Tommy Löfstedt

Docent and associate professor in computing science. Doing research in machine learning and computer vision, often with applications in medical image analysis or the life sciences.

Research qualifications: Docent

0000-0001-7119-7646

Contact

tommy.lofstedt@umu.se

Works as

Associate professor at Department of Computing Science

MIT-huset, Umeå universitet, MIT.E.252 Umeå universitet, 901 87 Umeå

Affiliated as other position at Department of Mathematics and Mathematical Statistics

MIT-huset, plan 3, Matematik och matematisk statistik Umeå universitet, 901 87 Umeå

Much of the recent successes within the field of artificial intelligence (AI) in the last 10-15 years can be attributed to a subarea of AI called machine learning. I am a researcher in machine learning and computer vision, often with applications in medical image analysis. Much of my research is developing novel methods for automatic segmentation, and to develop novel regularisation methods for machine learning to, among other things, improve the possibilities to interpret the results.

We are currently looking for a post-doctoral researcher in generative modelling in machine learning:

https://www.umu.se/en/work-with-us/postdoctoral-scholarships/6-1545-24/

We are currently offering master's thesis projects in the following areas:

Medical image segmentation
Reinforcement learning

From promise to practice: a study of common pitfalls behind the generalization gap in machine learning

Transactions on Machine Learning Research

Ghanbari Azar, Saeideh; Tronchin, Lorenzo; Simkó, Attila; et al.

Structured regularization with object size selection using mathematical morphology

Pattern Analysis and Applications, Springer Nature 2025, Vol. 28

Lin, Disi; Hägg, Linus; Wadbro, Eddie; et al.

Structured regularization using approximate morphology for Alzheimer's disease classification

2025 IEEE 22nd International Symposium on Biomedical Imaging (ISBI) : 1-4

Lin, Disi; Hägg, Linus; Wadbro, Eddie; et al.

Compressing regularized dynamics improves link prediction with the map equation in sparse networks

Physical review. E, American Physical Society 2025, Vol. 111, (5)

Lindström, Maja; Blöcker, Christopher; Löfstedt, Tommy; et al.

Matematikångest och bristande förkunskaper

Didaktisk dialog: i kursen Didaktik för universitetslärare, Umeå: Umeå University 2025 : 112-117

Löfstedt, Tommy

A training rate and survival heuristic for inference and robustness evaluation (Trashfire)

Proceedings of 2024 International Conference on Machine Learning and Cybernetics, IEEE 2025 : 613-623

Meyers, Charles; Saleh Sedghpour, Mohammad Reza; Löfstedt, Tommy; et al.

Using the Krylov subspace formulation to improve regularisation and interpretation in partial least squares regression

Computational statistics (Zeitschrift)

Löfstedt, Tommy

Massively parallel evasion attacks and the pitfalls of adversarial retraining

EAI Endorsed Transactions on Internet of Things, Gent EAI 2024, Vol. 10

Meyers, Charles; Löfstedt, Tommy; Elmroth, Erik

Towards MR contrast independent synthetic CT generation

Zeitschrift für Medizinische Physik, Elsevier 2024, Vol. 34, (2) : 270-277

Simkó, Attila; Bylund, Mikael; Jönsson, Gustav; et al.

Compressing the Activation Maps in Deep Neural Networks and Its Regularizing Effect

Transactions on Machine Learning Research

Vu, Minh Hoang; Garpebring, Anders; Nyholm, Tufve; et al.

Transformer-based deep learning for predicting protein properties in the life sciences

eLIFE, eLife Sciences Publications 2023, Vol. 12

Chandra, Abel; Tünnermann, Laura; Löfstedt, Tommy; et al.

Denoising and uncertainty estimation in parameter mapping with approximate Bayesian deep image priors

Magnetic Resonance in Medicine, John Wiley & Sons 2023, Vol. 90, (6) : 2557-2571

Hellström, Max; Löfstedt, Tommy; Garpebring, Anders

Safety-critical computer vision: an empirical survey of adversarial evasion attacks and defenses on computer vision systems

Artificial Intelligence Review, Elsevier 2023, Vol. 56 : 217-251

Meyers, Charles; Löfstedt, Tommy; Elmroth, Erik

Reproducibility of the methods in medical imaging with deep learning

Medical imaging with deep learning 2023, ML Research Press 2023 : 95-106

Simkó, Attila; Garpebring, Anders; Jonsson, Joakim; et al.

Improving MR image quality with a multi-task model, using convolutional losses

BMC Medical Imaging, BioMed Central (BMC) 2023, Vol. 23, (1)

Simkó, Attila; Ruiter, Simone; Löfstedt, Tommy; et al.

QU-BraTS: MICCAI BraTS 2020 Challenge on QuantifyingUncertainty in Brain Tumor Segmentation - Analysis of Ranking Scores and Benchmarking Results

Journal of Machine Learning for Biomedical Imaging : 1-54

Mehta, Raghav; Filos, Angelos; Baid, Ujjwal; et al.

MRI bias field correction with an implicitly trained CNN

Proceedings of the 5th international conference on medical imaging with deep learning, ML Research Press 2022 : 1125-1138

Simkó, Attila; Löfstedt, Tommy; Garpebring, Anders; et al.

A Data-Adaptive Loss Function for Incomplete Data and Incremental Learning in Semantic Image Segmentation

IEEE Transactions on Medical Imaging, IEEE 2022, Vol. 41, (6) : 1320-1330

Vu, Minh Hoang; Norman, Gabriella; Nyholm, Tufve; et al.

Changing the Contrast of Magnetic Resonance Imaging Signals using Deep Learning

Proceedings of the Fourth Conference on Medical Imaging with Deep Learning, PMLR, Lübeck University; Hamburg University of Technology 2021 : 713-727

Simkó, Attila; Löfstedt, Tommy; Garpebring, Anders; et al.

Multi-decoder Networks with Multi-denoising Inputs for Tumor Segmentation

Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries, Springer 2021 : 412-423

Vu, Minh Hoang; Nyholm, Tufve; Löfstedt, Tommy

Latent space manipulation for high-resolution medical image synthesis via the StyleGAN

Zeitschrift für Medizinische Physik, Elsevier 2020, Vol. 30, (4) : 305-314

Fetty, Lukas; Bylund, Mikael; Kuess, Peter; et al.

Investigating conditional GAN performance with different generator architectures, an ensemble model, and different MR scanners for MR-sCT conversion

Physics in Medicine and Biology, Institute of Physics Publishing (IOPP) 2020, Vol. 65, (10)

Fetty, Lukas; Löfstedt, Tommy; Heilemann, Gerd; et al.

Bayesian non-linear regression with spatial priors for noise reduction and error estimation in quantitative MRI with an application in T1 estimation

Physics in Medicine and Biology, Institute of Physics (IOP) 2020, Vol. 65, (22)

Löfstedt, Tommy; Hellström, Max; Bylund, Mikael; et al.

Evaluation of multislice inputs to convolutional neural networks for medical image segmentation

Medical physics (Lancaster), John Wiley & Sons 2020, Vol. 47, (12) : 6216-6231

Vu, Minh H.; Grimbergen, Guus; Nyholm, Tufve; et al.

A Question-Centric Model for Visual Question Answering in Medical Imaging

IEEE Transactions on Medical Imaging, IEEE 2020, Vol. 39, (9) : 2856-2868

Vu, Minh Hoang; Löfstedt, Tommy; Nyholm, Tufve; et al.

TuNet: End-to-End Hierarchical Brain Tumor Segmentation Using Cascaded Networks

Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries: 5th International Workshop, BrainLes 2019, Held in Conjunction with MICCAI 2019, Shenzhen, China, October 17, 2019, Revised Selected Papers, Part I, Cham: Springer 2020 : 174-186

Vu, Minh Hoang; Nyholm, Tufve; Löfstedt, Tommy

Network Regularization in Imaging Genetics Improves Prediction Performances and Model Interpretability on Alzheimer’s Disease

2019 IEEE 16th International Symposium on Biomedical Imaging (ISBI 2019), IEEE 2019 : 1403-1406

Guigui, N.; Philippe, C.; Gloaguen, A.; et al.

A constrained singular value decomposition method that integrates sparsity and orthogonality

PLOS ONE, Vol. 14, (3)

Guillemot, Vincent; Beaton, Derek; Gloaguen, Arnaud; et al.

Gray-level invariant Haralick texture features

PLOS ONE, Public Library of Science 2019, Vol. 14, (2)

Löfstedt, Tommy; Brynolfsson, Patrik; Nyholm, Tufve; et al.

Localization Network and End-to-End Cascaded U-Nets for Kidney Tumor Segmentation

MICCAI 2019, Shenzhen, China, Oct 13-17, 2019

Vu, Minh; Grimbergen, Guus; Simkó, Attila; et al.

Ensemble of Streamlined Bilinear Visual Question Answering Models for the ImageCLEF 2019 Challenge in the Medical Domain

CLEF 2019: Working Notes of CLEF 2019 - Conference and Labs of the Evaluation Forum

Vu, Minh Hoang; Sznitman, Raphael; Nyholm, Tufve; et al.

Gray-level invariant Haralick texture features

Radiotherapy and Oncology, Elsevier 2018, Vol. 127 : S279-S280

Brynolfsson, Patrik; Löfstedt, Tommy; Asklund, Thomas; et al.

Using deep learning to generate synthetic CTs for radiotherapy treatment planning

Radiotherapy and Oncology, Elsevier 2018, Vol. 127 : S283-S283

Bylund, Mikael; Jonsson, Joakim; Lundman, Josef; et al.

Prediction of activation patterns preceding hallucinations in patients with schizophrenia using machine learning with structured sparsity

Human Brain Mapping, Vol. 39, (4) : 1777-1788

de Pierrefeu, Amicie; Fovet, Thomas; Hadj-Selem, Fouad; et al.

Structured Sparse Principal Components Analysis With the TV-Elastic Net Penalty

IEEE Transactions on Medical Imaging, IEEE 2018, Vol. 37, (2) : 396-407

de Pierrefeu, Amicie; Löfstedt, Tommy; Hadj-Selem, Fouad; et al.

Identifying a neuroanatomical signature of schizophrenia, reproducible across sites and stages, using machine-learning with structured sparsity

Acta Psychiatrica Scandinavica, John Wiley & Sons 2018, Vol. 138 : 571-580

de Pierrefeu, Amicie; Löfstedt, Tommy; Laidi, C.; et al.

Interpretable and stable prediction of schizophrenia on a large multisite dataset using machine learning with structured sparsity

2018 International Workshop on Pattern Recognition in Neuroimaging (PRNI)

De Pierrefeu, Amicie; Löfstedt, Tommy; Laidi, Charles; et al.

Density Estimation of Grey-Level Co-Occurrence Matrices for Image Texture Analysis

Physics in Medicine and Biology, Institute of Physics and Engineering in Medicine 2018, Vol. 63, (19) : 9-15

Garpebring, Anders; Brynolfsson, Patrik; Kuess, Peter; et al.

Parameter estimation using weighted total least squares in the two-compartment exchange model

Magnetic Resonance in Medicine, John Wiley & Sons 2018, Vol. 79, (1) : 561-567

Garpebring, Anders; Tommy, Löfstedt

Continuation of Nesterov's Smoothing for Regression With Structured Sparsity in High-Dimensional Neuroimaging

IEEE Transactions on Medical Imaging, IEEE 2018, Vol. 37, (11) : 2403-2413

Hadj-Selem, Fouad; Löfstedt, Tommy; Dohmatob, Elvis; et al.

MLCN 2018 preface

Understanding and interpreting machine learning in medical image computing applications: First International Workshops MLCN 2018, DLF 2018, and iMIMIC 2018, Held in Conjunction with MICCAI 2018, Granada, Spain, September 16-20, 2018, Proceedings

Kia, Seyed Mostafa; Marquand, Andre; Duchesnay, Edouard; et al.

Simulated Data for Linear Regression with Structured and Sparse Penalties: Introducing pylearn-simulate

Journal of Statistical Software, Foundation for Open Access Statistics 2018, Vol. 87, (3)

Löfstedt, Tommy; Guillemot, Vincent; Frouin, Vincent; et al.

Structured Variable Selection for Regularized Generalized Canonical Correlation Analysis

MULTIPLE FACETS OF PARTIAL LEAST SQUARES AND RELATED METHODS, SPRINGER INT PUBLISHING AG 2016 : 129-139

Löfstedt, Tommy; Hadj-Selem, Fouad; Guillemot, Vincent; et al.

Near infrared and skin impedance spectroscopy: a possible support in the diagnostic process of skin tumours in primary health care

Skin research and technology, John Wiley & Sons 2015, Vol. 21, (4) : 493-499

Hägerlind, E.; Falk, M.; Löfstedt, Tommy; et al.

Preliminary evaluation of dorsal muscle activity during resisted cervical extension in patients with longstanding pain and disability following anterior cervical decompression and fusion surgery

Physiotherapy, Vol. 101, (1) : 69-74