Posted Content•
Virufy: A Multi-Branch Deep Learning Network for Automated Detection of COVID-19
TL;DR: In this paper, a multi-branch deep learning network was proposed to detect COVID-19 infection status using audio samples recorded in clinical settings, where most of the data has not been manually processed and cleaned.
Abstract: Fast and affordable solutions for COVID-19 testing are necessary to contain the spread of the global pandemic and help relieve the burden on medical facilities. Currently, limited testing locations and expensive equipment pose difficulties for individuals trying to be tested, especially in low-resource settings. Researchers have successfully presented models for detecting COVID-19 infection status using audio samples recorded in clinical settings [5, 15], suggesting that audio-based Artificial Intelligence models can be used to identify COVID-19. Such models have the potential to be deployed on smartphones for fast, widespread, and low-resource testing. However, while previous studies have trained models on cleaned audio samples collected mainly from clinical settings, audio samples collected from average smartphones may yield suboptimal quality data that is different from the clean data that models were trained on. This discrepancy may add a bias that affects COVID-19 status predictions. To tackle this issue, we propose a multi-branch deep learning network that is trained and tested on crowdsourced data where most of the data has not been manually processed and cleaned. Furthermore, the model achieves state-of-art results for the COUGHVID dataset [16]. After breaking down results for each category, we have shown an AUC of 0.99 for audio samples with COVID-19 positive labels.
Citations
More filters
TL;DR: In this paper, the authors used the Coswara dataset where each user has recorded 9 different types of sounds as cough, breathing, and speech labeled with COVID-19 status and showed that using simple binary classifiers can achieve an AUC of 96.4% and an accuracy of 96% by averaging the predictions of multiple models trained separately on different sound types.
Abstract: Since the outbreak of COVID-19, many efforts have been made to utilize the respiratory sounds and coughs collected by smartphones for training Machine Learning models to classify and distinguish COVID-19 sounds from healthy ones. Embedding those models into mobile applications or Internet of things devices can make effective COVID-19 pre-screening tools afforded by anyone anywhere. Most of the previous researchers trained their classifiers with respiratory sounds such as breathing or coughs, and they achieved promising results. We claim that using special voice patterns besides other respiratory sounds can achieve better performance. In this study, we used the Coswara dataset where each user has recorded 9 different types of sounds as cough, breathing, and speech labeled with COVID-19 status. A combination of models trained on different sounds can diagnose COVID-19 more accurately than a single model trained on cough or breathing only. Our results show that using simple binary classifiers can achieve an AUC of 96.4% and an accuracy of 96% by averaging the predictions of multiple models trained and evaluated separately on different sound types. Finally, this study aims to draw attention to the importance of the human voice alongside other respiratory sounds for the sound-based COVID-19 diagnosis.
23 citations
TL;DR: In this paper, features extracted from an ensemble of 5 different CNNs (MobilenetV2, Shufflenet, Xception, Darknet53 and EfficientnetB0) in combination with kernel support vector machine is used for the diagnosis of COVID-19 from CT scans.
Abstract: Corona Virus Disease-2019 (COVID-19) is a global pandemic which is spreading briskly across the globe. The gold standard for the diagnosis of COVID-19 is viral nucleic acid detection with real-time polymerase chain reaction (RT-PCR). However, the sensitivity of RT-PCR in the diagnosis of early-stage COVID-19 is less. Recent research works have shown that computed tomography (CT) scans of the chest are effective for the early diagnosis of COVID-19. Convolutional neural networks (CNNs) are proven successful for diagnosing various lung diseases from CT scans. CNNs are composed of multiple layers which represent a hierarchy of features at each level. CNNs require a big number of labeled instances for training from scratch. In medical imaging tasks like the detection of COVID-19 where there is a difficulty in acquiring a large number of labeled CT scans, pre-trained CNNs trained on a huge number of natural images can be employed for extracting features. Feature representation of each CNN varies and an ensemble of features generated from various pre-trained CNNs can increase the diagnosis capability significantly. In this paper, features extracted from an ensemble of 5 different CNNs (MobilenetV2, Shufflenet, Xception, Darknet53 and EfficientnetB0) in combination with kernel support vector machine is used for the diagnosis of COVID-19 from CT scans. The method was tested using a public dataset and it attained an area under the receiver operating characteristic curve of 0.963, accuracy of 0.916, kappa score of 0.8305, F-score of 0.91, sensitivity of 0.917 and positive predictive value of 0.904 in the prediction of COVID-19.
7 citations
Posted Content•
TL;DR: In this paper, a deep learning approach was presented to analyze the acoustic dataset provided in Track 1 of the DiCOVA 2021 Challenge containing cough sound recordings belonging to both COVID-19 positive and negative examples.
Abstract: With the periodic rise and fall of COVID-19 and countries being inflicted by its waves, an efficient, economic, and effortless diagnosis procedure for the virus has been the utmost need of the hour. COVID-19 positive individuals may even be asymptomatic making the diagnosis difficult, but amongst the infected subjects, the asymptomatic ones need not be entirely free of symptoms caused by the virus. They might not show any observable symptoms like the symptomatic subjects, but they may differ from uninfected ones in the way they cough. These differences in the coughing sounds are minute and indiscernible to the human ear, however, these can be captured using machine learning-based statistical models. In this paper, we present a deep learning approach to analyze the acoustic dataset provided in Track 1 of the DiCOVA 2021 Challenge containing cough sound recordings belonging to both COVID-19 positive and negative examples. To perform the classification on the sound recordings as belonging to a COVID-19 positive or negative examples, we propose a ConvNet model. Our model achieved an AUC score percentage of 72.23 on the blind test set provided by the same for an unbiased evaluation of the models. The ConvNet model incorporated with Data Augmentation further increased the AUC-ROC percentage from 72.23 to 87.07. It also outperformed the DiCOVA 2021 Challenge's baseline model by 23% thus, claiming the top position on the DiCOVA 2021 Challenge leaderboard. This paper proposes the use of Mel frequency cepstral coefficients as the feature input for the proposed model.
6 citations
Posted Content•
TL;DR: In this article, an ensemble-based multi-criteria decision-making (MCDM) method was proposed for selecting top performance machine learning technique(s) for COVID-19 cough classification.
Abstract: The objectives of this research are analysing the performance of the state-of-the-art machine learning techniques for classifying COVID-19 from cough sound and identifying the model(s) that consistently perform well across different cough datasets. Different performance evaluation metrics (such as precision, sensitivity, specificity, AUC, accuracy, etc.) make it difficult to select the best performance model. To address this issue, in this paper, we propose an ensemble-based multi-criteria decision making (MCDM) method for selecting top performance machine learning technique(s) for COVID-19 cough classification. We use four cough datasets, namely Cambridge, Coswara, Virufy, and NoCoCoDa to verify the proposed method. At first, our proposed method uses the audio features of cough samples and then applies machine learning (ML) techniques to classify them as COVID-19 or non-COVID-19. Then, we consider a multi-criteria decision-making (MCDM) method that combines ensemble technologies (i.e., soft and hard) to select the best model. In MCDM, we use the technique for order preference by similarity to ideal solution (TOPSIS) for ranking purposes, while entropy is applied to calculate evaluation criteria weights. In addition, we apply the feature reduction process through recursive feature elimination with cross-validation under different estimators. The results of our empirical evaluations show that the proposed method outperforms the state-of-the-art models.
2 citations
17 Oct 2021
1 citations
References
More filters
27 Jun 2016
TL;DR: In this article, the authors proposed a residual learning framework to ease the training of networks that are substantially deeper than those used previously, which won the 1st place on the ILSVRC 2015 classification task.
Abstract: Deeper neural networks are more difficult to train. We present a residual learning framework to ease the training of networks that are substantially deeper than those used previously. We explicitly reformulate the layers as learning residual functions with reference to the layer inputs, instead of learning unreferenced functions. We provide comprehensive empirical evidence showing that these residual networks are easier to optimize, and can gain accuracy from considerably increased depth. On the ImageNet dataset we evaluate residual nets with a depth of up to 152 layers—8× deeper than VGG nets [40] but still having lower complexity. An ensemble of these residual nets achieves 3.57% error on the ImageNet test set. This result won the 1st place on the ILSVRC 2015 classification task. We also present analysis on CIFAR-10 with 100 and 1000 layers. The depth of representations is of central importance for many visual recognition tasks. Solely due to our extremely deep representations, we obtain a 28% relative improvement on the COCO object detection dataset. Deep residual nets are foundations of our submissions to ILSVRC & COCO 2015 competitions1, where we also won the 1st places on the tasks of ImageNet detection, ImageNet localization, COCO detection, and COCO segmentation.
123,388 citations
TL;DR: The outbreak of the 2019 novel coronavirus disease (COVID-19) has induced a considerable degree of fear, emotional stress and anxiety among individuals around the world.
Abstract: The outbreak of the 2019 novel coronavirus disease (COVID-19) has induced a considerable degree of fear, emotional stress and anxiety among individuals around t
8,336 citations
TL;DR: Diagnostic and surveillance technologies for SARS-CoV-2 and their performance characteristics are described and point-of-care diagnostics that are on the horizon are described to encourage academics to advance their technologies beyond conception.
Abstract: COVID-19 has spread globally since its discovery in Hubei province, China in December 2019. A combination of computed tomography imaging, whole genome sequencing, and electron microscopy were initially used to screen and identify SARS-CoV-2, the viral etiology of COVID-19. The aim of this review article is to inform the audience of diagnostic and surveillance technologies for SARS-CoV-2 and their performance characteristics. We describe point-of-care diagnostics that are on the horizon and encourage academics to advance their technologies beyond conception. Developing plug-and-play diagnostics to manage the SARS-CoV-2 outbreak would be useful in preventing future epidemics.
1,335 citations
TL;DR: It is confirmed that fever and cough are the most prevalent symptoms of adults infected by SARS-CoV-2, however, there is a large proportion of infected adults which symptoms-alone do not identify.
Abstract: Background To limit the spread of SARS-CoV-2, an evidence-based understanding of the symptoms is critical to inform guidelines for quarantining and testing. The most common features are purported to be fever and a new persistent cough, although the global prevalence of these symptoms remains unclear. The aim of this systematic review is to determine the prevalence of symptoms associated with COVID-19 worldwide. Methods We searched PubMed, Embase, CINAHL, AMED, medRxiv and bioRxiv on 5th April 2020 for studies of adults (>16 years) with laboratory test confirmed COVID-19. No language or publication status restrictions were applied. Data were independently extracted by two review authors into standardised forms. All datapoints were independently checked by three other review authors. A random-effects model for pooling of binomial data was applied to estimate the prevalence of symptoms, subgrouping estimates by country. I2 was used to assess inter-study heterogeneity. Results Of 851 unique citations, 148 articles were included which comprised 24,410 adults with confirmed COVID-19 from 9 countries. The most prevalent symptoms were fever (78% [95% CI 75%-81%]; 138 studies, 21,701 patients; I2 94%), a cough (57% [95% CI 54%-60%]; 138 studies, 21,682 patients; I2 94%) and fatigue (31% [95% CI 27%-35%]; 78 studies, 13,385 patients; I2 95%). Overall, 19% of hospitalised patients required non-invasive ventilation (44 studies, 6,513 patients), 17% required intensive care (33 studies, 7504 patients), 9% required invasive ventilation (45 studies, 6933 patients) and 2% required extra-corporeal membrane oxygenation (12 studies, 1,486 patients). The mortality rate was 7% (73 studies, 10,402 patients). Conclusions We confirm that fever and cough are the most prevalent symptoms of adults infected by SARS-CoV-2. However, there is a large proportion of infected adults which symptoms-alone do not identify.
435 citations
29 Sep 2020
TL;DR: An AI speech processing framework that leverages acoustic biomarker feature extractors to pre-screen for COVID-19 from cough recordings, and provide a personalized patient saliency map to longitudinally monitor patients in real-time, non-invasively, and at essentially zero variable cost is developed.
Abstract: Goal: We hypothesized that COVID-19 subjects, especially including asymptomatics, could be accurately discriminated only from a forced-cough cell phone recording using Artificial Intelligence. To train our MIT Open Voice model we built a data collection pipeline of COVID-19 cough recordings through our website (opensigma.mit.edu) between April and May 2020 and created the largest audio COVID-19 cough balanced dataset reported to date with 5,320 subjects. Methods: We developed an AI speech processing framework that leverages acoustic biomarker feature extractors to pre-screen for COVID-19 from cough recordings, and provide a personalized patient saliency map to longitudinally monitor patients in real-time, non-invasively, and at essentially zero variable cost. Cough recordings are transformed with Mel Frequency Cepstral Coefficient and inputted into a Convolutional Neural Network (CNN) based architecture made up of one Poisson biomarker layer and 3 pre-trained ResNet50's in parallel, outputting a binary pre-screening diagnostic. Our CNN-based models have been trained on 4256 subjects and tested on the remaining 1064 subjects of our dataset. Transfer learning was used to learn biomarker features on larger datasets, previously successfully tested in our Lab on Alzheimer's, which significantly improves the COVID-19 discrimination accuracy of our architecture. Results: When validated with subjects diagnosed using an official test, the model achieves COVID-19 sensitivity of 98.5% with a specificity of 94.2% (AUC: 0.97). For asymptomatic subjects it achieves sensitivity of 100% with a specificity of 83.2% . Conclusions: AI techniques can produce a free, non-invasive, real-time, any-time, instantly distributable, large-scale COVID-19 asymptomatic screening tool to augment current approaches in containing the spread of COVID-19. Practical use cases could be for daily screening of students, workers, and public as schools, jobs, and transport reopen, or for pool testing to quickly alert of outbreaks in groups. General speech biomarkers may exist that cover several disease categories, as we demonstrated using the same ones for COVID-19 and Alzheimer's.
403 citations