An Approach for COVID-19 Identification from Chest X-ray Images Using High-Resolution Networks

In this paper, we present an approach for COVID-19 identification from chest X-ray images by using high-resolution neural networks. These networks allow to connect high-to-low convolution streams in parallel. They can maintain high-resolution representations and generate different resolutions throughout the whole process. The high-resolution based models have shown the superior performance in several applications. The experiments were evaluated on a collection of three data sources containing 24,786 lung X-ray images, which were categorized into three classes including covid pneumonia, non-pneumonia, and viral pneumonia. The proposed approach can attain the overall accuracy of 98.2% and 97.56% for the training and testing set, respectively. The accuracy for each class is 99.37%, 94.83%, and 97.27%, respectively, for non-pneumonia, covid-pneumonia, and viral-pneumonia. © 2022 IEEE.

MODELING OF PRE-TRAINED NEURAL NETWORK EMBEDDINGS LEARNED FROM RAW WAVEFORM FOR COVID-19 INFECTION DETECTION

COVID-19 is a respiratory system disorder that can disrupt the function of lungs. Effects of dysfunctional respiratory mechanism can reflect upon other modalities which function in close coupling. Audio signals result from modulation of respiration through speech production system, and hence acoustic information can be modeled for detection of COVID-19. In that direction, this paper is addressing the second DiCOVA challenge that deals with COVID-19 detection based on speech, cough and breathing. We investigate modeling of (a) ComParE LLD representations derived at frame- and turn-level resolutions and (b) neural representations obtained from pre-trained neural networks trained to recognize phones and estimate breathing patterns. On Track 1, the ComParE LLD representations yield a best performance of 78.05% area under the curve (AUC). Experimental studies on Track 2 and Track 3 demonstrate that neural representations tend to yield better detection than ComParE LLD representations. Late fusion of different utterance level representations of neural embeddings yielded a best performance of 80.64% AUC. © 2022 IEEE

Effective hybrid deep learning model for COVID-19 patterns identification using CT images.

Coronavirus disease 2019 (COVID-19) has attracted significant attention of researchers from various disciplines since the end of 2019. Although the global epidemic situation is stabilizing due to vaccination, new COVID-19 cases are constantly being discovered around the world. As a result, lung computed tomography (CT) examination, an aggregated identification technique, has been used to ameliorate diagnosis. It helps reveal missed diagnoses due to the ambiguity of nucleic acid polymerase chain reaction. Therefore, this study investigated how quickly and accurately hybrid deep learning (DL) methods can identify infected individuals with COVID-19 on the basis of their lung CT images. In addition, this study proposed a developed system to create a reliable COVID-19 prediction network using various layers starting with the segmentation of the lung CT scan image and ending with disease prediction. The initial step of the system starts with a proposed technique for lung segmentation that relies on a no-threshold histogram-based image segmentation method. Afterward, the GrabCut method was used as a post-segmentation method to enhance segmentation outcomes and avoid over-and under-segmentation problems. Then, three pre-trained models of standard DL methods, including Visual Geometry Group Network, convolutional deep belief network, and high-resolution network, were utilized to extract the most affective features from the segmented images that can help to identify COVID-19. These three described pre-trained models were combined as a new mechanism to increase the system's overall prediction capabilities. A publicly available dataset, namely, COVID-19 CT, was used to test the performance of the proposed model, which obtained a 95% accuracy rate. On the basis of comparison, the proposed model outperformed several state-of-the-art studies. Because of its effectiveness in accurately screening COVID-19 CT images, the developed model will potentially be valuable as an additional diagnostic tool for leading clinical professionals.

Detection of COVID-19 from CT scan images using deep neural networks

The coronavirus pandemic brought the world to a standstill of historic significance. Countries over the world have imposed lockdowns, quarantines and travel bans in an effort to stop the further spread of the disease. Healthcare systems worldwide are under extreme pressure due to the influx of a large amount of patients suffering from COVID-19. Moreover, there is a dearth of doctors, nurses, and support staff in hospitals of many countries. In such a predicament, it is imperative to leverage the advances made in computer vision and deep learning technologies to create a system that attempts to ease the burden on worldwide healthcare. In this research, ten state-of-the-art pre-trained convolutional neural networks were used to identify COVID-19 in chest Computed Tomography (CT) scan images. After extensive experimental testing and tuning, comprehensive comparative analysis was done and very promising results were obtained in this classification task. © 2021 IEEE

On-field evaluation of a ultra-rapid fluorescence immunoassay as a frontline test for SARS-CoV-2 diagnostic.

BACKGROUND: Viral RNA amplification by real-time RT-PCR still represents the gold standard for the detection of SARS-CoV-2, but the development of rapid, reliable and easy-to-perform diagnostic methods is crucial for public health, because of the need of shortening the time of result-reporting with a cost-efficient approach. OBJECTIVES: The aim of our research was to assess the performance of FREND™ COVID-19 Ag assay (NanoEntek, South Korea) as a ultra-rapid frontline test for SARS-CoV-2 identification, in comparison with RT-PCR and another COVID-19 antigen fluorescence immunoassay (FIA). STUDY DESIGN: The qualitative FIA FREND™ test, designed to detect within 3 min the Nucleocapsid protein of SARS-CoV-2, was evaluated using nasopharyngeal swabs in Universal Transport Medium (UTM™, Copan Diagnostics Inc, US) from suspected COVID-19 cases who accessed the Emergency Room of the Ospedale Policlinico San Martino, Genoa, Liguria, Northwest Italy. Diagnostic accuracy was determined in comparison with SARS-CoV-2 RT-PCR and STANDARD F™ COVID-19 Ag FIA test (SD BIOSENSOR Inc., Republic of Korea). RESULTS: In November 2020, 110 nasopharyngeal samples were collected consecutively; 60 resulted RT-PCR positive. With respect to RT-PCR results, sensitivity and specificity of FREND™ COVID-19 Ag test were 93.3 % (95 % CI: 83.8-98.2) and 100 % (95 % CI: 92.9-100), respectively. FREND™and STANDARD F™ COVID-19 Ag FIA assays showed a concordance of 96.4 % (Cohen's k = 0.93, 95 % CI: 0.86-0.99). CONCLUSIONS: FREND™ FIA test showed high sensitivity and specificity in nasopharyngeal swabs. The assay has the potential to become an important tool for an ultra-rapid identification of SARS-CoV-2 infection, particularly in situations with limited access to molecular diagnostics.