The latest trends in clinical care and telemedicine suggest a demand for a reliable automated electrocardiogram (ECG) signal classification methods. In this paper, we present customized deep learning model for ECG classification as a part of the Physionet/CinC Challenge 2020. The method is based on modified ResNet type convolutional neural network and is capable to automatically recognize 24 cardiac abnormalities using 12-lead ECG. We have adopted several preprocessing and learning techniques including custom tailored loss function, dedicated classification layer and Bayesian threshold optimization which have major positive impact on the model performance. At the official phase of the Challenge, our team-BUTTeam-reached a challenge validation score of 0.696, and the full test score of 0.202, placing us 21 out of 40 in the official ranking. This implies that our method performed well on data from the same source (reached first place with validation score), however, it has very poor generalization to data from different sources.
This paper aims to present a methodology for sepsis prediction from clinical time-series data. Sepsis is one of the most threatening states which could occur while treating a patient at the intensive care unit. Therefore its prediction could significantly improve the quality of the patient treatment.In this work, we address the problem of sepsis prediction with Long Short-Term Memory (LSTM) network with specialized deep architecture with residual connections. The output of the network is sepsis prediction score at each point in time.Feature normalization into the fixed range of values is applied including replacing missing values with numerical representation from outside the normalized range. Therefore, the LSTM network is able to include missing values in the learning process. Also, the rarity of sepsis occurrence in the provided dataset is a challenging problem. This problem is addressed by the application of dice loss providing automatically weighted classes by the occurrence of the feature.The proposed method leads to 0.281 normalized utility score on the full test set as the best official Phys-ioNet/Computing in Cardiology (CinC) Challenge 2019 entry of ECGuru10 team.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.