Prof. Nils Daniel Forkert Profile

Prof. Nils Daniel Forkert

Professor at Univ. of Calgary

SPIE Involvement:

Author

Publications (22)

SPIE Journal Paper | 20 September 2024 Open Access

Open Access

HarmonyTM: multi-center data harmonization applied to distributed learning for Parkinson’s disease classification

Raissa Souza, Emma A. Stanley, Vedant Gulve, Jasmine Moore, Chris Kang, Richard Camicioli, Oury Monchi, Zahinoor Ismail, Matthias Wilms, Nils D. Forkert

JMI, Vol. 11, Issue 05, 054502, (September 2024) https://doi.org/10.1117/12.10.1117/1.JMI.11.5.054502

KEYWORDS: Scanners, Data modeling, Education and training, Machine learning, Principal component analysis, Diseases and disorders, Neuroimaging, Data acquisition, Image acquisition, Head

Read Abstract +

Proceedings Article | 2 April 2024 Presentation + Paper

Presentation + Paper

MACAW 3D: A masked causal normalizing flow method for counterfactual 3D brain image generation

Erik Ohara, Finn Vamosi, Harsh Patil, Vibujithan Vigneshwaran, Matthias Wilms, Nils Forkert

Proceedings Volume 12931, 129310K (2024) https://doi.org/10.1117/12.3006482

KEYWORDS: 3D modeling, 3D image processing, Brain, Data modeling, Neuroimaging, Medical imaging, Deep learning, Image processing, Artificial intelligence

Read Abstract +

Proceedings Article | 2 April 2024 Presentation + Paper

Presentation + Paper

CNN-derived brain age gaps of different neurological and cardiovascular diseases in the UK Biobank and identifying affected brain regions

Elizabeth Mcavoy, Matthias Wilms, Nils Forkert

Proceedings Volume 12931, 129310A (2024) https://doi.org/10.1117/12.3006650

KEYWORDS: Brain, Cardiovascular disorders, Brain diseases, Mental disorders, Bias correction, Neurological disorders, Image processing, Artificial intelligence, Deep convolutional neural networks, Brain imaging

Read Abstract +

Proceedings Article | 10 April 2023 Presentation + Paper

Presentation + Paper

A pipeline for multivariate genome-wide associations studies with morphological brain features

Gabrielle Dagasso, Matthias Wilms, Nils Forkert

Proceedings Volume 12469, 124690L (2023) https://doi.org/10.1117/12.2654600

KEYWORDS: Brain, Genetics, Brain diseases, Neuroimaging, Mental disorders, Biological imaging, Statistical analysis, Neurological disorders, Medicine

Read Abstract +

Proceedings Article | 7 April 2023 Poster + Paper

Poster + Paper

An analysis of intensity harmonization techniques for Parkinson’s multi-site MRI datasets

Raissa Souza, Emma A. Stanley, Milton Camacho, Matthias Wilms, Nils Forkert

Proceedings Volume 12465, 124652B (2023) https://doi.org/10.1117/12.2653948

KEYWORDS: Data modeling, Histograms, Performance modeling, Magnetic resonance imaging, Education and training, Parkinson disease, Neuroimaging, Machine learning, Solid modeling, Scanners

Read Abstract +

SPIE Journal Paper | 26 August 2022

Fairness-related performance and explainability effects in deep learning models for brain image analysis

Emma A. Stanley, Matthias Wilms, Pauline Mouches, Nils D. Forkert

JMI, Vol. 9, Issue 06, 061102, (August 2022) https://doi.org/10.1117/12.10.1117/1.JMI.9.6.061102

KEYWORDS: Brain, Performance modeling, Brain mapping, Cerebellum, Magnetic resonance imaging, Data modeling, Amygdala, Neuroimaging, Solid modeling, Artificial intelligence

Read Abstract +

Purpose: Explainability and fairness are two key factors for the effective and ethical clinical implementation of deep learning-based machine learning models in healthcare settings. However, there has been limited work on investigating how unfair performance manifests in explainable artificial intelligence (XAI) methods, and how XAI can be used to investigate potential reasons for unfairness. Thus, the aim of this work was to analyze the effects of previously established sociodemographic-related confounders on classifier performance and explainability methods.Approach: A convolutional neural network (CNN) was trained to predict biological sex from T1-weighted brain MRI datasets of 4547 9- to 10-year-old adolescents from the Adolescent Brain Cognitive Development study. Performance disparities of the trained CNN between White and Black subjects were analyzed and saliency maps were generated for each subgroup at the intersection of sex and race.Results: The classification model demonstrated a significant difference in the percentage of correctly classified White male (90.3 % ± 1.7 % ) and Black male (81.1 % ± 4.5 % ) children. Conversely, slightly higher performance was found for Black female (89.3 % ± 4.8 % ) compared with White female (86.5 % ± 2.0 % ) children. Saliency maps showed subgroup-specific differences, corresponding to brain regions previously associated with pubertal development. In line with this finding, average pubertal development scores of subjects used in this study were significantly different between Black and White females (p < 0.001) and males (p < 0.001).Conclusions: We demonstrate that a CNN with significantly different sex classification performance between Black and White adolescents can identify different important brain regions when comparing subgroup saliency maps. Importance scores vary substantially between subgroups within brain structures associated with pubertal development, a race-associated confounder for predicting sex. We illustrate that unfair models can produce different XAI results between subgroups and that these results may explain potential reasons for biased performance.

Proceedings Article | 4 April 2022 Presentation + Paper

Presentation + Paper

A comparative analysis of the impact of data distribution on distributed learning with a traveling model for brain age prediction

Raissa Souza, Agampreet Aulakh, Pauline Mouches, Anup Tuladhar, Matthias Wilms, Sönke Langner, Nils Forkert

Proceedings Volume 12037, 1203702 (2022) https://doi.org/10.1117/12.2612728

KEYWORDS: Data modeling, Brain, Neuroimaging, Performance modeling, Machine learning, Data centers, Magnetic resonance imaging, Solid modeling, Medical research, Feature extraction

Read Abstract +

Proceedings Article | 4 April 2022 Presentation + Paper

Presentation + Paper

A fully convolutional neural network for explainable classification of attention deficit hyperactivity disorder

Emma A. Stanley, Deepthi Rajashekar, Pauline Mouches, Matthias Wilms, Kira Plettl, Nils Forkert

Proceedings Volume 12033, 1203315 (2022) https://doi.org/10.1117/12.2607509

KEYWORDS: Brain, Data modeling, Convolutional neural networks, Magnetic resonance imaging, Cognitive modeling, Neuroimaging, Diagnostics, Amygdala, Computer aided diagnosis and therapy, Machine learning

Read Abstract +

Proceedings Article | 4 April 2022 Presentation + Paper

Presentation + Paper

Lesion-preserving unpaired image-to-image translation between MRI and CT from ischemic stroke patients

Alejandro Gutierrez, Anup Tuladhar, Deepthi Rajashekar, Nils Forkert

Proceedings Volume 12033, 1203317 (2022) https://doi.org/10.1117/12.2613203

KEYWORDS: Magnetic resonance imaging, Computed tomography, Ischemic stroke, Brain, Tissues, Image analysis, Neuroimaging, Data modeling, Process modeling, Artificial intelligence, Realistic image synthesis

Read Abstract +

Proceedings Article | 4 April 2022 Presentation + Paper

Presentation + Paper

Stroke lesion localization in 3D MRI datasets with deep reinforcement learning

Samuel Robertson, Anup Tuladhar, Deepthi Rajashekar, Nils Forkert

Proceedings Volume 12033, 120330M (2022) https://doi.org/10.1117/12.2613039

KEYWORDS: Magnetic resonance imaging, Medical imaging, Ischemic stroke, Image segmentation, Data modeling, Machine vision, Machine learning, Computer vision technology

Read Abstract +

Proceedings Article | 15 February 2021 Presentation + Paper

Presentation + Paper

Understanding privacy risks in typical deep learning models for medical image analysis

Nagesh Subbanna, Anup Tuladhar, Matthias Wilms, Nils D. Forkert

Proceedings Volume 11601, 116010E (2021) https://doi.org/10.1117/12.2582014

KEYWORDS: Data modeling, Visual process modeling, Medical imaging, Image segmentation, Magnetic resonance imaging, Image restoration, Machine vision, Ischemic stroke, Image compression, Databases

Read Abstract +

Deep learning in medical imaging typically requires sensitive and confidential patient data for model training. Recent research in computer vision has shown that it is possible to recover training data from trained models using model inversion techniques. In this paper, we investigate the degree to which encoder-decoder like architectures (U-Nets, etc) commonly used in medical imaging are vulnerable to simple model inversion attacks. Utilising a database consisting of 20 MRI datasets from acute ischemic stroke patients, we trained an autoencoder model for image reconstruction and a U-Net model for lesion segmentation. In the second step, model inversion decoders were developed and trained to reconstruct the original MRIs from the low dimensional representation of the trained autoencoder and the U-Net model. The inversion decoders were trained using 24 independent MRI datasets of acute stroke patients not used for training of the original models. Skull-stripped as well as the full original datasets including the skull and other non-brain tissues were used for model training and evaluation. The results show that the trained inversion decoder can be used to reconstruct training datasets after skull stripping given the latent space of the autoencoder trained for image reconstruction (mean correlation coefficient= 0.49), while it was not possible to fully reconstruct the original image used for training of a segmentation task UNet (mean correlation coefficient=0.18). These results are further supported by the structural similarity index measure (SSIM) scores, which show a mean SSIM score of 0.51± 0.14 for the autoencoder trained for image reconstruction, while the average SSIM score for the U-Net trained for the lesion segmentation task was 0.28±0.12. The same experiments were then conducted on the same images but without skull stripping. In this case, the U-Net trained for segmentation shows significantly worse results, while the autoencoder trained for image reconstruction is not affected. Our results suggest that an autoencoder model trained for image compression can be inverted with high accuracy while this is much harder to achieve for a U-Net trained for lesion segmentation.

Proceedings Article | 3 March 2017 Paper

Paper

Automatic classification of cardioembolic and arteriosclerotic ischemic strokes from apparent diffusion coefficient datasets using texture analysis and deep learning

Javier Villafruela, Sebastian Crites, Bastian Cheng, Christian Knaack, Götz Thomalla, Bijoy Menon, Nils Forkert

Proceedings Volume 10134, 101342K (2017) https://doi.org/10.1117/12.2255012

KEYWORDS: Ischemic stroke, Diffusion, Image classification, Magnetic resonance imaging, Pattern recognition, Machine learning, Feature extraction, Neural networks, Image filtering, Diffusion weighted imaging

Read Abstract +

Proceedings Article | 3 March 2017 Paper

Paper

Automatic classification of patients with idiopathic Parkinson's disease and progressive supranuclear palsy using diffusion MRI datasets

Sahand Talai, Kai Boelmans, Jan Sedlacik, Nils Forkert

Proceedings Volume 10134, 101342H (2017) https://doi.org/10.1117/12.2254418

KEYWORDS: Parkinson's disease, Diffusion magnetic resonance imaging, Image classification, Brain, Computer aided diagnosis and therapy, Diffusion, Neuroimaging, Magnetic resonance imaging, Feature extraction, Feature selection, Diffusion tensor imaging, Brain mapping, Decision support systems, Computing systems

Read Abstract +

Proceedings Article | 3 March 2017 Paper

Paper

Comparison of classification methods for voxel-based prediction of acute ischemic stroke outcome following intra-arterial intervention

Anthony Winder, Susanne Siemonsen, Fabian Flottmann, Jens Fiehler, Nils Forkert

Proceedings Volume 10134, 101344B (2017) https://doi.org/10.1117/12.2254118

KEYWORDS: Tissues, Ischemic stroke, Magnetic resonance imaging, Feature extraction, Machine learning, Brain, Ischemia, Diffusion magnetic resonance imaging, Clinical research, Data modeling, Artificial intelligence, Diffusion weighted imaging, Diffusion

Read Abstract +

Proceedings Article | 3 March 2017 Presentation + Paper

Presentation + Paper

Vessel segmentation in 4D arterial spin labeling magnetic resonance angiography images of the brain

Renzo Phellan, Thomas Lindner, Alexandre Falcão, Nils Forkert

Proceedings Volume 10134, 101341B (2017) https://doi.org/10.1117/12.2254119

KEYWORDS: Image processing algorithms and systems, Binary data, Tissues, Visualization, Image segmentation, Neuroimaging, Brain, Magnetic resonance angiography, Blood, Blood circulation

Read Abstract +

Proceedings Article | 17 March 2015 Paper

Paper

Effect of sample size on multi-parametric prediction of tissue outcome in acute ischemic stroke using a random forest classifier

Nils Daniel Forkert, Jens Fiehler

Proceedings Volume 9417, 94172H (2015) https://doi.org/10.1117/12.2082686

KEYWORDS: Tissues, Magnetic resonance imaging, Diffusion weighted imaging, Ischemic stroke, Brain, Blood, Machine learning, Clinical research, Brain mapping, Binary data

Read Abstract +

Proceedings Article | 13 March 2014 Paper

Paper

Is there more valuable information in PWI datasets for a voxel-wise acute ischemic stroke tissue outcome prediction than what is represented by typical perfusion maps?

Nils Daniel Forkert, Susanne Siemonsen, Michael Dalski, Tobias Verleger, Andre Kemmling, Jens Fiehler

Proceedings Volume 9038, 90381O (2014) https://doi.org/10.1117/12.2043490

KEYWORDS: Tissues, Ischemic stroke, Diffusion weighted imaging, Brain, Magnetic resonance imaging, Image segmentation, Blood, Neuroimaging, Data modeling, Brain mapping

Read Abstract +

The acute ischemic stroke is a leading cause for death and disability in the industry nations. In case of a present acute ischemic stroke, the prediction of the future tissue outcome is of high interest for the clinicians as it can be used to support therapy decision making. Within this context, it has already been shown that the voxel-wise multi-parametric tissue outcome prediction leads to more promising results compared to single channel perfusion map thresholding. Most previously published multi-parametric predictions employ information from perfusion maps derived from perfusion-weighted MRI together with other image sequences such as diffusion-weighted MRI. However, it remains unclear if the typically calculated perfusion maps used for this purpose really include all valuable information from the PWI dataset for an optimal tissue outcome prediction. To investigate this problem in more detail, two different methods to predict tissue outcome using a k-nearest-neighbor approach were developed in this work and evaluated based on 18 datasets of acute stroke patients with known tissue outcome. The first method integrates apparent diffusion coefficient and perfusion parameter (Tmax, MTT, CBV, CBF) information for the voxel-wise prediction, while the second method employs also apparent diffusion coefficient information but the complete perfusion information in terms of the voxel-wise residue functions instead of the perfusion parameter maps for the voxel-wise prediction. Overall, the comparison of the results of the two prediction methods for the 18 patients using a leave-one-out cross validation revealed no considerable differences. Quantitatively, the parameter-based prediction of tissue outcome led to a mean Dice coefficient of 0.474, while the prediction using the residue functions led to a mean Dice coefficient of 0.461. Thus, it may be concluded from the results of this study that the perfusion parameter maps typically derived from PWI datasets include all valuable perfusion information required for a voxel-based tissue outcome prediction, while the complete analysis of the residue functions does not add further benefits for the voxel-wise tissue outcome prediction and is also computationally more expensive.

Proceedings Article | 29 March 2013 Paper

Paper

Multiparametric prediction of acute ischemic stroke tissue outcome using CT perfusion datasets

Nils Daniel Forkert, Jens Fiehler, Susanne Siemonsen, Andre Kemmling

Proceedings Volume 8672, 86721Y (2013) https://doi.org/10.1117/12.2004442

KEYWORDS: Tissues, Ischemic stroke, Computed tomography, Targeting Task Performance metric, Brain, Blood, Brain mapping, Magnetic resonance imaging, Data modeling, Arteries

Read Abstract +

Acute ischemic strokes are a major cause for death and severe neurologic deficits in the western hemisphere. The prediction of tissue outcome in case of an acute ischemic stroke is an important variable for treatment decision. An estimation of the expected outcome is typically obtained by thresholding a single perfusion parameter map, which is calculated from a perfusion CT dataset. However, cerebral perfusion is complex and the severity of perfusion impairment is not consistent within the penumbra of an acute ischemic stroke. Therefore, the application of only one parameter for acute stroke tissue outcome prediction may oversimplify the given problem. The aim of this study was to develop and evaluate the feasibility of a multiparametric approach for estimating tissue outcome in acute ischemic stroke patients using 15 CT perfusion datasets. For this purpose, perfusion parameter maps of cerebral blood flow, cerebral blood volume and mean transit time were calculated based on the concentration time curves derived from perfusion CT datasets. The parameter maps of ten patients were employed for a voxel-wise training of a support vector machine using ground-truth final infarct segmentations, whereas the remaining five patient datasets were used for evaluation of the voxel-wise prediction of tissue outcome using the trained support vector machine. Furthermore, tissue outcome was also predicted by optimal thresholding of corresponding time-to-peak (TTP) maps for comparison purposes. Both predictions were compared to ground-truth final infarct lesions for the five datasets used for evaluation. The proposed multiparametric tissue outcome prediction lead to superior prediction results in all cases. More precisely, the multiparametric prediction lead to a mean Dice coefficient of 0.556, while optimal thresholding of TTP maps lead to an average Dice-coefficient of 0.444 compared to the ground-truth infarct lesions. In conclusion, the evaluation results of the proposed method suggest that a multiparametric tissue outcome prediction may be feasible for CT perfusion datasets but needs to be evaluated in more detail.

Proceedings Article | 16 April 2012 Paper

Paper

Comparison of TTP and Tmax estimation techniques in perfusion-weighted MR datasets for tissue-at-risk definition

Nils Daniel Forkert, Philipp Kaesemann, Jens Fiehler, Götz Thomalla

Proceedings Volume 8317, 83171R (2012) https://doi.org/10.1117/12.911017

KEYWORDS: Targeting Task Performance metric, Tissues, Diffusion weighted imaging, Hemodynamics, Brain mapping, Brain, Magnetic resonance imaging, Data modeling, Neuroimaging, Arteries

Read Abstract +

Proceedings Article | 15 March 2011 Paper

Paper

Direction-dependent level set segmentation of cerebrovascular structures

Nils Daniel Forkert, Dennis Säring, Till Illies, Jens Fiehler, Jan Ehrhardt, Heinz Handels, Alexander Schmidt-Richberg

Proceedings Volume 7962, 79623S (2011) https://doi.org/10.1117/12.877942

KEYWORDS: Image segmentation, Image enhancement, Fuzzy logic, Visualization, Brain, 3D image processing, Image filtering, Tissues, 3D-TOF imaging, Health informatics

Read Abstract +

Proceedings Article | 13 March 2010 Paper

Paper

Closing of interrupted vascular segmentations: an automatic approach based on shortest paths and level sets

Nils Daniel Forkert, Alexander Schmidt-Richberg, Dennis Säring, Till Illies, Jens Fiehler, Heinz Handels

Proceedings Volume 7623, 76233G (2010) https://doi.org/10.1117/12.844210

KEYWORDS: Image segmentation, Visualization, Medical research, 3D image processing, Vascular diseases, Analytical research, Finite element methods, 3D-TOF imaging, Binary data, Neuroimaging

Read Abstract +

Proceedings Article | 13 March 2009 Paper

Paper

Analysis and dynamic 3D visualization of cerebral blood flow combining 3D and 4D MR image sequences

Nils Daniel Forkert, Dennis Säring, Jens Fiehler, Till Illies, Dietmar Möller, Heinz Handels

Proceedings Volume 7261, 726133 (2009) https://doi.org/10.1117/12.813742

KEYWORDS: Visualization, 3D modeling, Hemodynamics, 3D image processing, Cerebral blood flow, Image visualization, Temporal resolution, 3D visualizations, Magnetic resonance imaging, Data modeling

Read Abstract +

Showing 5 of 22 publications

Conference Committee Involvement (1)

Imaging Informatics

17 February 2025 | San Diego, California, United States

SIGN IN TO:

View contact details

UPDATE YOUR PROFILE

Is this your profile? Update it now.

Sign into your SPIE.org account

Don’t have a profile and want one?

Create an account on SPIE.org

Advertisement

Advertisement