Stroke tissue outcome prediction

Acute ischemic stroke occurs when blood flow through a cerebral artery is interrupted, depriving the brain of oxygen in a localized area. Until this interruption (typically a blood clot) is resolved, a patient may lose 120 million neurons and 830 billion synapses per hour, which is equivalent to nearly 4 years of natural aging. When “time is brain,” access to the most efficient treatment options and the ability to make rapid treatment decisions are two important factors determining a patient’s outcome.

Using machine learning, we are developing models to predict acute ischemic stroke outcomes under different treatment conditions using the patient’s clinical and brain imaging data. These models can be applied prospectively to new stroke patients to help clinicians make personalized treatment decisions quickly and effectively. Furthermore, these can also be applied retrospectively to large databases of stroke patients to help researchers identify promising new stroke treatments. Currently, we are exploring deep learning models for the prediction of both voxel-wise tissue outcomes and patient-level functional outcomes.

Voxelwise classification

Publications

André Kemmling, Fabian Flottmann, Nils D. Forkert, Jens Minnerup, Götz Thomalla, Walter Heindel, Bernd Eckert, Michael Knauth, Marios Psychogios, Sönke Langner, Jens Fiehler:
Multivariate dynamic prediction of ischemic infarction and tissue salvage as a function of time and degree of recanalization. Journal of Cerebral Blood Flow & Metabolism, 35(9), 1397-1405, 2015.

Nils D. Forkert, Jens Fiehler: Effect of sample size on multi-parametric prediction of tissue outcome in acute ischemic stroke using a random forest classifier. Gimi B., Molthen R.C. (eds), Biomedical Applications in Molecular, Structural, and Functional Imaging, SPIE Medical Imaging 2015, Vol. 9417, 94172H1-7, Orlando, USA, 2015.
 
Nils D. Forkert, Susanne Siemonsen, Michael Dalski, Tobias Verleger, Andre Kemmling, Jens Fiehler: 
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? Molthen R.C., Weaver J.B. (eds.), Biomedical Applications in Molecular, Structural, and Functional Imaging, SPIE Medical Imaging 2014, Vol. 9038, 90381O1-7, San Diego, USA, 2014.

Nils D. Forkert, Jens Fiehler, Susanne Siemonsen, Andre Kemmling: Multiparametric prediction of acute ischemic stroke tissue outcome using CT perfusion datasets. Weaver J.B., Molthen R.C. (eds.), Biomedical Applications in Molecular, Structural, and Functional Imaging, SPIE Medical Imaging 2013, Vol. 8672, 86721Y1-7, Orlando, USA, 2013.

Anthony J. Winder, Susanne Siemonsen, Fabian Flottmann, Götz Thomalla, Jens Fiehler, Nils D. Forkert:
Technical considerations of multi-parametric tissue outcome prediction methods in acute ischemic stroke patients. Scientific Reports, 9, article 13208, 2019. https://doi.org/10.1038/s41598-019-49460-y

Anthony J. Winder, Christopher d’Esterre, Bijoy Menon, Jens Fiehler, Nils D. Forkert: Automatic arterial input function selection in CT and MR perfusion datasets using deep convolutional neural networks. Medical Physics, 47(9), 2020. https://doi.org/10.1002/mp.14351 

Malte Grosser, Susanne Gellissen, Patrick Borchert, Jan Sedlacik, Jawed Nawabi, Jens Fiehler, Nils D. Forkert: Improved multi-parametric prediction of tissue outcome in acute ischemic stroke patients using spatial features. PLOS One, 15(1), e0228113, 2020.

Malte Grosser, Susanne Gellissen, Patrick Borchert, Jan Sedlacik, Jawed Nawabi, Jens Fiehler, Nils D. Forkert: Localized prediction of tissue outcome in acute ischemic stroke patients using diffusion- and perfusion-weighted MRI datasets. PLOS One, 15(11), e0241917, 2020.

Connor McDougall, Leona Chan, Surbhi Sachan, Jen Guo, Rani G. Sah, Bijoy Menon, Andrew M. Demchuk, Michael D. Hill, Nils D. Forkert, Christopher D. d’Esterre, Philip A. Barber: Dynamic CTA-derived perfusion maps predict final infarct volume – the simple perfusion reconstruction algorithm. American Journal of Neuroradiology, 2020 [Epub ahead of print]
 

Team members

  • Anthony Winder

  • Kimberly Amador

  • Nils D. Forkert