Your browser doesn't support javascript.
Show: 20 | 50 | 100
Results 1 - 2 de 2
Filter
1.
TrAC - Trends in Analytical Chemistry ; 157 (no pagination), 2022.
Article in English | EMBASE | ID: covidwho-2236003

ABSTRACT

Polymerase chain reaction (PCR) amplifies specific fragment of DNA molecules and has been extensively applied in fields of pathogens and gene mutation detection, food safety and clinical diagnosis which on the other hand, holds the drawbacks of large size instrument, high heat dissipation etc. It has been demonstrated that microfluidics technique coupling with PCR reaction exhibits characteristics of integration, automatization, miniaturization, and portability. Meanwhile, various designed fabrication of microchip could contribute to diverse applications. In this review, we summarized major works about a variety of microfluidic chips equipped with several kinds of PCR techniques (PCR, RT-PCR, mPCR, dPCR) and detection methods like fluorescence, electrochemistry, and electrophoresis detection. The development and application of PCR-based microfluidic chip in pathogen and gene mutation detection, diseases prevention and diagnosis, DNA hybridization and low-volume sample treatment were also discussed. Copyright © 2022 Elsevier B.V.

2.
5th International Conference on Future Networks and Distributed Systems: The Premier Conference on Smart Next Generation Networking Technologies, ICFNDS 2021 ; : 155-161, 2021.
Article in English | Scopus | ID: covidwho-1832589

ABSTRACT

The newly detected Coronavirus pneumonia, dubbed COVID-19, is highly contagious and pathogenic. To combat this disease, the diagnostic step is mostly carried out utilizing the RT-PCR technique on nasopharyngeal and throat samples with sensitivity values ranging from 30 to 70%. Biomedical imaging, on the other hand, has sensitivity levels of 98 and 69 percent, respectively. In this paper, a machine learning model is built using convolutional neural networks (CNN) with 5 CNN architectures: VGG16, MobileNetV2, NASNetMobile, and ResNet-50. The presented model scored a precision rate of 81%, a recall rate of 72%, and an f1-score of 71%. Moreover, this research paper accommodates a proposed expansion to the existing model. The Expansion suggested is to create a lightweight version of the model for smartphones © 2021 ACM.

SELECTION OF CITATIONS
SEARCH DETAIL