Damian Sabas Roqueiro

Last Name

Roqueiro

First Name

Damian Sabas

ORCID

0000-0002-9195-5915

Organisational unit

01630 - Lehre HEST

Show all metadata

Search Results

Publications 1 - 10 of 19

Machine Learning Successfully Detects Patients with COVID-19 Prior to PCR Results and Predicts Their Survival Based on Standard Laboratory Parameters in an Observational Study
Styrzynski, Filip; Zhakparov, Damir; Schmid, Marco; et al. (2023)
Infectious Diseases and Therapy
Introduction In the current COVID-19 pandemic, clinicians require a manageable set of decisive parameters that can be used to (i) rapidly identify SARS-CoV-2 positive patients, (ii) identify patients with a high risk of a fatal outcome on hospital admission, and (iii) recognize longitudinal warning signs of a possible fatal outcome. Methods This comparative study was performed in 515 patients in the Maria Sklodowska-Curie Specialty Voivodeship Hospital in Zgierz, Poland. The study groups comprised 314 patients with COVID-like symptoms who tested negative and 201 patients who tested positive for SARS-CoV-2 infection; of the latter, 72 patients with COVID-19 died and 129 were released from hospital. Data on which we trained several machine learning (ML) models included clinical findings on admission and during hospitalization, symptoms, epidemiological risk, and reported comorbidities and medications. Results We identified a set of eight on-admission parameters: white blood cells, antibody-synthesizing lymphocytes, ratios of basophils/lymphocytes, platelets/neutrophils, and monocytes/lymphocytes, procalcitonin, creatinine, and C-reactive protein. The medical decision tree built using these parameters differentiated between SARS-CoV-2 positive and negative patients with up to 90-100% accuracy. Patients with COVID-19 who on hospital admission were older, had higher procalcitonin, C-reactive protein, and troponin I levels together with lower hemoglobin and platelets/neutrophils ratio were found to be at highest risk of death from COVID-19. Furthermore, we identified longitudinal patterns in C-reactive protein, white blood cells, and D dimer that predicted the disease outcome. Conclusions Our study provides sets of easily obtainable parameters that allow one to assess the status of a patient with SARS-CoV-2 infection, and the risk of a fatal disease outcome on hospital admission and during the course of the disease.
DeePhys: A machine learning–assisted platform for electrophysiological phenotyping of human neuronal networks
Hornauer, Philipp; Prack, Gustavo; Anastasi, Nadia; et al. (2024)
Stem Cell Reports
Reproducible functional assays to study in vitro neuronal networks represent an important cornerstone in the quest to develop physiologically relevant cellular models of human diseases. Here, we introduce DeePhys, a MATLAB-based analysis tool for data-driven functional phenotyping of in vitro neuronal cultures recorded by high-density microelectrode arrays. DeePhys is a modular workflow that offers a range of techniques to extract features from spike-sorted data, allowing for the examination of functional phenotypes both at the individual cell and network levels, as well as across development. In addition, DeePhys incorporates the capability to integrate novel features and to use machine-learning-assisted approaches, which facilitates a comprehensive evaluation of pharmacological interventions. To illustrate its practical application, we apply DeePhys to human induced pluripotent stem cell–derived dopaminergic neurons obtained from both patients and healthy individuals and showcase how DeePhys enables phenotypic screenings.
DeePhys, a machine learning-driven platform for electrophysiological phenotype screening of human stem-cell derived neuronal networks
Hornauer, Philipp; Prack, Gustavo; Anastasi, Nadia; et al. (2022)
MEA Meeting 2022 Abstract Book
Comparison of connectivity inference algorithms for classification of neuronal cultures using graph kernels
Kim, Taehoon; Hornauer, Philipp; Donner, Christian; et al. (2020)
Neurons derived from human induced pluripotent stem cells (iPSCs), provide new means to study aspects of severe neurological diseases in vitro. Network features extracted from electrophysiological recordings of iPSC-derived neurons could be useful to better understand and study disease phenotypes. However, up to this date, there is no fully-validated method to infer connectivity between neurons when using spike trains as input. In this study, we compare two types of human iPSC- derived dopaminergic neurons: wild type cells and cells with a genetic mutation associated with Parkinson’s disease. Moreover, we use graph kernels to train a classifier on the inferred functional networks and probe which connectivity inference parameters generate networks with more discriminative features.
Predicting in vitro single-neuron firing rates upon pharmacological perturbation using Graph Neural Networks
Kim, Taehoon; Chen, Dexiong; Hornauer, Philipp; et al. (2023)
Frontiers in Neuroinformatics
Modern Graph Neural Networks (GNNs) provide opportunities to study the determinants underlying the complex activity patterns of biological neuronal networks. In this study, we applied GNNs to a large-scale electrophysiological dataset of rodent primary neuronal networks obtained by means of high-density microelectrode arrays (HD-MEAs). HD-MEAs allow for long-term recording of extracellular spiking activity of individual neurons and networks and enable the extraction of physiologically relevant features at the singleneuron and population level. We employed established GNNs to generate a combined representation of single-neuron and connectivity features obtained from HD-MEA data, with the ultimate goal of predicting changes in single-neuron firing rate induced by a pharmacological perturbation. The aim of the main prediction task was to assess whether single-neuron and functional connectivity features, inferred under baseline conditions, were informative for predicting changes in neuronal activity in response to a perturbation with Bicuculline, a GABAA receptor antagonist. Our results suggest that the joint representation of node features and functional connectivity, extracted from a baseline recording, was informative for predicting firing rate changes of individual neurons after the perturbation. Specifically, our implementation of a GNN model with inductive learning capability (GraphSAGE) outperformed other prediction models that relied only on single-neuron features. We tested the generalizability of the results on two additional datasets of HD-MEA recordings–a second dataset with cultures perturbed with Bicuculline and a dataset perturbed with the GABAA receptor antagonist Gabazine. GraphSAGE models showed improved prediction accuracy over other prediction models. Our results demonstrate the added value of taking into account the functional connectivity between neurons and the potential of GNNs to study complex interactions between neurons.
Association mapping in biomedical time series via statistically significant shapelet mining
Bock, Christian; Gumbsch, Thomas; Moor, Michael; et al. (2018)
Bioinformatics
Motivation Most modern intensive care units record the physiological and vital signs of patients. These data can be used to extract signatures, commonly known as biomarkers, that help physicians understand the biological complexity of many syndromes. However, most biological biomarkers suffer from either poor predictive performance or weak explanatory power. Recent developments in time series classification focus on discovering shapelets, i.e. subsequences that are most predictive in terms of class membership. Shapelets have the advantage of combining a high predictive performance with an interpretable component—their shape. Currently, most shapelet discovery methods do not rely on statistical tests to verify the significance of individual shapelets. Therefore, identifying associations between the shapelets of physiological biomarkers and patients that exhibit certain phenotypes of interest enables the discovery and subsequent ranking of physiological signatures that are interpretable, statistically validated and accurate predictors of clinical endpoints. Results We present a novel and scalable method for scanning time series and identifying discriminative patterns that are statistically significant. The significance of a shapelet is evaluated while considering the problem of multiple hypothesis testing and mitigating it by efficiently pruning untestable shapelet candidates with Tarone’s method. We demonstrate the utility of our method by discovering patterns in three of a patient’s vital signs: heart rate, respiratory rate and systolic blood pressure that are indicators of the severity of a future sepsis event, i.e. an inflammatory response to an infective agent that can lead to organ failure and death, if not treated in time. Availability and implementation We make our method and the scripts that are required to reproduce the experiments publicly available at https://github.com/BorgwardtLab/S3M. Supplementary information Supplementary data are available at Bioinformatics online.
DeePhys, a machine learning-driven platform for electrophysiological phenotype screening of human stem-cell derived neuronal networks
Hornauer, Philipp; Prack, Gustavo; Anastasi, Nadia; et al. (2022)
Machine learning for deep electrophysiological phenotyping of human dopaminergic neurons
Schröter, Manuel; Roqueiro, Damian Sabas; Prack, Gustavo; et al. (2019)
Poster Abstract Book. ISSCR 2019 Annual Meeting
Methods and Tools in Genome-wide Association Studies
Gumpinger, Anja C.; Roqueiro, Damian Sabas; Grimm, Dominik G.; et al. (2018)
Methods in Molecular Biology ~ Computational Cell Biology
Electrophysiological characterization of α-synuclein function in human iPSC-derived dopaminergic neurons using high-density microelectrode arrays
Hornauer, Philipp; Prack, Gustavo; Fiscella, Michele; et al. (2020)
ISSCR 2020 Poster Abstract Guide

Publications 1 - 10 of 19

Damian Sabas Roqueiro

Last Name

First Name

ORCID

Organisational unit

Filters

Search Results