arXiv subjects

Christian Kothe

Publications and source records attributed to Christian Kothe.

6 recordsLinked to original sources

arXiv2026

EEGPrep: a validated Python implementation of the EEGLAB preprocessing pipeline

Objective. Automated EEG preprocessing is common in research and clinical work, but few pipelines have been tested systematically. In a recent benchmark, the default EEGLAB pipeline was the only pipeline that significantly outperformed simple high pass filtering. Its MATLAB implementation, however, complicates deployment in Python and cloud workflows. We developed EEGPrep to reproduce this pipeline in Python while supporting BIDS data. The main technical problem was numerical: small floating point differences can accumulate during recursive filtering and ICA. Approach. EEGPrep covers the default EEGLAB workflow: average rereferencing, artifact removal with the clean rawdata plugin, Picard ICA, ICLabel component classification, channel interpolation, and epoching. We compared each stage with MATLAB reference output on ARM arm64, the primary analysis, and Intel x86 64, the supplementary analysis. The test dataset contained 64 channel P300 auditory oddball recordings from 13 participants. We measured maximum absolute difference, RMS error, AMARI distance for ICA, and agreement between ICLabel decisions. Main Results. On ARM arm64, import and rereferencing matched exactly for all 12 analysed subjects. The clean rawdata and Picard ICA stages remained at numerical zero, with maximum RMS equal to 1.5 x 10^-12 microvolts, AMARI distance less than or equal to 0.000001, and mean correlation equal to 1.000. ICLabel neural network inference introduced the only measurable difference, with maximum RMS equal to 2.0 x 10^-5 microvolts. Rejection decisions nevertheless agreed for every subject, and the difference did not increase through interpolation and epoching, with end to end maximum RMS less than or equal to 2.1 x 10^-5 microvolts. The Intel x86 64 analysis matched to the same precision. Significance. EEGPrep reproduces the validated EEGLAB pipeline and can be installed from PyPI or run in Docker.

eess.SP↗

arXiv2023

Decoding Working-Memory Load During n-Back Task Performance from High Channel NIRS Data

Near-infrared spectroscopy (NIRS) can measure neural activity through blood oxygenation changes in the brain in a wearable form factor, enabling unique applications for research in and outside the lab. NIRS has proven capable of measuring cognitive states such as mental workload, often using machine learning (ML) based brain-computer interfaces (BCIs). To date, NIRS research has largely relied on probes with under ten to several hundred channels, although recently a new class of wearable NIRS devices with thousands of channels has emerged. This poses unique challenges for ML classification, as NIRS is typically limited by few training trials which results in severely under-determined estimation problems. So far, it is not well understood how such high-resolution data is best leveraged in practical BCIs and whether state-of-the-art (SotA) or better performance can be achieved. To address these questions, we propose an ML strategy to classify working-memory load that relies on spatio-temporal regularization and transfer learning from other subjects in a combination that has not been used in previous NIRS BCIs. The approach can be interpreted as an end-to-end generalized linear model and allows for a high degree of interpretability using channel-level or cortical imaging approaches. We show that using the proposed methodology, it is possible to achieve SotA decoding performance with high-resolution NIRS data. We also replicated several SotA approaches on our dataset of 43 participants wearing a 3198 dual-channel NIRS device while performing the n-Back task and show that these existing methods struggle in the high-channel regime and are largely outperformed by the proposed method. Our approach helps establish high-channel NIRS devices as a viable platform for SotA BCI and opens new applications using this class of headset while also enabling high-resolution model imaging and interpretation.

eess.SP↗

arXiv2012

Experimental determination of the degree of quantum polarisation of continuous variable states

We demonstrate excitation-manifold resolved polarisation characterisation of continuous-variable (CV) quantum states. In contrast to traditional characterisation of polarisation that is based on the Stokes parameters, we experimentally determine the Stokes vector of each excitation manifold separately. Only for states with a given photon number does the methods coincide. For states with an indeterminate photon number, for example Gaussian states, the employed method gives a richer and more accurate description. We apply the method both in theory and in experiment to some common states to demonstrate its advantages.

quant-ph↗

arXiv2011

On the efficiency of quantum lithography

Quantum lithography promises, in principle, unlimited feature resolution, independent of wavelength. However, in the literature at least two different theoretical descriptions of quantum lithography exist. They differ in to which extent they predict that the photons retain spatial correlation from generation to the absorption, and while both predict the same feature size, they differ vastly in predicting how efficiently a quantum lithographic pattern can be exposed. Until recently, essentially all experiments reported have been performed in such a way that it is difficult to distinguish between the two theoretical explanations. However, last year an experiment was performed which gives different outcomes for the two theories. We comment on the experiment and show that the model that fits the data unfortunately indicates that the trade-off between resolution and efficiency in quantum lithography is very unfavourable.

quant-ph↗

arXiv2010

Arbitrarily High Super-Resolving Phase Measurements at Telecommunication Wavelengths

We present two experiments that achieve phase super-resolution at telecommunication wavelengths. One of the experiments is realized in the space domain and the other in the time domain. Both experiments show high visibilities and are performed with standard lasers and single-photon detectors. The first experiment uses six-photon coincidences, whereas the latter needs no coincidence measurements, is easy to perform, and achieves, in principle, arbitrarily high phase super-resolution. Here, we demonstrate a 30-fold increase of the resolution. We stress that neither entanglement nor joint detection is needed in these experiments, demonstrating that neither is necessary to achieve phase super-resolution.

quant-ph↗

arXiv2006

Entanglement quantification through local observable correlations

We present a significantly improved scheme of entanglement detection inspired by local uncertainty relations for a system consisting of two qubits. Developing the underlying idea of local uncertainty relations, namely correlations, we demonstrate that it's possible to define a measure which is invariant under local unitary transformations and which is based only on local measurements. It is quite simple to implement experimentally and it allows entanglement quantification in a certain range for mixed states and exactly for pure states, without first obtaining full knowledge (e.g. through tomography) of the state.

quant-ph↗