Performance comparison between two computer-aided detection colonoscopy models by trainees using different false positive thresholds: a cross-sectional study in Thailand

BACKGROUND/AIMS

This study aims to compare polyp detection performance of "Deep-GI," a newly developed artificial intelligence (AI) model, to a previously validated AI model computer-aided polyp detection (CADe) using various false positive (FP) thresholds and determining the best threshold for each model.

METHODS

Colonoscopy videos were collected prospectively and reviewed by three expert endoscopists (gold standard), trainees, CADe (CAD EYE; Fujifilm Corp.), and Deep-GI. Polyp detection sensitivity (PDS), polyp miss rates (PMR), and false-positive alarm rates (FPR) were compared among the three groups using different FP thresholds for the duration of bounding boxes appearing on the screen.

RESULTS

In total, 170 colonoscopy videos were used in this study. Deep-GI showed the highest PDS (99.4% vs. 85.4% vs. 66.7%, p<0.01) and the lowest PMR (0.6% vs. 14.6% vs. 33.3%, p<0.01) when compared to CADe and trainees, respectively. Compared to CADe, Deep-GI demonstrated lower FPR at FP thresholds of ≥0.5 (12.1 vs. 22.4) and ≥1 second (4.4 vs. 6.8) (both p<0.05). However, when the threshold was raised to ≥1.5 seconds, the FPR became comparable (2 vs. 2.4, p=0.3), while the PMR increased from 2% to 10%.

CONCLUSION

Compared to CADe, Deep-GI demonstrated a higher PDS with significantly lower FPR at ≥0.5- and ≥1-second thresholds. At the ≥1.5-second threshold, both systems showed comparable FPR with increased PMR.

Long-Term Dynamic Changes in Hybrid Immunity over Six Months after Inactivated and Adenoviral Vector Vaccination in Individuals with Previous SARS-CoV-2 Infection

Numerous studies have largely focused on short-term immunogenicity in recovered individuals post mRNA vaccination. However, understanding the long-term durability, particularly in inactivated and adenoviral vectored vaccines, remains limited. We evaluated antibody responses, omicron variant neutralization, and IFN-γ responses in 119 previously infected individuals vaccinated with CoronaVac or ChAdOx1 up to six months post-vaccination. Both vaccines elicited robust immune responses in recovered individuals, surpassing those who were infection-naïve, and these persisted above pre-vaccination levels for six months. However, antibody levels declined over time (geometric mean ratio (GMR) = 0.52 for both vaccines). Notably, neutralizing activities against omicron declined faster in ChAdOx1 (GMR = 0.6) compared to CoronaVac recipients (GMR = 1.03). While the first dose of ChAdOx1 adequately induced immune responses in recovered individuals, a second dose demonstrated advantages in omicron variant neutralization and slower decline. Although both vaccines induced T cell responses, the median IFN-γ level at six months returned to pre-vaccination levels. However, more individuals exhibited reactive T cell responses. Extending the interval (13-15 months) between infection and vaccination could enhance antibody levels and broaden neutralization. Together, these findings demonstrate a robust humoral and cellular response that was sustained for at least six months after vaccination, thus guiding optimal vaccination strategies based on prior infection and vaccine platforms.

An amino acid substitution in HCV core antigen limits its use as a reliable measure of HCV infection compared with HCV RNA

Hepatitis C virus (HCV) is a viral pathogen that causes chronic hepatitis, which can lead to cirrhosis and hepatocellular carcinoma. Detection of HCV RNA is the standard method used to diagnose the disease and monitor antiviral treatment. A quantification assay for the HCV core antigen (HCVcAg) has been proposed as a simplified alternative to the HCV RNA test for predicting active HCV infection, with the aim of achieving the global goal of eliminating hepatitis. The objective of this study was to determine the correlation between HCV RNA and HCVcAg, as well as the impact of amino acid sequence heterogeneity on HCVcAg quantification. Our findings demonstrated a strong positive correlation between HCV RNA and HCVcAg across all HCV genotypes (1a, 1b, 3a, and 6), with correlation coefficients ranging from 0.88 to 0.96 (p < 0.001). However, in some cases, samples with genotypes 3a and 6 exhibited lower HCVcAg levels than expected based on the corresponding HCV RNA values. Upon the core amino acid sequence alignment, it was observed that samples exhibiting low core antigen levels had an amino acid substitution at position 49, where threonine was replaced by either alanine or valine. Core mutation at this position may correlate with one of the epitope regions recognized by anti-HCV monoclonal antibodies. The present findings suggest that the utilization of HCVcAg as a standalone marker for HCV RNA might not provide adequate sensitivity for the detection of HCV infection, especially in cases where there are variations in the amino acid sequence of the core region and a low viral load of HCV RNA.

Effectiveness of internet-based CBT-I for the treatment of chronic subthreshold to moderate insomnia

Study objectives

To study the effectiveness of the first internet-based cognitive behavioral therapy for insomnia (CBT-i) in Thailand, using the Nitra application, for chronic subthreshold to moderate insomnia treatment.

Methods

An interventional study without a control group was conducted between January and June 2022. Participants were adults aged 18 years old and older with subthreshold to moderate severity of chronic insomnia (insomnia severity index (ISI) of 8-21) and had mean sleep efficiency <85% from baseline sleep diaries. Baseline sleep characteristics were obtained from questionnaires and sleep diaries from the Nitra application for 2 weeks. Eligible participants continued using the Nitra application for 4 weeks during the intervention period. Interventions including sleep restriction, stimulus control, cognitive restructuring, relaxation techniques, and sleep hygiene education were implemented via the pre-programmed Nitra application. Post-intervention sleep characteristics were also obtained from questionnaires and sleep diaries from the Nitra application for another 1 week.

Results

A total of 40 participants completed the study. All participants had a baseline sleep efficiency of less than 85% with the majority of the participants having a sleep-onset insomnia problem (98%). For the primary outcome, sleep efficiency was significantly improved after using the Nitra application (p < 0.001). Self-reported total sleep time, sleep onset latency, wake after sleep onset, early morning awakening, ISI, Pittsburgh Sleep Quality Index (PSQI), and average subjective sleep quality were also significantly improved (p < 0.001 for all parameters except p = 0.017 for total sleep time and p = 0.018 for wake after sleep onset). Participants who had a low baseline ISI and went to bed and woke up within 30 minutes of a designated bedtime and wake-up time recommended by the Nitra application for ≥70% of all nights demonstrated an increased chance of achieving normal sleep efficiency after using the Nitra application.

Conclusion

This first internet-based CBT-i in Thailand, using the Nitra application, effectively improved sleep efficiency and other sleep parameters in chronic subthreshold to moderate insomnia.

Two decades of one health surveillance of Nipah virus in Thailand

BACKGROUND

Nipah virus (NiV) infection causes encephalitis and has > 75% mortality rate, making it a WHO priority pathogen due to its pandemic potential. There have been NiV outbreak(s) in Malaysia, India, Bangladesh, and southern Philippines. NiV naturally circulates among fruit bats of the genus Pteropus and has been detected widely across Southeast and South Asia. Both Malaysian and Bangladeshi NiV strains have been found in fruit bats in Thailand. This study summarizes 20 years of pre-emptive One Health surveillance of NiV in Thailand, including triangulated surveillance of bats, and humans and pigs in the vicinity of roosts inhabited by NiV-infected bats.

METHODS

Samples were collected periodically and tested for NiV from bats, pigs and healthy human volunteers from Wat Luang village, Chonburi province, home to the biggest P. lylei roosts in Thailand, and other provinces since 2001. Archived cerebrospinal fluid specimens from encephalitis patients between 2001 and 2012 were also tested for NiV. NiV RNA was detected using nested reverse transcription polymerase chain reaction (RT-PCR). NiV antibodies were detected using enzyme-linked immunosorbent assay or multiplex microsphere immunoassay.

RESULTS

NiV RNA (mainly Bangladesh strain) was detected every year in fruit bats by RT-PCR from 2002 to 2020. The whole genome sequence of NiV directly sequenced from bat urine in 2017 shared 99.17% identity to NiV from a Bangladeshi patient in 2004. No NiV-specific IgG antibodies or RNA have been found in healthy volunteers, encephalitis patients, or pigs to date. During the sample collection trips, 100 community members were trained on how to live safely with bats.

CONCLUSIONS

High identity shared between the NiV genome from Thai bats and the Bangladeshi patient highlights the outbreak potential of NiV in Thailand. Results from NiV cross-sectoral surveillance were conveyed to national authorities and villagers which led to preventive control measures, increased surveillance of pigs and humans in vicinity of known NiV-infected roosts, and increased vigilance and reduced risk behaviors at the community level. This proactive One Health approach to NiV surveillance is a success story; that increased collaboration between the human, animal, and wildlife sectors is imperative to staying ahead of a zoonotic disease outbreak.

Early detection of neutralizing antibodies against SARS-CoV-2 in COVID-19 patients in Thailand

BACKGROUND

The presence of neutralizing antibodies (NAbs) is an indicator of protective immunity for most viral infections. A newly developed surrogate viral neutralization assay (sVNT) offers the ability to detect total receptor binding domain-targeting NAbs in an isotype-independent manner, increasing the test sensitivity. Thus, specimens with low IgM/ IgG antibody levels showed strong neutralization activity in sVNT.

METHODS

This study aimed to measure the %inhibition of NAbs measured by sVNT in PCR-confirmed COVID-19 patients. The sensitivity of sVNT for the diagnosis of SARS-CoV-2 infection and its kinetics were determined.

RESULTS

Ninety-seven patients with PCR-confirmed SARS-CoV-2 infection were included in this study. Majority of the patients were 21-40 years old (67%) and 63% had mild symptoms. The sensitivity of sVNT for the diagnosis of SARS-CoV-2 infection was 99% (95% confidence interval (CI) 94.4-100%) and the specificity was 100% (95% CI 98.3-100%). The negative predictive value of sVNT from the samples collected before and after 7 days of symptom onset was 99.5% (95% CI 97.4-100%) and 100% (95% CI 93.8-100%), respectively. The level of inhibition at days 8-14 were significantly higher than days 0-7 (p<0.001). The median %inhibition values by severity of COVID-19 symptoms were 79.9% (interquartile range (IQR) 49.7-91.8%); 89.0% (IQR 71.2-92.4%); and 86.6% (IQR 69.5-92.8%), for mild, moderate and severe/critical symptoms respectively. The median level of sVNT %inhibition of severe was significantly higher than the mild group (p = 0.05).

CONCLUSION

The sVNT is a practical and robust serological test for SARS-CoV-2 infection and does not require specialized biosafety containment. It can be used clinically to aid diagnosis in both early and late infection especially in cases when the real-time RT-PCR results in weakly negative or weakly positive, and to determine the protective immune response from SARS-CoV-2 infection in patients.

SARS-CoV-2 RNA shedding in recovered COVID-19 cases and the presence of antibodies against SARS-CoV-2 in recovered COVID-19 cases and close contacts, Thailand, April-June 2020

Coronavirus disease 2019 (COVID-19) is caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Although Thailand has been fairly effective at controlling the spread of COVID-19, continued disease surveillance and information on antibody response in recovered patients and their close contacts remain necessary in the absence of approved vaccines and antivirals. Here, we examined 217 recovered COVID-19 patients to assess their viral RNA shedding and residual antibodies against SARS-CoV-2. We also evaluated antibodies in blood samples from 308 close contacts of recovered COVID-19 patients. We found that viral RNA remained detectable in 6.6% of recovered COVID-19 cases and up to 105 days. IgM, IgG, and IgA antibodies against SARS-CoV-2 were detected in 13.8%, 88.5%, and 83.4% of the recovered cases 4-12 weeks after disease onset, respectively. Higher levels of antibodies detected were associated with severe illness patients experienced while hospitalized. Fifteen of the 308 contacts (4.9%) of COVID-19 cases tested positive for IgG antibodies, suggesting probable exposure. Viral clearance and the pattern of antibody responses in infected individuals are both crucial for effectively combating SARS-CoV-2. Our study provides additional information on the natural history of this newly emerging disease related to both natural host defenses and antibody duration.