Virus Detection and Identification in Minutes Using Single-Particle Imaging and Deep Learning

Development of a dual-component biosensor for rapid and sensitive detection of influenza H7 and H5 subtypes

The outbreak of highly pathogenic influenza virus subtypes, such as H7 and H5, presents a significant global health challenge, necessitating the development of rapid and sensitive diagnostic methods. In this study, we have developed a novel dual-component biosensor assembly, each component of which incorporates an antibody fused with a nano-luciferase subunit. Our results demonstrate the effectiveness of this biosensor in enabling the rapid and sensitive detection of influenza H7 and other subtypes. Additionally, we successfully applied the biosensor in paper-based assay and lateral flow assay formats, expanding its versatility and potential for field-deployable applications. Notably, we achieved effective detection of the H7N9 virus using this biosensor. Furthermore, we designed and optimized a dedicated biosensor to the sensitive detection of the influenza H5 subtype. Collectively, our findings underscore the significant potential of this dual-component biosensor assembly as a valuable and versatile tool for accurate and timely diagnosis of influenza virus infections, promising to advance the field of influenza diagnostics and contribute to outbreak management and surveillance efforts.

Significance of Artificial Intelligence in the Study of Virus-Host Cell Interactions

A highly critical event in a virus's life cycle is successfully entering a given host. This process begins when a viral glycoprotein interacts with a target cell receptor, which provides the molecular basis for target virus-host cell interactions for novel drug discovery. Over the years, extensive research has been carried out in the field of virus-host cell interaction, generating a massive number of genetic and molecular data sources. These datasets are an asset for predicting virus-host interactions at the molecular level using machine learning (ML), a subset of artificial intelligence (AI). In this direction, ML tools are now being applied to recognize patterns in these massive datasets to predict critical interactions between virus and host cells at the protein-protein and protein-sugar levels, as well as to perform transcriptional and translational analysis. On the other end, deep learning (DL) algorithms-a subfield of ML-can extract high-level features from very large datasets to recognize the hidden patterns within genomic sequences and images to develop models for rapid drug discovery predictions that address pathogenic viruses displaying heightened affinity for receptor docking and enhanced cell entry. ML and DL are pivotal forces, driving innovation with their ability to perform analysis of enormous datasets in a highly efficient, cost-effective, accurate, and high-throughput manner. This review focuses on the complexity of virus-host cell interactions at the molecular level in light of the current advances of ML and AI in viral pathogenesis to improve new treatments and prevention strategies.

Integrated Antigenic and Nucleic Acid Detection in Single Virions and Extracellular Vesicles with Viral Content

Virion-mediated outbreaks are imminent and despite rapid responses, continue to cause adverse symptoms and death. Therefore, tunable, sensitive, high-throughput assays are needed to help diagnose future virion-mediated outbreaks. Herein, it is developed a tunable in situ assay to selectively enrich virions and extracellular vesicles (EVs) and simultaneously detect antigens and nucleic acids at a single-particle resolution. The Biochip Antigen and RNA Assay (BARA) enhanced sensitivities compared to quantitative reverse-transcription polymerase chain reaction (qRT-PCR), enabling the detection of virions in asymptomatic patients, genetic mutations in single virions, and enabling the continued long-term expression of viral RNA in the EV-enriched subpopulation in the plasma of patients with post-acute sequelae of the coronavirus disease of 2019 (COVID-19). BARA revealed highly accurate diagnoses of COVID-19 by simultaneously detecting the spike glycoprotein and nucleocapsid-encoding RNA in saliva and nasopharyngeal swab samples. Altogether, the single-particle detection of antigens and viral RNA provides a tunable framework for the diagnosis, monitoring, and mutation screening of current and future outbreaks.

Tackling the Antimicrobial Resistance "Pandemic" with Machine Learning Tools: A Summary of Available Evidence

Antimicrobial resistance is recognised as one of the top threats healthcare is bound to face in the future. There have been various attempts to preserve the efficacy of existing antimicrobials, develop new and efficient antimicrobials, manage infections with multi-drug resistant strains, and improve patient outcomes, resulting in a growing mass of routinely available data, including electronic health records and microbiological information that can be employed to develop individualised antimicrobial stewardship. Machine learning methods have been developed to predict antimicrobial resistance from whole-genome sequencing data, forecast medication susceptibility, recognise epidemic patterns for surveillance purposes, or propose new antibacterial treatments and accelerate scientific discovery. Unfortunately, there is an evident gap between the number of machine learning applications in science and the effective implementation of these systems. This narrative review highlights some of the outstanding opportunities that machine learning offers when applied in research related to antimicrobial resistance. In the future, machine learning tools may prove to be superbugs' kryptonite. This review aims to provide an overview of available publications to aid researchers that are looking to expand their work with new approaches and to acquaint them with the current application of machine learning techniques in this field.

Tools and methods for high-throughput single-cell imaging with the mother machine

Despite much progress, image processing remains a significant bottleneck for high-throughput analysis of microscopy data. One popular platform for single-cell time-lapse imaging is the mother machine, which enables long-term tracking of microbial cells under precisely controlled growth conditions. While several mother machine image analysis pipelines have been developed in the past several years, adoption by a non-expert audience remains a challenge. To fill this gap, we implemented our own software, MM3, as a plugin for the multidimensional image viewer napari. napari-MM3 is a complete and modular image analysis pipeline for mother machine data, which takes advantage of the high-level interactivity of napari. Here, we give an overview of napari-MM3 and test it against several well-designed and widely used image analysis pipelines, including BACMMAN and DeLTA. Researchers often analyze mother machine data with custom scripts using varied image analysis methods, but a quantitative comparison of the output of different pipelines has been lacking. To this end, we show that key single-cell physiological parameter correlations and distributions are robust to the choice of analysis method. However, we also find that small changes in thresholding parameters can systematically alter parameters extracted from single-cell imaging experiments. Moreover, we explicitly show that in deep learning-based segmentation, 'what you put is what you get' (WYPIWYG) - that is, pixel-level variation in training data for cell segmentation can propagate to the model output and bias spatial and temporal measurements. Finally, while the primary purpose of this work is to introduce the image analysis software that we have developed over the last decade in our lab, we also provide information for those who want to implement mother machine-based high-throughput imaging and analysis methods in their research.

Rapid and specific detection of single nanoparticles and viruses in microfluidic laminar flow via confocal fluorescence microscopy

Mainstream virus detection relies on the specific amplification of nucleic acids via polymerase chain reaction, a process that is slow and requires extensive laboratory expertise and equipment. Other modalities, such as antigen-based tests, allow much faster virus detection but have reduced sensitivity. In this study, we report the development of a flow virometer for the specific and rapid detection of single nanoparticles based on confocal microscopy. The combination of laminar flow and multiple dyes enable the detection of correlated fluorescence signals, providing information on nanoparticle volumes and specific chemical composition properties, such as viral envelope proteins. We evaluated and validated the assay using fluorescent beads and viruses, including SARS-CoV-2. Additionally, we demonstrate how hydrodynamic focusing enhances the assay sensitivity for detecting clinically-relevant virus loads. Based on our results, we envision the use of this technology for clinically relevant bio-nanoparticles, supported by the implementation of the assay in a portable and user-friendly setup.

Mosaic quadrivalent influenza vaccine single nanoparticle characterization

Recent work by our laboratory and others indicates that co-display of multiple antigens on protein-based nanoparticles may be key to induce cross-reactive antibodies that provide broad protection against disease. To reach the ultimate goal of a universal vaccine for seasonal influenza, a mosaic influenza nanoparticle vaccine (FluMos-v1) was developed for clinical trial (NCT04896086). FluMos-v1 is unique in that it is designed to co-display four recently circulating haemagglutinin (HA) strains; however, current vaccine analysis techniques are limited to nanoparticle population analysis, thus, are unable to determine the valency of an individual nanoparticle. For the first time, we demonstrate by total internal reflection fluorescence microscopy and supportive physical-chemical methods that the co-display of four antigens is indeed achieved in single nanoparticles. Additionally, we have determined percentages of multivalent (mosaic) nanoparticles with four, three, or two HA proteins. The integrated imaging and physicochemical methods we have developed for single nanoparticle multivalency will serve to further understand immunogenicity data from our current FluMos-v1 clinical trial.

Tools and methods for high-throughput single-cell imaging with the mother machine

Despite much progress, image processing remains a significant bottleneck for high-throughput analysis of microscopy data. One popular platform for single-cell time-lapse imaging is the mother machine, which enables long-term tracking of microbial cells under precisely controlled growth conditions. While several mother machine image analysis pipelines have been developed in the past several years, adoption by a non-expert audience remains a challenge. To fill this gap, we implemented our own software, MM3, as a plugin for the multidimensional image viewer napari. napari-MM3 is a complete and modular image analysis pipeline for mother machine data, which takes advantage of the high-level interactivity of napari. Here, we give an overview of napari-MM3 and test it against several well-designed and widely-used image analysis pipelines, including BACMMAN and DeLTA. Researchers often analyze mother machine data with custom scripts using varied image analysis methods, but a quantitative comparison of the output of different pipelines has been lacking. To this end, we show that key single-cell physiological parameter correlations and distributions are robust to the choice of analysis method. However, we also find that small changes in thresholding parameters can systematically alter parameters extracted from single-cell imaging experiments. Moreover, we explicitly show that in deep learning based segmentation, "what you put is what you get" (WYPIWYG) - i.e., pixel-level variation in training data for cell segmentation can propagate to the model output and bias spatial and temporal measurements. Finally, while the primary purpose of this work is to introduce the image analysis software that we have developed over the last decade in our lab, we also provide information for those who want to implement mother-machine-based high-throughput imaging and analysis methods in their research.

COVID-19 lateral flow test image classification using deep CNN and StyleGAN2

Introduction

Artificial intelligence (AI) in healthcare can enhance clinical workflows and diagnoses, particularly in large-scale operations like COVID-19 mass testing. This study presents a deep Convolutional Neural Network (CNN) model for automated COVID-19 RATD image classification.

Methods

To address the absence of a RATD image dataset, we crowdsourced 900 real-world images focusing on positive and negative cases. Rigorous data augmentation and StyleGAN2-ADA generated simulated images to overcome dataset limitations and class imbalances.

Results

The best CNN model achieved a 93% validation accuracy. Test accuracies were 88% for simulated datasets and 82% for real datasets. Augmenting simulated images during training did not significantly improve real-world test image performance but enhanced simulated test image performance.

Discussion

The findings of this study highlight the potential of the developed model in expediting COVID-19 testing processes and facilitating large-scale testing and tracking systems. The study also underscores the challenges in designing and developing such models, emphasizing the importance of addressing dataset limitations and class imbalances.

Conclusion

This research contributes to the deployment of large-scale testing and tracking systems, offering insights into the potential applications of AI in mitigating outbreaks similar to COVID-19. Future work could focus on refining the model and exploring its adaptability to other healthcare scenarios.

Impact of Artificial Intelligence (AI) Technology in Healthcare Sector: A Critical Evaluation of Both Sides of the Coin

The influence of artificial intelligence (AI) has drastically risen in recent years, especially in the field of medicine. Its influence has spread so greatly that it is determined to become a pillar in the future medical world. A comprehensive literature search related to AI in healthcare was performed in the PubMed database and retrieved the relevant information from suitable ones. AI excels in aspects such as rapid adaptation, high diagnostic accuracy, and data management that can help improve workforce productivity. With this potential in sight, the FDA has continuously approved more machine learning (ML) software to be used by medical workers and scientists. However, there are few controversies such as increased chances of data breaches, concern for clinical implementation, and potential healthcare dilemmas. In this article, the positive and negative aspects of AI implementation in healthcare are discussed, as well as recommended some potential solutions to the potential issues at hand.

High Confidence Single Particle Analysis with Machine Learning

Single particle analysis can effectively determine the heterogeneity between particles based on the local information on a single particle, which is utilized extensively for monitoring chemical reactions and biological activities. However, the study of obtaining ensemble reaction information at the single particle level, which can obtain both the structural and functional heterogeneity of particles as well as the ensemble reaction information, is challenging because the selection of a single particle mainly depends on experience, which will lead to a certain randomness when analyzing the ensemble reaction with single particles. Using machine learning, it is demonstrated that the proposed intelligent single particle analysis strategy can provide single particle and ensemble analyses with high confidence. Convolutional neural network and Gaussian mixture model were utilized to develop a machine learning model for resonance scattering imaging analysis of plasmonic nanoparticles. It can identify the scattered light of single particles and select representative or diverse particles. When single particle scattering imaging is used to obtain ensemble information on the reaction, the error caused by the selection of individual particles can be significantly reduced by selecting representative particles. In addition, the real situation of the reaction can be better revealed by selecting diverse particles. These results indicate that the intelligent single particle analysis strategy has great potential for imaging analysis and biological sensing.

Advanced fluorescence microscopy in respiratory virus cell biology

Respiratory viruses are a major public health burden across all age groups around the globe, and are associated with high morbidity and mortality rates. They can be transmitted by multiple routes, including physical contact or droplets and aerosols, resulting in efficient spreading within the human population. Investigations of the cell biology of virus replication are thus of utmost importance to gain a better understanding of virus-induced pathogenicity and the development of antiviral countermeasures. Light and fluorescence microscopy techniques have revolutionized investigations of the cell biology of virus infection by allowing the study of the localization and dynamics of viral or cellular components directly in infected cells. Advanced microscopy including high- and super-resolution microscopy techniques available today can visualize biological processes at the single-virus and even single-molecule level, thus opening a unique view on virus infection. We will highlight how fluorescence microscopy has supported investigations on virus cell biology by focusing on three major respiratory viruses: respiratory syncytial virus (RSV), Influenza A virus (IAV) and SARS-CoV-2. We will review our current knowledge of virus replication and highlight how fluorescence microscopy has helped to improve our state of understanding. We will start by introducing major imaging and labeling modalities and conclude the chapter with a perspective discussion on remaining challenges and potential opportunities.

Solid-Phase Optical Sensing Techniques for Sensitive Virus Detection

Viral infections can pose a major threat to public health by causing serious illness, leading to pandemics, and burdening healthcare systems. The global spread of such infections causes disruptions to every aspect of life including business, education, and social life. Fast and accurate diagnosis of viral infections has significant implications for saving lives, preventing the spread of the diseases, and minimizing social and economic damages. Polymerase chain reaction (PCR)-based techniques are commonly used to detect viruses in the clinic. However, PCR has several drawbacks, as highlighted during the recent COVID-19 pandemic, such as long processing times and the requirement for sophisticated laboratory instruments. Therefore, there is an urgent need for fast and accurate techniques for virus detection. For this purpose, a variety of biosensor systems are being developed to provide rapid, sensitive, and high-throughput viral diagnostic platforms, enabling quick diagnosis and efficient control of the virus's spread. Optical devices, in particular, are of great interest due to their advantages such as high sensitivity and direct readout. The current review discusses solid-phase optical sensing techniques for virus detection, including fluorescence-based sensors, surface plasmon resonance (SPR), surface-enhanced Raman scattering (SERS), optical resonators, and interferometry-based platforms. Then, we focus on an interferometric biosensor developed by our group, the single-particle interferometric reflectance imaging sensor (SP-IRIS), which has the capability to visualize single nanoparticles, to demonstrate its application for digital virus detection.

Machine Learning-Assisted Real-Time Polymerase Chain Reaction and High-Resolution Melt Analysis for SARS-CoV-2 Variant Identification

Since the declaration of COVID-19 as a pandemic in early 2020, multiple variants of the severe acute respiratory syndrome-related coronavirus (SARS-CoV-2) have been detected. The emergence of multiple variants has raised concerns due to their impact on public health. Therefore, it is crucial to distinguish between different viral variants. Here, we developed a machine learning web-based application for SARS-CoV-2 variant identification via duplex real-time polymerase chain reaction (PCR) coupled with high-resolution melt (qPCR-HRM) analysis. As a proof-of-concept, we investigated the platform's ability to identify the Alpha, Delta, and wild-type strains using two sets of primers. The duplex qPCR-HRM could identify the two variants reliably in as low as 100 copies/μL. Finally, the platform was validated with 167 nasopharyngeal swab samples, which gave a sensitivity of 95.2%. This work demonstrates the potential for use as automated, cost-effective, and large-scale viral variant surveillance.