no code implementations • 22 Feb 2024 • Tiezhi Wang, Nils Strodthoff
This study aims to elucidate the significance of long-range correlations for deep-learning-based sleep staging.
1 code implementation • 12 Jan 2024 • Stefan Blücher, Johanna Vielhaben, Nils Strodthoff
The R-OMS score enables a systematic comparison of occlusion strategies and resolves the disagreement problem by grouping consistent PF rankings.
1 code implementation • 18 Dec 2023 • Nils Strodthoff, Juan Miguel Lopez Alcaraz, Wilhelm Haverkamp
Current deep learning algorithms designed for automatic ECG analysis have exhibited notable accuracy.
1 code implementation • 11 Oct 2023 • Gabriel Ott, Yannik Schaubelt, Juan Miguel Lopez Alcaraz, Wilhelm Haverkamp, Nils Strodthoff
In this paper we present the following contributions: (1) We employ a deep-learning model and a tree-based model to analyze ECG data from a robust dataset of healthy individuals across varying ages in both raw signals and ECG feature format.
1 code implementation • 10 Oct 2023 • Tiezhi Wang, Nils Strodthoff
Scoring sleep stages in polysomnography recordings is a time-consuming task plagued by significant inter-rater variability.
1 code implementation • 7 Sep 2023 • Markus Wenzel, Erik Grüner, Nils Strodthoff
Motivation: We explored how explainable artificial intelligence (XAI) can help to shed light into the inner workings of neural networks for protein function prediction, by extending the widely used XAI method of integrated gradients such that latent representations inside of transformer models, which were finetuned to Gene Ontology term and Enzyme Commission number prediction, can be inspected too.
Explainable artificial intelligence Explainable Artificial Intelligence (XAI) +1
1 code implementation • 29 Aug 2023 • Temesgen Mehari, Nils Strodthoff
These components consistently enhance performance beyond the existing state-of-the-art, which is predominantly based on convolutional models.
1 code implementation • 26 May 2023 • Patrick Wagner, Temesgen Mehari, Wilhelm Haverkamp, Nils Strodthoff
Furthermore, we demonstrate how these XAI techniques can be utilized for knowledge discovery, such as identifying subtypes of myocardial infarction.
1 code implementation • 5 Apr 2023 • Temesgen Mehari, Ashish Sundar, Alen Bosnjakovic, Peter Harris, Steven E. Williams, Axel Loewe, Olaf Doessel, Claudia Nagel, Nils Strodthoff, Philip J. Aston
Feature importance methods promise to provide a ranking of features according to importance for a given classification task.
1 code implementation • 27 Jan 2023 • Johanna Vielhaben, Stefan Blücher, Nils Strodthoff
For the trustworthy application of XAI, in particular for high-stake decisions, a more global model understanding is required.
1 code implementation • 19 Jan 2023 • Juan Miguel Lopez Alcaraz, Nils Strodthoff
Synthetic data generation is a promising solution to address privacy issues with the distribution of sensitive health data.
1 code implementation • 14 Nov 2022 • Temesgen Mehari, Nils Strodthoff
The field of deep-learning-based ECG analysis has been largely dominated by convolutional architectures.
1 code implementation • 19 Aug 2022 • Juan Miguel Lopez Alcaraz, Nils Strodthoff
The imputation of missing values represents a significant obstacle for many real-world data analysis pipelines.
1 code implementation • 11 Apr 2022 • Maximilian Springenberg, Annika Frommholz, Markus Wenzel, Eva Weicken, Jackie Ma, Nils Strodthoff
While machine learning is currently transforming the field of histopathology, the domain lacks a comprehensive evaluation of state-of-the-art models based on essential but complementary quality requirements beyond a mere classification accuracy.
no code implementations • 11 Mar 2022 • Johanna Vielhaben, Stefan Blücher, Nils Strodthoff
We empirically demonstrate the soundness of the proposed Sparse Subspace Clustering for Concept Discovery (SSCCD) method for a variety of different image classification tasks.
no code implementations • 25 Jun 2021 • Vignesh Srinivasan, Nils Strodthoff, Jackie Ma, Alexander Binder, Klaus-Robert Müller, Wojciech Samek
Our results indicate that models initialized from ImageNet pretraining report a significant increase in performance, generalization and robustness to image distortions.
1 code implementation • 16 Apr 2021 • Johanna Vielhaben, Markus Wenzel, Eva Weicken, Nils Strodthoff
Predicting the binding of viral peptides to the major histocompatibility complex with machine learning can potentially extend the computational immunology toolkit for vaccine development, and serve as a key component in the fight against a pandemic.
1 code implementation • 23 Mar 2021 • Temesgen Mehari, Nils Strodthoff
In a first step, we learn contrastive representations and evaluate their quality based on linear evaluation performance on a recently established, comprehensive, clinical ECG classification task.
2 code implementations • 26 Feb 2021 • Stefan Blücher, Johanna Vielhaben, Nils Strodthoff
PredDiff is a model-agnostic, local attribution method that is firmly rooted in probability theory.
1 code implementation • 18 Dec 2020 • Johanna Vielhaben, Nils Strodthoff
Generative neural samplers offer a complementary approach to Monte Carlo methods for problems in statistical physics and quantum field theory.
no code implementations • 19 Oct 2020 • Nils Strodthoff, Claas Strodthoff, Tobias Becher, Norbert Weiler, Inéz Frerichs
Electrical impedance tomography (EIT) is a noninvasive imaging modality that allows a continuous assessment of changes in regional bioimpedance of different organs.
2 code implementations • 28 Apr 2020 • Nils Strodthoff, Patrick Wagner, Tobias Schaeffter, Wojciech Samek
Electrocardiography is a very common, non-invasive diagnostic procedure and its interpretation is increasingly supported by automatic interpretation algorithms.
no code implementations • 3 Mar 2020 • Stefan Bluecher, Lukas Kades, Jan M. Pawlowski, Nils Strodthoff, Julian M. Urban
Machine learning has the potential to aid our understanding of phase structures in lattice quantum field theories through the statistical analysis of Monte Carlo samples.
no code implementations • 29 Oct 2019 • Kim A. Nicoli, Shinichi Nakajima, Nils Strodthoff, Wojciech Samek, Klaus-Robert Müller, Pan Kessel
We propose a general framework for the estimation of observables with generative neural samplers focusing on modern deep generative neural networks that provide an exact sampling probability.
no code implementations • 3 Jun 2019 • Jan Laermann, Wojciech Samek, Nils Strodthoff
We study the recently introduced stability training as a general-purpose method to increase the robustness of deep neural networks against input perturbations.
no code implementations • 26 Mar 2019 • Kim Nicoli, Pan Kessel, Nils Strodthoff, Wojciech Samek, Klaus-Robert Müller, Shinichi Nakajima
In this comment on "Solving Statistical Mechanics Using Variational Autoregressive Networks" by Wu et al., we propose a subtle yet powerful modification of their approach.
no code implementations • 27 Jul 2018 • Nils Strodthoff, Barış Göktepe, Thomas Schierl, Cornelius Hellge, Wojciech Samek
We investigate Early Hybrid Automatic Repeat reQuest (E-HARQ) feedback schemes enhanced by machine learning techniques as a path towards ultra-reliable and low-latency communication (URLLC).
no code implementations • 18 Jun 2018 • Nils Strodthoff, Claas Strodthoff
Objective: We aim to provide an algorithm for the detection of myocardial infarction that operates directly on ECG data without any preprocessing and to investigate its decision criteria.
1 code implementation • 28 Oct 2016 • Anton K. Cyrol, Mario Mitter, Nils Strodthoff
We present FormTracer, a high-performance, general purpose, easy-to-use Mathematica tracing package which uses FORM.
High Energy Physics - Phenomenology High Energy Physics - Theory