for an automatic computer-aided diagnosis approach.

Details DOI

YNICL Journal 2015 Journal Article

Mapping causal functional contributions derived from the clinical assessment of brain damage after stroke

Melissa Zavaglia
Nils D. Forkert
Bastian Cheng
Christian Gerloff
Götz Thomalla
Claus C. Hilgetag

Lesion analysis reveals causal contributions of brain regions to mental functions, aiding the understanding of normal brain function as well as rehabilitation of brain-damaged patients. We applied a novel lesion inference technique based on game theory, Multi-perturbation Shapley value Analysis (MSA), to a large clinical lesion dataset. We used MSA to analyze the lesion patterns of 148 acute stroke patients together with their neurological deficits, as assessed by the National Institutes of Health Stroke Scale (NIHSS). The results revealed regional functional contributions to essential behavioral and cognitive functions as reflected in the NIHSS, particularly by subcortical structures. There were also side specific differences of functional contributions between the right and left hemispheric brain regions which may reflect the dominance of the left hemispheric syndrome aphasia in the NIHSS. Comparison of MSA to established lesion inference methods demonstrated the feasibility of the approach for analyzing clinical data and indicated its capability for objectively inferring functional contributions from multiple injured, potentially interacting sites, at the cost of having to predict the outcome of unknown lesion configurations. The analysis of regional functional contributions to neurological symptoms measured by the NIHSS contributes to the interpretation of this widely used standardized stroke scale in clinical practice as well as clinical trials and provides a first approximation of a 'map of stroke'.

Details DOI