Search

Photoplethysmography (PPG) is a widely used non-invasive physiological sensing technique, suitable for various clinical applications. Such clinical applications are increasingly supported by machine learning methods, raising the question of the most appropriate input representation and model choice. Comprehensive comparisons, in particular across different input representations, are scarce. We address this gap in the research landscape by a comprehensive benchmarking study covering three kinds of input representations, interpretable features, image representations and raw waveforms, across prototypical regression and classification use cases: blood pressure and atrial fibrillation prediction. In both cases, the best results are achieved by deep neural networks operating on raw time series as input representations. Within this model class, best results are achieved by modern convolutional neural networks (CNNs). but depending on the task setup, shallow CNNs are often also very competitive. We envision that these results will be insightful for researchers to guide their choice on machine learning tasks for PPG data, even beyond the use cases presented in this work., Comment: 39 pages, 9 figures, code available at https://gitlab.com/qumphy/d1-code

Photoplethysmography (PPG)-based blood pressure (BP) estimation represents a promising alternative to cuff-based BP measurements. Recently, an increasing number of deep learning models have been proposed to infer BP from the raw PPG waveform. However, these models have been predominantly evaluated on in-distribution test sets, which immediately raises the question of the generalizability of these models to external datasets. To investigate this question, we trained five deep learning models on the recently released PulseDB dataset, provided in-distribution benchmarking results on this dataset, and then assessed out-of-distribution performance on several external datasets. The best model (XResNet1d101) achieved in-distribution MAEs of 9.4 and 6.0 mmHg for systolic and diastolic BP respectively on PulseDB (with subject-specific calibration), and 14.0 and 8.5 mmHg respectively without calibration. Equivalent MAEs on external test datasets without calibration ranged from 15.0 to 25.1 mmHg (SBP) and 7.0 to 10.4 mmHg (DBP). Our results indicate that the performance is strongly influenced by the differences in BP distributions between datasets. We investigated a simple way of improving performance through sample-based domain adaptation and put forward recommendations for training models with good generalization properties. With this work, we hope to educate more researchers for the importance and challenges of out-of-distribution generalization., Comment: 20 pages, 5 figures, code available at https://github.com/AI4HealthUOL/ppg-ood-generalization

The development of electronic stethoscopes and wearable recording sensors opened the door to the automated analysis of bowel sound (BS) signals. This enables a data-driven analysis of bowel sound patterns, their interrelations, and their correlation to different pathologies. This work leverages a BS dataset collected from 16 healthy subjects that was annotated according to four established BS patterns. This dataset is used to evaluate the performance of machine learning models to detect and/or classify BS patterns. The selection of considered models covers models using tabular features, convolutional neural networks based on spectrograms and models pre-trained on large audio datasets. The results highlight the clear superiority of pre-trained models, particularly in detecting classes with few samples, achieving an AUC of 0.89 in distinguishing BS from non-BS using a HuBERT model and an AUC of 0.89 in differentiating bowel sound patterns using a Wav2Vec 2.0 model. These results pave the way for an improved understanding of bowel sounds in general and future machine-learning-driven diagnostic applications for gastrointestinal examinations, Comment: 9 pages, 6 figures and 1 table

Electrocardiogram (ECG) analysis has emerged as a promising tool for identifying physiological changes associated with neuropsychiatric conditions. The relationship between cardiovascular health and neuropsychiatric disorders suggests that ECG abnormalities could serve as valuable biomarkers for more efficient detection, therapy monitoring, and risk stratification. However, the potential of the ECG to accurately distinguish neuropsychiatric conditions, particularly among diverse patient populations, remains underexplored. This study utilized ECG markers and basic demographic data to predict neuropsychiatric conditions using machine learning models, with targets defined through ICD-10 codes. Both internal and external validation were performed using the MIMIC-IV and ECG-View datasets respectively. Performance was assessed using AUROC scores. To enhance model interpretability, Shapley values were applied to provide insights into the contributions of individual ECG features to the predictions. Significant predictive performance was observed for conditions within the neurological and psychiatric groups. For the neurological group, Alzheimer's disease (G30) achieved an internal AUROC of 0.813 (0.812-0.814) and an external AUROC of 0.868 (0.867-0.868). In the psychiatric group, unspecified dementia (F03) showed an internal AUROC of 0.849 (0.848-0.849) and an external AUROC of 0.862 (0.861-0.863). Discriminative features align with known ECG markers but also provide hints on potentially new markers. ECG offers significant promise for diagnosing and monitoring neuropsychiatric conditions, with robust predictive performance across internal and external cohorts. Future work should focus on addressing potential confounders, such as therapy-related cardiotoxicity, and expanding the scope of ECG applications, including personalized care and early intervention strategies., Comment: 9 pages, 2 figures, source code under https://github.com/AI4HealthUOL/CardioDiag

Background: Neoplasms remains a leading cause of mortality worldwide, with timely diagnosis being crucial for improving patient outcomes. Current diagnostic methods are often invasive, costly, and inaccessible to many populations. Electrocardiogram (ECG) data, widely available and non-invasive, has the potential to serve as a tool for neoplasms diagnosis by using physiological changes in cardiovascular function associated with neoplastic prescences. Methods: This study explores the application of machine learning models to analyze ECG features for the diagnosis of neoplasms. We developed a pipeline integrating tree-based models with Shapley values for explainability. The model was trained and internally validated and externally validated on a second large-scale independent external cohort to ensure robustness and generalizability. Findings: The results demonstrate that ECG data can effectively capture neoplasms-associated cardiovascular changes, achieving high performance in both internal testing and external validation cohorts. Shapley values identified key ECG features influencing model predictions, revealing established and novel cardiovascular markers linked to neoplastic conditions. This non-invasive approach provides a cost-effective and scalable alternative for the diagnosis of neoplasms, particularly in resource-limited settings. Similarly, useful for the management of secondary cardiovascular effects given neoplasms therapies. Interpretation: This study highlights the feasibility of leveraging ECG signals and machine learning to enhance neoplasms diagnostics. By offering interpretable insights into cardio-neoplasms interactions, this approach bridges existing gaps in non-invasive diagnostics and has implications for integrating ECG-based tools into broader neoplasms diagnostic frameworks, as well as neoplasms therapy management., Comment: 9 pages, 2 figures, code under https://github.com/AI4HealthUOL/CardioDiag

Vision transformers (ViTs) can be trained using various learning paradigms, from fully supervised to self-supervised. Diverse training protocols often result in significantly different feature spaces, which are usually compared through alignment analysis. However, current alignment measures quantify this relationship in terms of a single scalar value, obscuring the distinctions between common and unique features in pairs of representations that share the same scalar alignment. We address this limitation by combining alignment analysis with concept discovery, which enables a breakdown of alignment into single concepts encoded in feature space. This fine-grained comparison reveals both universal and unique concepts across different representations, as well as the internal structure of concepts within each of them. Our methodological contributions address two key prerequisites for concept-based alignment: 1) For a description of the representation in terms of concepts that faithfully capture the geometry of the feature space, we define concepts as the most general structure they can possibly form - arbitrary manifolds, allowing hidden features to be described by their proximity to these manifolds. 2) To measure distances between concept proximity scores of two representations, we use a generalized Rand index and partition it for alignment between pairs of concepts. We confirm the superiority of our novel concept definition for alignment analysis over existing linear baselines in a sanity check. The concept-based alignment analysis of representations from four different ViTs reveals that increased supervision correlates with a reduction in the semantic structure of learned representations., Comment: 19 pages, 17 figures, code: https://github.com/jvielhaben/NLMCD-ALIGN

Background: Liver diseases are a major global health concern, often diagnosed using resource-intensive methods. Electrocardiogram (ECG) data, widely accessible and non-invasive, offers potential as a diagnostic tool for liver diseases, leveraging the physiological connections between cardiovascular and hepatic health. Methods: This study applies machine learning models to ECG data for the diagnosis of liver diseases. The pipeline, combining tree-based models with Shapley values for explainability, was trained, internally validated, and externally validated on an independent cohort, demonstrating robust generalizability. Findings: Our results demonstrate the potential of ECG to derive biomarkers to diagnose liver diseases. Shapley values revealed key ECG features contributing to model predictions, highlighting already known connections between cardiovascular biomarkers and hepatic conditions as well as providing new ones. Furthermore, our approach holds promise as a scalable and affordable solution for liver disease detection, particularly in resource-limited settings. Interpretation: This study underscores the feasibility of leveraging ECG features and machine learning to enhance the diagnosis of liver diseases. By providing interpretable insights into cardiovascular-liver interactions, the approach bridges existing gaps in non-invasive diagnostics, offering implications for broader systemic disease monitoring., Comment: 8 pages, 3 images, code under https://github.com/AI4HealthUOL/CardioDiag

Background: Laboratory values are fundamental to medical diagnosis and management, but acquiring these values can be costly, invasive, and time-consuming. While electrocardiogram (ECG) patterns have been linked to certain laboratory abnormalities, the comprehensive modeling of these relationships remains underexplored. Methods: We utilize MIMIC-IV dataset to develop multimodal deep-learning models to demonstrate the feasibility of estimating (real-time) and monitoring (predict at future intervals) laboratory value abnormalities from ECG waveforms, demographics, biometrics, and vital signs. Results: The models exhibit a strong predictive performance with AUROC scores above 0.70 in a statistically significant manner for 23 laboratory values in the estimation setting and up to 26 values in the monitoring setting. Most notably, the accurately predictable values encompassing abnormalities across diverse physiological categories such as cardiac, renal, hematological, metabolic, immunological and coagulation. To name examples, for estimation NTproBNP (>353 pg/mL) with 0.882, whereas for monitoring at 30 minutes Urea nitrogen (<6 mg/dL) with 0.851, at 60 minutes creatinine (<0.5 mg/dL) with 0.85, and at 120 minutes hemoglobin (>17.5 g/dL) with 0.821. Conclusions: This study provides first evidence for the feasibility of using ECG data alongside clinical routine data for the real-time estimation and monitoring of laboratory value abnormalities, which could provide a non-invasive, cost-effective supplement to traditional laboratory testing, with strong implications for enhanced patient monitoring and early intervention. Further validation could facilitate their integration into routine clinical practice., Comment: 7 pages, 1 figure, code under https://github.com/AI4HealthUOL/CardioLab

Introduction: Ensuring timely and accurate diagnosis of medical conditions is paramount for effective patient care. Electrocardiogram (ECG) signals are fundamental for evaluating a patient's cardiac health and are readily available. Despite this, little attention has been given to the remarkable potential of ECG data in detecting non-cardiac conditions. Methods: In our study, we used publicly available datasets (MIMIC-IV-ECG-ICD and ECG-VIEW II) to investigate the feasibility of inferring general diagnostic conditions from ECG features. To this end, we trained a tree-based model (XGBoost) based on ECG features and basic demographic features to estimate a wide range of diagnoses, encompassing both cardiac and non-cardiac conditions. Results: Our results demonstrate the reliability of estimating 23 cardiac as well as 21 non-cardiac conditions above 0.7 AUROC in a statistically significant manner across a wide range of physiological categories. Our findings underscore the predictive potential of ECG data in identifying well-known cardiac conditions. However, even more striking, this research represents a pioneering effort in systematically expanding the scope of ECG-based diagnosis to conditions not traditionally associated with the cardiac system., Comment: 4 pages, source code under https://github.com/AI4HealthUOL/CardioDiag

Introduction: Laboratory value represents a cornerstone of medical diagnostics, but suffers from slow turnaround times, and high costs and only provides information about a single point in time. The continuous estimation of laboratory values from non-invasive data such as electrocardiogram (ECG) would therefore mark a significant frontier in healthcare monitoring. Despite its transformative potential, this domain remains relatively underexplored within the medical community. Methods: In this preliminary study, we used a publicly available dataset (MIMIC-IV-ECG) to investigate the feasibility of inferring laboratory values from ECG features and patient demographics using tree-based models (XGBoost). We define the prediction task as a binary prediction problem of predicting whether the lab value falls into low or high abnormalities. The model performance can then be assessed using AUROC. Results: Our findings demonstrate promising results in the estimation of laboratory values related to different organ systems based on a small yet comprehensive set of features. While further research and validation are warranted to fully assess the clinical utility and generalizability of ECG-based estimation in healthcare monitoring, our findings lay the groundwork for future investigations into approaches to laboratory value estimation using ECG data. Such advancements hold promise for revolutionizing predictive healthcare applications, offering faster, non-invasive, and more affordable means of patient monitoring., Comment: 4 pages, (updated dataset features set description version) code under https://github.com/AI4HealthUOL/CardioLab

Background: A clinically meaningful comparative assessment of medical decision support in emergency care is challenging due to a lack of appropriate datasets with multimodal input modalities and comprehensive prediction task. This hampers measurable progress in the field. Results: We introduce a dataset based on MIMIC-IV, a benchmarking protocol, and initial results for evaluating multimodal decision support in the emergency department (ED). We use diverse data modalities from the first 1.5 hours after patient arrival, including demographics, biometrics, vital signs, lab values, and electrocardiogram waveforms. We analyze 1443 clinical labels across two contexts: predicting diagnoses and patient deterioration. Our diagnostic model achieves an AUROC score over 0.8 in a statistically significant manner for 609 out of 1428 conditions, including cardiac conditions like myocardial infarction and non-cardiac conditions such as renal disease and diabetes. The deterioration model scores above 0.8 in a statistically significant manner for 14 out of 15 targets, including critical events like cardiac arrest, mechanical ventilation, intensive care unit admission, as well as short- and long-term mortality. Furthermore, we provide one of the first robust demonstrations of the significant impact of raw waveform input data on model performance. Conclusions: This study highlights the proposed dataset as a unique resource to foster progress towards measurable progress in the domain of algorithmic decision support in emergency care. The presented multimodal baseline models showcase the potential of diagnostic decision support in the field and provide strong incentives for including raw waveform data., Comment: 14 pages, 1 figure, code available under https://github.com/AI4HealthUOL/MDS-ED, dataset available under https://physionet.org/content/multimodal-emergency-benchmark/

Despite the excelling performance of machine learning models, understanding the decisions of machine learning models remains a long-standing goal. While commonly used attribution methods in explainable AI attempt to address this issue, they typically rely on associational rather than causal relationships. In this study, within the context of time series classification, we introduce a novel framework to assess the causal effect of concepts, i.e., predefined segments within a time series, on specific classification outcomes. To achieve this, we leverage state-of-the-art diffusion-based generative models to estimate counterfactual outcomes. Our approach compares these causal attributions with closely related associational attributions, both theoretically and empirically. We demonstrate the insights gained by our approach for a diverse set of qualitatively different time series classification tasks. Although causal and associational attributions might often share some similarities, in all cases they differ in important details, underscoring the risks associated with drawing causal conclusions from associational data alone. We believe that the proposed approach is widely applicable also in other domains, particularly where predefined segmentations are available, to shed some light on the limits of associational attributions., Comment: 17 pages, 8 figures. Source code under https://github.com/AI4HealthUOL/CausalConceptTS

This study aims to elucidate the significance of long-range correlations for deep-learning-based sleep staging. It is centered around S4Sleep(TS), a recently proposed model for automated sleep staging. This model utilizes electroencephalography (EEG) as raw time series input and relies on structured state space sequence (S4) models as essential model component. Although the model already surpasses state-of-the-art methods for a moderate number of 15 input epochs, recent literature results suggest potential benefits from incorporating very long correlations spanning hundreds of input epochs. In this submission, we explore the possibility of achieving further enhancements by systematically scaling up the model's input size, anticipating potential improvements in prediction accuracy. In contrast to findings in literature, our results demonstrate that augmenting the input size does not yield a significant enhancement in the performance of S4Sleep(TS). These findings, coupled with the distinctive ability of S4 models to capture long-range dependencies in time series data, cast doubt on the diagnostic relevance of very long-range interactions for sleep staging., Comment: 3 pages, 1 figure, Accepted at Workshop Biosignals, 28.2.-1.3.2024, G\"ottingen, Germany

Feature removal is a central building block for eXplainable AI (XAI), both for occlusion-based explanations (Shapley values) as well as their evaluation (pixel flipping, PF). However, occlusion strategies can vary significantly from simple mean replacement up to inpainting with state-of-the-art diffusion models. This ambiguity limits the usefulness of occlusion-based approaches. For example, PF benchmarks lead to contradicting rankings. This is amplified by competing PF measures: Features are either removed starting with most influential first (MIF) or least influential first (LIF). This study proposes two complementary perspectives to resolve this disagreement problem. Firstly, we address the common criticism of occlusion-based XAI, that artificial samples lead to unreliable model evaluations. We propose to measure the reliability by the R(eference)-Out-of-Model-Scope (OMS) score. The R-OMS score enables a systematic comparison of occlusion strategies and resolves the disagreement problem by grouping consistent PF rankings. Secondly, we show that the insightfulness of MIF and LIF is conversely dependent on the R-OMS score. To leverage this, we combine the MIF and LIF measures into the symmetric relevance gain (SRG) measure. This breaks the inherent connection to the underlying occlusion strategy and leads to consistent rankings. This resolves the disagreement problem, which we verify for a set of 40 different occlusion strategies., Comment: 28 pages, 8 figures

Current deep learning algorithms designed for automatic ECG analysis have exhibited notable accuracy. However, akin to traditional electrocardiography, they tend to be narrowly focused and typically address a singular diagnostic condition. In this exploratory study, we specifically investigate the capability of a single model to predict a diverse range of both cardiac and non-cardiac discharge diagnoses based on a sole ECG collected in the emergency department. We find that 253, 81 cardiac, and 172 non-cardiac, ICD codes can be reliably predicted in the sense of exceeding an AUROC score of 0.8 in a statistically significant manner. This underscores the model's proficiency in handling a wide array of cardiac and non-cardiac diagnostic scenarios which demonstrates potential as a screening tool for diverse medical encounters., Comment: Accepted version EHJDH. 30 pages, 6 figures, code available under https://github.com/AI4HealthUOL/ECG-MIMIC

Cardiovascular diseases remain the leading global cause of mortality. Age is an important covariate whose effect is most easily investigated in a healthy cohort to properly distinguish the former from disease-related changes. Traditionally, most of such insights have been drawn from the analysis of electrocardiogram (ECG) feature changes in individuals as they age. However, these features, while informative, may potentially obscure underlying data relationships. 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. (2) We use explainable AI methods to identify the most discriminative ECG features across age groups.(3) Our analysis with tree-based classifiers reveals age-related declines in inferred breathing rates and identifies notably high SDANN values as indicative of elderly individuals, distinguishing them from younger adults. (4) Furthermore, the deep-learning model underscores the pivotal role of the P-wave in age predictions across all age groups, suggesting potential changes in the distribution of different P-wave types with age. These findings shed new light on age-related ECG changes, offering insights that transcend traditional feature-based approaches., Comment: Accepted version by PLOS ONE. 10 pages, 5 figures, code available under https://github.com/AI4HealthUOL/ECG-aging. Publication under https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0302024

Scoring sleep stages in polysomnography recordings is a time-consuming task plagued by significant inter-rater variability. Therefore, it stands to benefit from the application of machine learning algorithms. While many algorithms have been proposed for this purpose, certain critical architectural decisions have not received systematic exploration. In this study, we meticulously investigate these design choices within the broad category of encoder-predictor architectures. We identify robust architectures applicable to both time series and spectrogram input representations. These architectures incorporate structured state space models as integral components and achieve statistically significant performance improvements compared to state-of-the-art approaches on the extensive Sleep Heart Health Study dataset. We anticipate that the architectural insights gained from this study along with the refined methodology for architecture search demonstrated herein will not only prove valuable for future research in sleep staging but also hold relevance for other time series annotation tasks., Comment: version accepted by Computers in Biology and Medicine, code available at https://github.com/AI4HealthUOL/s4sleep

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. Results: The approach enabled us to identify amino acids in the sequences that the transformers pay particular attention to, and to show that these relevant sequence parts reflect expectations from biology and chemistry, both in the embedding layer and inside of the model, where we identified transformer heads with a statistically significant correspondence of attribution maps with ground truth sequence annotations (e.g. transmembrane regions, active sites) across many proteins. Availability and Implementation: Source code can be accessed at https://github.com/markuswenzel/xai-proteins ., Comment: 26 pages, 12 figures, 5 tables, source code available at https://github.com/markuswenzel/xai-proteins

Deep learning has emerged as the preferred modeling approach for automatic ECG analysis. In this study, we investigate three elements aimed at improving the quantitative accuracy of such systems. These components consistently enhance performance beyond the existing state-of-the-art, which is predominantly based on convolutional models. Firstly, we explore more expressive architectures by exploiting structured state space models (SSMs). These models have shown promise in capturing long-term dependencies in time series data. By incorporating SSMs into our approach, we not only achieve better performance, but also gain insights into long-standing questions in the field. Specifically, for standard diagnostic tasks, we find no advantage in using higher sampling rates such as 500Hz compared to 100Hz. Similarly, extending the input size of the model beyond 3 seconds does not lead to significant improvements. Secondly, we demonstrate that self-supervised learning using contrastive predictive coding can further improve the performance of SSMs. By leveraging self-supervision, we enable the model to learn more robust and representative features, leading to improved analysis accuracy. Lastly, we depart from synthetic benchmarking scenarios and incorporate basic demographic metadata alongside the ECG signal as input. This inclusion of patient metadata departs from the conventional practice of relying solely on the signal itself. Remarkably, this addition consistently yields positive effects on predictive performance. We firmly believe that all three components should be considered when developing next-generation ECG analysis algorithms., Comment: extended version of arXiv:2211.07579

Deep neural networks have become increasingly popular for analyzing ECG data because of their ability to accurately identify cardiac conditions and hidden clinical factors. However, the lack of transparency due to the black box nature of these models is a common concern. To address this issue, explainable AI (XAI) methods can be employed. In this study, we present a comprehensive analysis of post-hoc XAI methods, investigating the local (attributions per sample) and global (based on domain expert concepts) perspectives. We have established a set of sanity checks to identify sensible attribution methods, and we provide quantitative evidence in accordance with expert rules. This dataset-wide analysis goes beyond anecdotal evidence by aggregating data across patient subgroups. Furthermore, we demonstrate how these XAI techniques can be utilized for knowledge discovery, such as identifying subtypes of myocardial infarction. We believe that these proposed methods can serve as building blocks for a complementary assessment of the internal validity during a certification process, as well as for knowledge discovery in the field of ECG analysis.

Feature importance methods promise to provide a ranking of features according to importance for a given classification task. A wide range of methods exist but their rankings often disagree and they are inherently difficult to evaluate due to a lack of ground truth beyond synthetic datasets. In this work, we put feature importance methods to the test on real-world data in the domain of cardiology, where we try to distinguish three specific pathologies from healthy subjects based on ECG features comparing to features used in cardiologists' decision rules as ground truth. Some methods generally performed well and others performed poorly, while some methods did well on some but not all of the problems considered.

The completeness axiom renders the explanation of a post-hoc XAI method only locally faithful to the model, i.e. for a single decision. For the trustworthy application of XAI, in particular for high-stake decisions, a more global model understanding is required. Recently, concept-based methods have been proposed, which are however not guaranteed to be bound to the actual model reasoning. To circumvent this problem, we propose Multi-dimensional Concept Discovery (MCD) as an extension of previous approaches that fulfills a completeness relation on the level of concepts. Our method starts from general linear subspaces as concepts and does neither require reinforcing concept interpretability nor re-training of model parts. We propose sparse subspace clustering to discover improved concepts and fully leverage the potential of multi-dimensional subspaces. MCD offers two complementary analysis tools for concepts in input space: (1) concept activation maps, that show where a concept is expressed within a sample, allowing for concept characterization through prototypical samples, and (2) concept relevance heatmaps, that decompose the model decision into concept contributions. Both tools together enable a detailed understanding of the model reasoning, which is guaranteed to relate to the model via a completeness relation. This paves the way towards more trustworthy concept-based XAI. We empirically demonstrate the superiority of MCD against more constrained concept definitions., Comment: v2: Version published by Transactions on Machine Learning Research in 2023 (TMLR ISSN 2835-8856) https://openreview.net/forum?id=KxBQPz7HKh. 25 pages, 11 figures. This work builds on an earlier manuscript (arXiv:2203.06043) and crucially extends it. Code is available at https://github.com/jvielhaben/MCD-XAI

Synthetic data generation is a promising solution to address privacy issues with the distribution of sensitive health data. Recently, diffusion models have set new standards for generative models for different data modalities. Also very recently, structured state space models emerged as a powerful modeling paradigm to capture long-term dependencies in time series. We put forward SSSD-ECG, as the combination of these two technologies, for the generation of synthetic 12-lead electrocardiograms conditioned on more than 70 ECG statements. Due to a lack of reliable baselines, we also propose conditional variants of two state-of-the-art unconditional generative models. We thoroughly evaluate the quality of the generated samples, by evaluating pretrained classifiers on the generated data and by evaluating the performance of a classifier trained only on synthetic data, where SSSD-ECG clearly outperforms its GAN-based competitors. We demonstrate the soundness of our approach through further experiments, including conditional class interpolation and a clinical Turing test demonstrating the high quality of the SSSD-ECG samples across a wide range of conditions., Comment: 12 pages, 9 figures. Accepted version by Computers in Biology and Medicine in 2023 under https://doi.org/10.1016/j.compbiomed.2023.107115. Source code under https://github.com/AI4HealthUOL/SSSD-ECG

The field of deep-learning-based ECG analysis has been largely dominated by convolutional architectures. This work explores the prospects of applying the recently introduced structured state space models (SSMs) as a particularly promising approach due to its ability to capture long-term dependencies in time series. We demonstrate that this approach leads to significant improvements over the current state-of-the-art for ECG classification, which we trace back to individual pathologies. Furthermore, the model's ability to capture long-term dependencies allows to shed light on long-standing questions in the literature such as the optimal sampling rate or window size to train classification models. Interestingly, we find no evidence for using data sampled at 500Hz as opposed to 100Hz and no advantages from extending the model's input size beyond 3s. Based on this very promising first assessment, SSMs could develop into a new modeling paradigm for ECG analysis., Comment: Extended Abstract presented at Machine Learning for Health (ML4H) symposium 2022, November 28th, 2022, New Orleans, United States & Virtual, http://www.ml4h.cc, 6+5 pages

The imputation of missing values represents a significant obstacle for many real-world data analysis pipelines. Here, we focus on time series data and put forward SSSD, an imputation model that relies on two emerging technologies, (conditional) diffusion models as state-of-the-art generative models and structured state space models as internal model architecture, which are particularly suited to capture long-term dependencies in time series data. We demonstrate that SSSD matches or even exceeds state-of-the-art probabilistic imputation and forecasting performance on a broad range of data sets and different missingness scenarios, including the challenging blackout-missing scenarios, where prior approaches failed to provide meaningful results., Comment: V3: Updated results for the solar dataset. 36 pages, 13 figures. Version published by Transactions on Machine Learning Research in 2022 (TMLR ISSN 2835-8856) https://openreview.net/forum?id=hHiIbk7ApW. Source code under https://github.com/AI4HealthUOL/SSSD

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. In order to fill this gap, we developed a new methodology to extensively evaluate a wide range of classification models, including recent vision transformers, and convolutional neural networks such as: ConvNeXt, ResNet (BiT), Inception, ViT and Swin transformer, with and without supervised or self-supervised pretraining. We thoroughly tested the models on five widely used histopathology datasets containing whole slide images of breast, gastric, and colorectal cancer and developed a novel approach using an image-to-image translation model to assess the robustness of a cancer classification model against stain variations. Further, we extended existing interpretability methods to previously unstudied models and systematically reveal insights of the models' classifications strategies that can be transferred to future model architectures., Comment: 14 pages, 6 figures, v2: version accepted by Medical Image Analysis, code available under https://github.com/hhi-aml/histobenchmark

Concepts are key building blocks of higher level human understanding. Explainable AI (XAI) methods have shown tremendous progress in recent years, however, local attribution methods do not allow to identify coherent model behavior across samples and therefore miss this essential component. In this work, we study concept-based explanations and put forward a new definition of concepts as low-dimensional subspaces of hidden feature layers. We novelly apply sparse subspace clustering to discover these concept subspaces. Moving forward, we derive insights from concept subspaces in terms of localized input (concept) maps, show how to quantify concept relevances and lastly, evaluate similarities and transferability between concepts. We empirically demonstrate the soundness of the proposed Sparse Subspace Clustering for Concept Discovery (SSCCD) method for a variety of different image classification tasks. This approach allows for deeper insights into the actual model behavior that would remain hidden from conventional input-level heatmaps., Comment: 24 pages, 24 figures, code will be made publicly available

There is an increasing number of medical use-cases where classification algorithms based on deep neural networks reach performance levels that are competitive with human medical experts. To alleviate the challenges of small dataset sizes, these systems often rely on pretraining. In this work, we aim to assess the broader implications of these approaches. For diabetic retinopathy grading as exemplary use case, we compare the impact of different training procedures including recently established self-supervised pretraining methods based on contrastive learning. To this end, we investigate different aspects such as quantitative performance, statistics of the learned feature representations, interpretability and robustness to image distortions. Our results indicate that models initialized from ImageNet pretraining report a significant increase in performance, generalization and robustness to image distortions. In particular, self-supervised models show further benefits to supervised models. Self-supervised models with initialization from ImageNet pretraining not only report higher performance, they also reduce overfitting to large lesions along with improvements in taking into account minute lesions indicative of the progression of the disease. Understanding the effects of pretraining in a broader sense that goes beyond simple performance comparisons is of crucial importance for the broader medical imaging community beyond the use-case considered in this work.

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. In this work, we adapt and extend USMPep, a recently proposed, conceptually simple prediction algorithm based on recurrent neural networks. Most notably, we combine regressors (binding affinity data) and classifiers (mass spectrometry data) from qualitatively different data sources to obtain a more comprehensive prediction tool. We evaluate the performance on a recently released SARS-CoV-2 dataset with binding stability measurements. USMPep not only sets new benchmarks on selected single alleles, but consistently turns out to be among the best-performing methods or, for some metrics, to be even the overall best-performing method for this task., Comment: Accepted at ICLR 2021 Workshop: Machine Learning for Preventing and Combating Pandemics; code available at https://github.com/nstrodt/USMPep

Clinical 12-lead electrocardiography (ECG) is one of the most widely encountered kinds of biosignals. Despite the increased availability of public ECG datasets, label scarcity remains a central challenge in the field. Self-supervised learning represents a promising way to alleviate this issue. In this work, we put forward the first comprehensive assessment of self-supervised representation learning from clinical 12-lead ECG data. To this end, we adapt state-of-the-art self-supervised methods based on instance discrimination and latent forecasting to the ECG domain. 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. In a second step, we analyze the impact of self-supervised pretraining on finetuned ECG classifiers as compared to purely supervised performance. For the best-performing method, an adaptation of contrastive predictive coding, we find a linear evaluation performance only 0.5% below supervised performance. For the finetuned models, we find improvements in downstream performance of roughly 1% compared to supervised performance, label efficiency, as well as robustness against physiological noise. This work clearly establishes the feasibility of extracting discriminative representations from ECG data via self-supervised learning and the numerous advantages when finetuning such representations on downstream tasks as compared to purely supervised training. As first comprehensive assessment of its kind in the ECG domain carried out exclusively on publicly available datasets, we hope to establish a first step towards reproducible progress in the rapidly evolving field of representation learning for biosignals., Comment: 15 pages, 12 figures, matches published version, code available under https://github.com/hhi-aml/ecg-selfsupervised

We discuss the emergence of a low-energy effective theory with quarks, mesons, diquarks and baryons at vanishing and finite baryon density from first principle QCD. The present work also includes an overview on diquarks at vanishing and finite density, and elucidates the physics of transitional changes from baryonic matter to quark matter including diquarks. This set-up is discussed within the functional renormalisation group approach with dynamical hadronisation. In this framework it is detailed how mesons, diquarks, and baryons emerge dynamically from the renormalisation flow of the QCD effective action. Moreover, the fundamental degrees of freedom of QCD, quarks and gluons, decouple from the dynamics of QCD below the respective mass gaps. The resulting global picture unifies the different low energy effective theories used for low and high densities within QCD, and allows for a determination of the respective low energy constants directly from QCD., Comment: 30 pages, 17 figures

PredDiff is a model-agnostic, local attribution method that is firmly rooted in probability theory. Its simple intuition is to measure prediction changes while marginalizing features. In this work, we clarify properties of PredDiff and its close connection to Shapley values. We stress important differences between classification and regression, which require a specific treatment within both formalisms. We extend PredDiff by introducing a new, well-founded measure for interaction effects between arbitrary feature subsets. The study of interaction effects represents an inevitable step towards a comprehensive understanding of black-box models and is particularly important for science applications. Equipped with our novel interaction measure, PredDiff is a promising model-agnostic approach for obtaining reliable, numerically inexpensive and theoretically sound attributions., Comment: 35 pages, 20 Figures, accepted journal version, code available at https://github.com/AI4HealthUOL/preddiff-interactions

Generative neural samplers offer a complementary approach to Monte Carlo methods for problems in statistical physics and quantum field theory. This work tests the ability of generative neural samplers to estimate observables for real-world low-dimensional spin systems. It maps out how autoregressive models can sample configurations of a quantum Heisenberg chain via a classical approximation based on the Suzuki-Trotter transformation. We present results for energy, specific heat and susceptibility for the isotropic XXX and the anisotropic XY chain that are in good agreement with Monte Carlo results within the same approximation scheme., Comment: 10 figures

Electrical impedance tomography (EIT) is a noninvasive imaging modality that allows a continuous assessment of changes in regional bioimpedance of different organs. One of its most common biomedical applications is monitoring regional ventilation distribution in critically ill patients treated in intensive care units. In this work, we put forward a proof-of-principle study that demonstrates how one can reconstruct synchronously measured respiratory or circulatory parameters from the EIT image sequence using a deep learning model trained in an end-to-end fashion. We demonstrate that one can accurately infer absolute volume, absolute flow, normalized airway pressure and within certain limitations even the normalized arterial blood pressure from the EIT signal alone, in a way that generalizes to unseen patients without prior calibration. As an outlook with direct clinical relevance, we furthermore demonstrate the feasibility of reconstructing the absolute transpulmonary pressure from a combination of EIT and absolute airway pressure, as a way to potentially replace the invasive measurement of esophageal pressure. With these results, we hope to stimulate further studies building on the framework put forward in this work., Comment: 6 pages, 3 figures

Electrocardiography is a very common, non-invasive diagnostic procedure and its interpretation is increasingly supported by automatic interpretation algorithms. The progress in the field of automatic ECG interpretation has up to now been hampered by a lack of appropriate datasets for training as well as a lack of well-defined evaluation procedures to ensure comparability of different algorithms. To alleviate these issues, we put forward first benchmarking results for the recently published, freely accessible PTB-XL dataset, covering a variety of tasks from different ECG statement prediction tasks over age and gender prediction to signal quality assessment. We find that convolutional neural networks, in particular resnet- and inception-based architectures, show the strongest performance across all tasks outperforming feature-based algorithms by a large margin. These results are complemented by deeper insights into the classification algorithm in terms of hidden stratification, model uncertainty and an exploratory interpretability analysis. We also put forward benchmarking results for the ICBEB2018 challenge ECG dataset and discuss prospects of transfer learning using classifiers pretrained on PTB-XL. With this resource, we aim to establish the PTB-XL dataset as a resource for structured benchmarking of ECG analysis algorithms and encourage other researchers in the field to join these efforts., Comment: 12 pages, 8 figures

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. Available algorithms, in particular those based on deep learning, often demonstrate remarkable performance in the search for previously unidentified features, but tend to lack transparency if applied naively. To address these shortcomings, we propose representation learning in combination with interpretability methods as a framework for the identification of observables. More specifically, we investigate action parameter regression as a pretext task while using layer-wise relevance propagation (LRP) to identify the most important observables depending on the location in the phase diagram. The approach is put to work in the context of a scalar Yukawa model in (2+1)d. First, we investigate a multilayer perceptron to determine an importance hierarchy of several predefined, standard observables. The method is then applied directly to the raw field configurations using a convolutional network, demonstrating the ability to reconstruct all order parameters from the learned filter weights. Based on our results, we argue that due to its broad applicability, attribution methods such as LRP could prove a useful and versatile tool in our search for new physical insights. In the case of the Yukawa model, it facilitates the construction of an observable that characterises the symmetric phase., Comment: 13 pages, 11 figures

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. In this framework, we present asymptotically unbiased estimators for generic observables, including those that explicitly depend on the partition function such as free energy or entropy, and derive corresponding variance estimators. We demonstrate their practical applicability by numerical experiments for the 2d Ising model which highlight the superiority over existing methods. Our approach greatly enhances the applicability of generative neural samplers to real-world physical systems., Comment: 5 figures

We study the recently introduced stability training as a general-purpose method to increase the robustness of deep neural networks against input perturbations. In particular, we explore its use as an alternative to data augmentation and validate its performance against a number of distortion types and transformations including adversarial examples. In our image classification experiments using ImageNet data stability training performs on a par or even outperforms data augmentation for specific transformations, while consistently offering improved robustness against a broader range of distortion strengths and types unseen during training, a considerably smaller hyperparameter dependence and less potentially negative side effects compared to data augmentation., Comment: 18 pages, 25 figures; Camera-ready version

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. We show that the inherent sampling error of their method can be corrected by using neural network-based MCMC or importance sampling which leads to asymptotically unbiased estimators for physical quantities. This modification is possible due to a singular property of VANs, namely that they provide the exact sample probability. With these modifications, we believe that their method could have a substantially greater impact on various important fields of physics, including strongly-interacting field theories and statistical physics., Comment: 6 pages, 4 figures

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). To this end, we propose machine learning methods to predict the outcome of the decoding process ahead of the end of the transmission. We discuss different input features and classification algorithms ranging from traditional methods to newly developed supervised autoencoders. These methods are evaluated based on their prospects of complying with the URLLC requirements of effective block error rates below $10^{-5}$ at small latency overheads. We provide realistic performance estimates in a system model incorporating scheduling effects to demonstrate the feasibility of E-HARQ across different signal-to-noise ratios, subcode lengths, channel conditions and system loads, and show the benefit over regular HARQ and existing E-HARQ schemes without machine learning., Comment: 14 pages, 15 figures; accepted version

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. Approach: We train an ensemble of fully convolutional neural networks on the PTB ECG dataset and apply state-of-the-art attribution methods. Main results: Our classifier reaches 93.3% sensitivity and 89.7% specificity evaluated using 10-fold cross-validation with sampling based on patients. The presented method outperforms state-of-the-art approaches and reaches the performance level of human cardiologists for detection of myocardial infarction. We are able to discriminate channel-specific regions that contribute most significantly to the neural network's decision. Interestingly, the network's decision is influenced by signs also recognized by human cardiologists as indicative of myocardial infarction. Significance: Our results demonstrate the high prospects of algorithmic ECG analysis for future clinical applications considering both its quantitative performance as well as the possibility of assessing decision criteria on a per-example basis, which enhances the comprehensibility of the approach., Comment: 11 pages, 4 figures

We compute correlation functions of three-dimensional Landau-gauge Yang-Mills theory with the Functional Renormalisation Group. Starting from the classical action as only input, we calculate the non-perturbative ghost and gluon propagators as well as the momentum-dependent ghost-gluon, three-gluon, and four-gluon vertices in a comprehensive truncation scheme. Compared to the physical case of four spacetime dimensions, we need more sophisticated truncations due to significant contributions from non-classical tensor structures. In particular, we apply a special technique to compute the tadpole diagrams of the propagator equations, which captures also all perturbative two-loop effects, and compare our correlators with lattice and Dyson-Schwinger results., Comment: 10 pages, 11 figures

We directly calculate spectral functions in the O(N)-model at finite temperature within the framework of the Functional Renormalization group. Special emphasis is put on a fully numerical framework involving four-dimensional regulators preserving Euclidean O(4) and Minkowski Lorentz invariance, an important prerequisite for future applications. Pion and sigma meson spectral functions are calculated for a wide range of temperatures across the phase transition illustrating the applicability of the general framework for finite temperature applications. In addition, various aspects concerning the interplay between the Euclidean and real time two-point function are discussed., Comment: 11 pages, 20 figures

We present non-perturbative correlation functions in Landau-gauge Yang-Mills theory at finite temperature. The results are obtained from the functional renormalisation group within a self-consistent approximation scheme. In particular, we compute the magnetic and electric components of the gluon propagator, and the three- and four-gluon vertices. We also show the ghost propagator and the ghost-gluon vertex at finite temperature. Our results for the propagators are confronted with lattice simulations and our Debye mass is compared to hard thermal loop perturbation theory., Comment: 13 pages + appendix

We present non-perturbative first-principle results for quark-, gluon- and meson $1$PI correlation functions of two-flavour Landau-gauge QCD in the vacuum. These correlation functions carry the full information about the theory. They are obtained by solving their Functional Renormalisation Group equations in a systematic vertex expansion, aiming at apparent convergence. This work represents a crucial prerequisite for quantitative first-principle studies of the QCD phase diagram and the hadron spectrum within this framework. In particular, we have computed the gluon, ghost, quark and scalar-pseudoscalar meson propagators, as well as gluon, ghost-gluon, quark-gluon, quark, quark-meson, and meson interactions. Our results stress the crucial importance of the quantitatively correct running of different vertices in the semi-perturbative regime for describing the phenomena and scales of confinement and spontaneous chiral symmetry breaking without phenomenological input.