A Smartphone-Based Platform Assisted by Artificial Intelligence for Reading and Reporting Rapid Diagnostic Tests: Evaluation Study in SARS-CoV-2 Lateral Flow Immunoassays

BACKGROUND

Rapid diagnostic tests (RDTs) are being widely used to manage COVID-19 pandemic. However, many results remain unreported or unconfirmed, altering a correct epidemiological surveillance.

OBJECTIVE

Our aim was to evaluate an artificial intelligence-based smartphone app, connected to a cloud web platform, to automatically and objectively read RDT results and assess its impact on COVID-19 pandemic management.

METHODS

Overall, 252 human sera were used to inoculate a total of 1165 RDTs for training and validation purposes. We then conducted two field studies to assess the performance on real-world scenarios by testing 172 antibody RDTs at two nursing homes and 96 antigen RDTs at one hospital emergency department.

RESULTS

Field studies demonstrated high levels of sensitivity (100%) and specificity (94.4%, CI 92.8%-96.1%) for reading IgG band of COVID-19 antibody RDTs compared to visual readings from health workers. Sensitivity of detecting IgM test bands was 100%, and specificity was 95.8% (CI 94.3%-97.3%). All COVID-19 antigen RDTs were correctly read by the app.

CONCLUSIONS

The proposed reading system is automatic, reducing variability and uncertainty associated with RDTs interpretation and can be used to read different RDT brands. The web platform serves as a real-time epidemiological tracking tool and facilitates reporting of positive RDTs to relevant health authorities.

T-cell response after first dose of BNT162b2 SARS-CoV-2 vaccine among healthcare workers with previous infection or cross-reactive immunity

Objectives

Antibody response to the first dose of BNT162b2 SARS‐CoV‐2 is greater in COVID‐19‐convalescent than in infection‐naïve individuals. However, there are no data about T‐cell response in individuals with pre‐existing cellular immunity.

Methods

We evaluated T‐cell responses in parallel with SARS‐CoV‐2 antibody level after first dose of BNT162b2 vaccine in 23 infection‐naïve and 27 convalescent healthcare workers (HCWs) previously included in a study about humoral and T‐cell immunity.

Results

Overall, the antibody response was lower in the infection‐naïve group than in convalescent individuals (18 895 vs 662.7 AU mL⁻¹, P < 0.001), and intermediate but significantly lower in convalescent HCWs with previous negative serology (25 174 vs 1793 AU mL⁻¹; P = 0.015). Indeed, anti‐spike IgG titres after the first dose correlated with baseline anti‐nucleocapsid IgG titres (rho = 0.689; P < 0.001). Pre‐existing T‐cell immunity was observed in 78% of convalescent and 65% of the infection‐naïve HCWs. T‐cell response after the first dose of the vaccine was observed in nearly all the cases with pre‐existing T‐cell immunity, reaching 94% in convalescent HCWs and 93% in those with cross‐reactive T cells. It was lower in the infection‐naïve group (50%; P = 0.087) and in convalescent HCWs with negative serology (56%; P = 0.085). Notably, systemic reactogenicity after vaccination was mainly observed in those with pre‐existing T‐cell immunity (P = 0.051).

Conclusion

Here, we report that the first dose of BTN162b2 elicits a similar S‐specific T‐cell response in cases of either past infection or cross‐reactive T cells, but lower in the rest of infection‐naïve individuals and in convalescent HCWs who have lost detectable specific antibodies during follow‐up.

High Prevalence of Co-Infections by Invasive and Non-Invasive Chlamydia trachomatis Genotypes during the Lymphogranuloma Venereum Outbreak in Spain

The evolution of Chlamydia trachomatis is mainly driven by recombination events. This fact can be fuelled by the coincidence in several European regions of the high prevalence of non-invasive urogenital genotypes and lymphogranuloma venereum (LGV) outbreaks. This scenario could modify the local epidemiology and favor the selection of new C. trachomatis variants. Quantifying the prevalence of co-infection could help to predict the potential risk in the selection of new variants with unpredictable results in pathogenesis or transmissibility. In the 2009-2013 period, 287 clinical samples with demonstrated presence of C. trachomatis were selected. They were divided in two groups. The first group was constituted by 137 samples with C. trachomatis of the LGV genotypes, and the second by the remaining 150 samples in which the presence of LGV genotypes was previously excluded. They were analyzed to detect the simultaneous presence of non-LGV genotypes based on pmpH and ompA genes. In the first group, co-infections were detected in 10.9% of the cases whereas in the second group the prevalence was 14.6%, which is the highest percentage ever described among European countries. Moreover, bioinformatic analyses suggested the presence among men who have sex with men of a pmpH-recombinant variant, similar to strains described in Seattle in 2002. This variant was the result of genetic exchange between genotypes belonging to LGV and members of G-genotype. Sequencing of other genes, phylogenetically related to pathotype, confirmed that the putative recombinant found in Madrid could have a common origin with the strains described in Seattle. Countries with a high prevalence of co-infections and high migration flows should enhance surveillance programs in at least their vulnerable population.