RESUMO
Deep neural networks must address the dual challenge of delivering high-accuracy predictions and providing user-friendly explanations. While deep models are widely used in the field of time series modeling, deciphering the core principles that govern the models' outputs remains a significant challenge. This is crucial for fostering the development of trusted models and facilitating domain expert validation, thereby empowering users and domain experts to utilize them confidently in high-risk decision-making contexts (e.g., decision-support systems in healthcare). In this work, we put forward a deep prototype learning model that supports interpretable and manipulable modeling and classification of medical time series (i.e., ECG signal). Specifically, we first optimize the representation of single heartbeat data by employing a bidirectional long short-term memory and attention mechanism, and then construct prototypes during the training phase. The final classification outcomes (i.e., normal sinus rhythm, atrial fibrillation, and other rhythm) are determined by comparing the input with the obtained prototypes. Moreover, the proposed model presents a human-machine collaboration mechanism, allowing domain experts to refine the prototypes by integrating their expertise to further enhance the model's performance (contrary to the human-in-the-loop paradigm, where humans primarily act as supervisors or correctors, intervening when required, our approach focuses on a human-machine collaboration, wherein both parties engage as partners, enabling more fluid and integrated interactions). The experimental outcomes presented herein delineate that, within the realm of binary classification tasks-specifically distinguishing between normal sinus rhythm and atrial fibrillation-our proposed model, albeit registering marginally lower performance in comparison to certain established baseline models such as Convolutional Neural Networks (CNNs) and bidirectional long short-term memory with attention mechanisms (Bi-LSTMAttns), evidently surpasses other contemporary state-of-the-art prototype baseline models. Moreover, it demonstrates significantly enhanced performance relative to these prototype baseline models in the context of triple classification tasks, which encompass normal sinus rhythm, atrial fibrillation, and other rhythm classifications. The proposed model manifests a commendable prediction accuracy of 0.8414, coupled with macro precision, recall, and F1-score metrics of 0.8449, 0.8224, and 0.8235, respectively, achieving both high classification accuracy as well as good interpretability.
Assuntos
Eletrocardiografia , Redes Neurais de Computação , Humanos , Eletrocardiografia/métodos , Fibrilação Atrial/fisiopatologia , Fibrilação Atrial/diagnóstico , Aprendizado Profundo , Frequência Cardíaca/fisiologia , Algoritmos , Processamento de Sinais Assistido por ComputadorRESUMO
The rapid dissemination of misinformation in social media during the COVID-19 pandemic triggers panic and threatens the pandemic preparedness and control. Correction is a crucial countermeasure to debunk misperceptions. However, the effective mechanism of correction on social media is not fully verified. Previous works focus on psychological theories and experimental studies, while the applicability of conclusions to the actual social media is unclear. This study explores determinants governing the effectiveness of misinformation corrections on social media with a combination of a data-driven approach and related theories on psychology and communication. Specifically, referring to the Backfire Effect, Source Credibility, and Audience's role in dissemination theories, we propose five hypotheses containing seven potential factors (regarding correction content and publishers' influence), e.g., the proportion of original misinformation and warnings of misinformation. Then, we obtain 1487 significant COVID-19 related corrections on Microblog between January 1st, 2020 and April 30th, 2020, and conduct annotations, which characterize each piece of correction based on the aforementioned factors. We demonstrate several promising conclusions through a comprehensive analysis of the dataset. For example, mentioning excessive original misinformation in corrections would not undermine people's believability within a short period after reading; warnings of misinformation in a demanding tone make correction worse; determinants of correction effectiveness vary among different topics of misinformation. Finally, we build a regression model to predict correction effectiveness. These results provide practical suggestions on misinformation correction on social media, and a tool to guide practitioners to revise corrections before publishing, leading to ideal efficacies.