Search

Deep learning has the potential to be used to improve and streamline EEG analysis. At the present, classifiers and supervised learning dominate the field. Supervised learning depends on target labels which most often are created by human experts manually classifying data. A problem with supervised learning is intra- and interrater agreement which in some instances are far from perfect. This can affect the training and make evaluation more difficult. This thesis includes three papers where self-supervised deep neural networks were developed. In self-supervised learning, the input data to the networks themselves contain structures that are used as targets for the training and no labeling is necessary. In paper I, deep neural networks were trained to increase the number of-, or to recreate missing EEG-channels. The performance was at least on the same level as that of spherical interpolation, but unlike in the case of interpolation, missing data does not have to be identified manually first. Papers II and III involved developing deep neural networks for clustering analysis. The networks produced two-dimensional representations of EEG data and the training strategy was based on the principle of t-distributed stochastic neighbor embedding (t-SNE). In paper II, comparisons were made to parametric t-SNE and EEG-features obtained from time-frequency methods. The deep neural networks produced more distinct clustering when tested on data annotated for epileptiform discharges, seizure activity, or sleep-wakefulness. In paper III, the newly developed method was used to compare annotations of epileptiform discharges. Two experts performed independent annotations and classifiers were trained on these, using supervised learning, which in turn produced new annotations. The agreement when comparing two sets of annotations was not larger between the two experts than between an expert and a classifier. The analysis showed that differences in the annotations by the experts infl, Funding: The work was funded by Saad Nagi (grants RÖ-974228, RÖ-962769, and RÖ-941377), Magnus Thordstein (grant RÖ-986017), and Håkan Olaus-son (grants LIO-936017 and RÖ-941359).

t-distributed stochastic neighbor embedding (t-SNE) is a method for reducing high-dimensional data to a low-dimensional representation, and is mostly used for visualizing data. In parametric t-SNE, a neural network learns to reproduce this mapping. When used for EEG analysis, the data are usually first transformed into a set of features, but it is not known which features are optimal. The principle of t-SNE was used to train convolutional neural network (CNN) encoders to learn to produce both a high- and a low-dimensional representation, eliminating the need for feature engineering. To evaluate the method, the Temple University EEG Corpus was used to create three datasets with distinct EEG characters: (1) wakefulness and sleep; (2) interictal epileptiform discharges; and (3) seizure activity. The CNN encoders produced low-dimensional representations of the datasets with a structure that conformed well to the EEG characters and generalized to new data. Compared to parametric t-SNE for either a short-time Fourier transform or wavelet representation of the datasets, the developed CNN encoders performed equally well in separating categories, as assessed by support vector machines. The CNN encoders generally produced a higher degree of clustering, both visually and in the number of clusters detected by k-means clustering. The developed principle is promising and could be further developed to create general tools for exploring relations in EEG data., Funding: Linkoping University; University Hospital of Linkoeping; ALF of Region OEstergoetland [LIO-936176, ROE-941359]; ITEA3/VINNOVA

Enterotoxigenic Escherichia coli (ETEC) attach to the host epithelium in the intestinal tract via specific adhesion organelles expressed on the cell membrane. We investigate, by force measuring optical tweezers, the intrinsic biomechanical properties and kinetics of the colonization factor I (CFA/I) at a single pilus level. The measurements indicate that CFA/I pili are helix-like structures that can both be unraveled to a linearized polymer by applying a small external force, 7.5 ± 1.5 pN but also regain its helix-like structure when the applied force is reduced. The data confirm that layer-to-layer interactions, that stabilize the helix-like structure, are much weaker than the interactions found in pili expressed by Uropathogenic Escherichia coli (UPEC). It is also found, contrary to previous results assessed from UPEC pili, that the CFA/I undergo in some cases a sudden structural change, a force drop of ~2 pN, when unraveled from the helix-like configuration to an open helical linearized fiber. These data suggest a rotation of the filament about its helical axis, followed by a region in which the force required to extend the pili further increases rapidly. During this final elongation to a super-extended fiber, CFA/I pili do not show any structural transition as seen for UPEC pili. In addition, the CFA/I pili show faster kinetics than UPEC pili that allows for a larger dynamic regime of in vivo shear forces. The unfolding and refolding possibility points toward an organelle that has evolved to allow for dynamic damping of external forces and handling of harsh motion without breaking.