OBJECTIVES:To evaluate continuity of once-weekly isoniazid plus rifapentine for 12 weeks (3HP) tuberculosis preventive treatment in a Paraguayan prison cohort, focusing on adherence, strict completion, safety and institutional transitions as drivers of non-completion. METHODS:We conducted a longitudinal implementation cohort study at Coronel Oviedo Penitentiary Center, Paraguay, during July-October 2025. Persons with QuantiFERON-TB Gold Plus-confirmed tuberculosis infection and active tuberculosis excluded by the program algorithm were offered 3HP under directly observed therapy. Weekly dose administration, missed-dose reasons and self-reported adverse reactions were recorded. RESULTS:Of 1846 people incarcerated in the facility, 240 accessed QuantiFERON testing and 170 tested positive. Among QFT-positive individuals, 50 did not initiate 3HP because of transfer, release or refusal. A total of 120 initiated 3HP; 90 achieved acceptable adherence, defined as receipt of at least 11 of 12 doses (75.0%), and 82 completed all 12 doses (68.3%). Medication was administered in 1195 of 1440 scheduled person-weeks (83.0%). Non-completion occurred in 30 participants; 28 cases (93.3%) were attributable to release or transfer. Fifty-seven participants (47.5%) reported at least one adverse reaction, mostly mild or moderate. CONCLUSION:High adherence was achieved with 3HP, but institutional transitions interrupted completion, highlighting the need for prison-to-community continuity mechanisms after incarceration.
In Loreto, Peru, unsuccessful treatment outcomes among patients with drug-resistant tuberculosis (DR-TB) remain high, underscoring the need to identify associated factors in this Amazonian region. This retrospective cohort study included 417 DR-TB cases registered in the national TB system in Loreto between 2015 and 2023. Loss to follow-up, death, and treatment failure were combined into a single composite outcome variable defined as unsuccessful treatment outcome. Bivariate and multivariable logistic regression analyses were performed to estimate associations between sociodemographic, clinical, and programmatic factors and this outcome. Among the 417 cases analyzed, 49.5% had unsuccessful treatment outcomes: 34.1% due to loss to follow-up, 3.6% due to treatment failure, and 11.8% due to death. Loss to follow-up was the main driver of unsuccessful outcomes, suggesting important gaps in continuity of care and adherence support within the regional health system. In bivariate analyses, unsuccessful outcomes were associated with age 18-30 years (OR: 3.7; 95% CI: 1.6-8.5; p = 0.002), TB-HIV coinfection (OR: 2.4; 95% CI: 1.2-4.6; p = 0.012), previous treatment after loss to follow-up (OR: 2.9; 95% CI: 1.6-5.2; p < 0.001), high bacillary load (+++) (OR: 1.9; 95% CI: 1.0-3.5; p = 0.041), alcoholism (OR: 4.6; 95% CI: 1.5-14.1; p = 0.006), and drug addiction (OR: 7.0; 95% CI: 2.0-24.2; p = 0.002). In the multivariable analysis, alcoholism remained independently associated with unsuccessful outcomes (aOR: 4.0; 95% CI: 1.1-14.2; p = 0.029), while individualized treatment was associated with lower odds of unsuccessful outcomes (aOR: 0.4; 95% CI: 0.1-0.9; p = 0.023). Unsuccessful treatment outcomes among DR-TB patients in Loreto were mainly driven by loss to follow-up, highlighting persistent gaps in treatment continuity and adherence support. Programmatic priorities should include decentralized access to drug-susceptibility testing, community-based adherence strategies, and integrated screening and management of alcohol use within DR-TB care.
ABSTRACT Introduction Lateral flow assays (LFAs) are indispensable rapid diagnostic tools in healthcare, enabling point-of-care diagnosis critical for patient management and support disease burden assessment and surveillance when results are properly recorded. However, misinterpretation errors and unreported cases remain a concern. A quality-assured, affordable Ai-powered tool, supporting the decision-making during result interpretation could promote proper disease monitoring and epidemiological surveillance. Here, we describe the performance of a universal AI model to digitize and interpret results from multiple LFA types through a smartphone application, a step that could ultimately enable standardized and digitally reportable test outcomes. Methods The AI algorithm was evaluated in 17 LFA types, including both 2-band and 3-band tests for different diseases and manufacturers. The model was trained on a dataset of 22,576 images captured under diverse lighting conditions with different smartphone models and using a custom mobile application, TiraSpot (Spotlab, Madrid, Spain). To assess generalizability, a leave-one-out cross-validation was applied, wherein each LFA type was iteratively excluded from training and used for testing. Model performance was evaluated using bootstrapping on the inference dataset. Results In the assessment of the model’s ability to generalize to new LFA types not previously analyzed (not included during development), the model achieved an overall AUC of 94.3% for second band detection. This overall performance was enhanced to 99.3% (Sensitivity=98,6%; Specificity=98%) after training with 50 images of each LFA type, highlighting the benefit of additional data for specific LFA types. For the third band detection, where less training data was available, the system achieved an overall AUC of 83.9% for unseen LFAs, improving to 94.2% (Sensitivity=92.9%; Specificity=87,9%) after training with 50 images of each LFA type. Conclusion This system demonstrates the feasibility of an AI-powered universal digital reader for interpreting LFA results from diverse test types using smartphone-captured images. Its compatibility with standard smartphones makes it a universal tool, enabling reliable LFA interpretation across devices and settings. By standardizing test interpretation and digitizing results, this tool could support decision making in result interpretation, enhancing epidemiological surveillance, particularly in resource-limited settings. Its adaptability across various infections highlights its potential to improve diagnostic consistency and support disease management in diverse healthcare settings.
Background: Colombia introduced the PCV7 in 2010, PCV10 in 2012, and PCV13 in 2022. Evidence on long-term pneumococcal serotype dynamics and the potential impact of new PCVs in Colombia remains limited. Methods: We conducted a retrospective analysis using national surveillance data on IPD isolates from children and adults collected between 2005 and 2023. The study period included a pre-vaccine era (2005-2009) and four subsequent periods corresponding to the sequential introduction of PCV7, PCV10, and PCV13. Temporal trends in serotype distribution were assessed using chi(2) tests. Theoretical vaccine coverage was estimated for currently licensed PCVs (10, 13, 15, 20, and 21) and for new formulations (PCV24 and PCV25). Results: A total of 7824 IPD isolates were included. Following universal introduction of PCV10, most PCV10 serotypes declined across all age groups. In contrast, serotype 19A increased substantially over the study period, rising from 3% of all isolates in 2005-2009 to 33% in 2023. Non-PCV13 serotypes particularly 6C, 23A , and 15A increased from 22% in 2005-2009 to 46% in 2013-2018. Among children <2 years, PCV21, PCV24, and PCV25 provided approximately 81-87% theoretical coverage of serotypes detected from 2019 to 2023; among adults, coverage ranged from 74% to 78%. Serotype 19A became the predominant antimicrobial-resistant serotype. PCV20-21 and PCV24-PCV25provide substantial coverage of resistant strains. Conclusions: In Colombia, PCV10 introduction was followed by a marked decline in PCV10 serotypes and a substantial increase in serotype 19A and other non-PCV13 serotypes. Continued nationwide surveillance is essential to monitor serotype trends and guide policy decisions regarding the introduction of next-generation PCVs.
Decision making in orthodontic treatment, especially regarding irreversible tooth extractions, is a complex and controversial challenge due to its impact on facial aesthetics and long-term functional stability. Artificial Intelligence and Machine Learning are advancing dentistry, aiding in tooth extraction decisions. Here we propose a Deep Learning solution to the Dental Extraction decision problem. We compare Intraoral and Extraoral image from a Dataset, which is an important contribution of this paper, with 1720 images from 215 patients, a small number considering the typical Deep Learning data sets. We compare five well known pretrained Convolutional Neural Network applying several balancing and optimization techniques. We significantly improved the baseline, achieving an F1-Score of 89.52
