Abbott realtime MTB assay for detecting Mycobacterium tuberculosis complex in respiratory specimens: a cost-benefit analysis.

PURPOSE: Molecular screening for Mycobacterium tuberculosis (MTB) can lead to rapid empirical treatment inception and reduce hospitalization time and complementary diagnostic tests. However, in low-prevalence settings, the cost-benefit balance remains controversial due to the high cost. METHODS: We used a Markov model to perform an economic analysis to evaluate the profit after implementing molecular MTB screening (Period B) compared with conventional culture testing (Period A) in respiratory samples from 7,452 consecutive subjects with presumed tuberculosis (TB). RESULTS: The proportion of positivity was comparable between both periods (P > 0.05), with a total of 2.16 and 1.78 samples/patient requested in periods A and B, respectively (P < 0.001). The mean length of hospital stay was 8.66 days (95%CI: 7.63-9.70) in Period B and 11.51 days (95%CI: 10.15-12.87) in Period A (P = 0.001). The healthcare costs associated with diagnosing patients with presumed TB were reduced by €717.95 per patient with PCR screening. The probability of remaining hospitalized and the need for a greater number of outpatient specialty care visits were the variables with the most weight in the model. CONCLUSION: Employing PCR as an MTB screening method in a low-prevalence setting may increase the profits to the system.

Genotypic study of Chlamydia trachomatis for lymphogranuloma venereum diagnosis in rectal specimens from men who have sex with men: a cost-effectiveness analysis.

PURPOSE: The significant proportion of asymptomatic patients and the scarcity of genotypic analysis of lymphogranuloma venereum (LGV), mainly among men who have sex with men (MSM), triggers a high incidence of underdiagnosed patients, highlighting the importance of determining the most appropriate strategy for LGV diagnosis, at both clinical and economical levels. MATERIALS AND METHODS: We conducted L1-L3 serovar detection by molecular biology in stored Chlamydia trachomatis-positive samples from MSM patients with HIV, another STI or belonging to a Pre-exposure prophylaxis program, to make a cost effectiveness study of four diagnostic strategies with a clinical, molecular, or mixed approach. RESULTS: A total of 85 exudates were analyzed: 35urethral (31 symptomatic/4 positive) and 50 rectal (22 symptomatic/25 positive), 70/85 belonging to MSM with associated risk factors. The average cost per patient was €77.09 and €159.55 for clinical (Strategy I) and molecular (Strategy IV) strategies respectively. For molecular diagnosis by genotyping of all rectal exudate samples previously positive for CT (Strategy II), the cost was €123.84. For molecular diagnosis by genotyping of rectal and/or urethral exudate samples from all symptomatic patients (proctitis or urethritis) with a previous positive result for CT (Strategy III), the cost was €129.39. The effectiveness ratios were 0.80, 0.95, 0.91, and 1.00 for each strategy respectively. The smallest ICER was €311.67 for Strategy II compared to Strategy I. CONCLUSIONS: With 30% asymptomatic patients, the most cost-effective strategy was based on genotyping all rectal exudates. With less restrictive selection criteria, thus increasing the number of patients with negative results, the most sensitive strategies tend to be the most cost-effective, but with a high incremental cost-effectiveness ratio.

LFighter: Defending against the label-flipping attack in federated learning.

Federated learning (FL) provides autonomy and privacy by design to participating peers, who cooperatively build a machine learning (ML) model while keeping their private data in their devices. However, that same autonomy opens the door for malicious peers to poison the model by conducting either untargeted or targeted poisoning attacks. The label-flipping (LF) attack is a targeted poisoning attack where the attackers poison their training data by flipping the labels of some examples from one class (i.e., the source class) to another (i.e., the target class). Unfortunately, this attack is easy to perform and hard to detect, and it negatively impacts the performance of the global model. Existing defenses against LF are limited by assumptions on the distribution of the peers' data and/or do not perform well with high-dimensional models. In this paper, we deeply investigate the LF attack behavior. We find that the contradicting objectives of attackers and honest peers on the source class examples are reflected on the parameter gradients corresponding to the neurons of the source and target classes in the output layer. This makes those gradients good discriminative features for the attack detection. Accordingly, we propose LFighter, a novel defense against the LF attack that first dynamically extracts those gradients from the peers' local updates and then clusters the extracted gradients, analyzes the resulting clusters, and filters out potential bad updates before model aggregation. Extensive empirical analysis on three data sets shows the effectiveness of the proposed defense regardless of the data distribution or model dimensionality. Also, LFighter outperforms several state-of-the-art defenses by offering lower test error, higher overall accuracy, higher source class accuracy, lower attack success rate, and higher stability of the source class accuracy. Our code and data are available for reproducibility purposes at https://github.com/NajeebJebreel/LFighter.

Validation of the anxiety subscale of the Spanish version of the Edinburgh Postnatal Depression Scale (EPDS-A).

Introduction: There is a need to implement routine perinatal mental health screening in Spain. Therefore, it is necessary to systematise the detection of depressive and anxious symptoms in pregnancy and postpartum using the same instrument. The Edinburgh Postnatal Stress Depression Scale (EPDS) is frequently used as a rapid, effective and cross-culturally validated screening tool for perinatal depression. In several countries, an Anxiety subscale, the EPDS-A, was identified within the EPDS. Although the factorial structure of the EPDS has been investigated in Spanish population, the EPDS-A has not yet been validated. This study aimed to validate the EPDS-A as a measure of perinatal anxiety in Spanish population. Methods: 161 women were evaluated with the EPDS and the State-Trait Anxiety Inventory (STAI) during pregnancy and postpartum. Confirmatory factor analysis (CFA) was used to confirm the trifactorial structure of the EPDS, comprising the dimensions of Depression, Anhedonia and Anxiety. Likewise, the invariance of the trifactorial model between pregnancy and postpartum was tested. Finally, the correlations between the EPDS-A and the STAI subscales (State Anxiety and Trait Anxiety) were calculated. Results: The Exploratory factor analysis (EFA) driven three-factor structure of the EPDS, consisting of an Anhedonia factor (Items 1, 2, and 10), an Anxiety factor (Items 3, 4, 5, and 6) and a Depression factor (Items 7, 8, and 9), was the best measurement model for the current data compared to the alternative model tested [χ2 = 34.592, df = 32, p = 0.34; χ2/df = 1.08; RMSEA = 0.023, 90% Confidence Interval [CI] [0.000, 0.064], CFI = 0.996, GFI = 0.960]. The model's invariance between pregnant and postpartum women was confirmed. The existence of an Anxiety subscale within the EPDS was also confirmed. The scores obtained with the EPDS-A correlated moderately with scores on both subscales of the STAI during pregnancy and after delivery. Using the STAI as a criterion and prioritising the instrument's sensitivity, a cut-off point of 4 points was established for the EPDS-A. Conclusion: Our results confirm the trifactorial structure of the EPDS in Spanish population. The Anxiety subscale was validated for routine perinatal mental health screening.

Interplay between Optical Emission and Magnetism in the van der Waals Magnetic Semiconductor CrSBr in the Two-Dimensional Limit.

The van der Waals semiconductor metamagnet CrSBr offers an ideal platform for studying the interplay between optical and magnetic properties in the two-dimensional limit. Here, we carried out an exhaustive optical characterization of this material by means of temperature- and magnetic-field-dependent photoluminescence (PL) on flakes of different thicknesses down to the monolayer. We found a characteristic emission peak that is quenched upon switching the ferromagnetic layers from an antiparallel to a parallel configuration and exhibits a temperature dependence different from that of the peaks commonly ascribed to excitons. The contribution of this peak to the PL is boosted around 30-40 K, coinciding with the hidden order magnetic transition temperature. Our findings reveal the connection between the optical and magnetic properties via the ionization of magnetic donor vacancies. This behavior enables a useful tool for the optical reading of the magnetic states in atomically thin layers of CrSBr and shows the potential of the design of 2D heterostructures with magnetic and excitonic properties.

Hexagonal Hybrid Bismuthene by Molecular Interface Engineering.

High-quality devices based on layered heterostructures are typically built from materials obtained by complex solid-state physical approaches or laborious mechanical exfoliation and transfer. Meanwhile, wet-chemically synthesized materials commonly suffer from surface residuals and intrinsic defects. Here, we synthesize using an unprecedented colloidal photocatalyzed, one-pot redox reaction a few-layers bismuth hybrid of "electronic grade" structural quality. Intriguingly, the material presents a sulfur-alkyl-functionalized reconstructed surface that prevents it from oxidation and leads to a tuned electronic structure that results from the altered arrangement of the surface. The metallic behavior of the hybrid is supported by ab initio predictions and room temperature transport measurements of individual nanoflakes. Our findings indicate how surface reconstructions in two-dimensional (2D) systems can promote unexpected properties that can pave the way to new functionalities and devices. Moreover, this scalable synthetic process opens new avenues for applications in plasmonics or electronic (and spintronic) device fabrication. Beyond electronics, this 2D hybrid material may be of interest in organic catalysis, biomedicine, or energy storage and conversion.

Fair detection of poisoning attacks in federated learning on non-i.i.d. data.

Reconciling machine learning with individual privacy is one of the main motivations behind federated learning (FL), a decentralized machine learning technique that aggregates partial models trained by clients on their own private data to obtain a global deep learning model. Even if FL provides stronger privacy guarantees to the participating clients than centralized learning collecting the clients' data in a central server, FL is vulnerable to some attacks whereby malicious clients submit bad updates in order to prevent the model from converging or, more subtly, to introduce artificial bias in the classification (poisoning). Poisoning detection techniques compute statistics on the updates to identify malicious clients. A downside of anti-poisoning techniques is that they might lead to discriminate minority groups whose data are significantly and legitimately different from those of the majority of clients. This would not only be unfair, but would yield poorer models that would fail to capture the knowledge in the training data, especially when data are not independent and identically distributed (non-i.i.d.). In this work, we strive to strike a balance between fighting poisoning and accommodating diversity to help learning fairer and less discriminatory federated learning models. In this way, we forestall the exclusion of diverse clients while still ensuring detection of poisoning attacks. Empirical work on three data sets shows that employing our approach to tell legitimate from malicious updates produces models that are more accurate than those obtained with state-of-the-art poisoning detection techniques. Additionally, we explore the impact of our proposal on the performance of models on non-i.i.d local training data.

The Biofire® Filmarray® Pneumonia Plus panel for management of lower respiratory tract infection in mechanically-ventilated patients in the COVID-19 era: a diagnostic and cost-benefit evaluation.

We assessed the diagnostic performance of the Biofire® Filmarray® Pneumonia Plus panel (FA-PP) compared to standard culture in Intensive Care Unit patients with suspected ventilator-associated lower respiratory tract infection in the COVID-19 era. We determined whether its implementation in routine diagnostic algorithms would be cost-beneficial from a hospital perspective. Of 163 specimens, 96 (59%) returned negative results with FA-PP and conventional culture, and 29 specimens (17.8%) were positive with both diagnostic methods and yielded concordant qualitative bacterial identification/isolation. Thirty-nine specimens (23.9%) gave discordant results (positive via FA-PP and negative via culture). Real-life adjustments of empirical antimicrobial therapy (EAT) after FA-PP results resulted in additional costs beyond EAT alone of 1868.7 €. Adequate EAT adjustments upon FA-PP results would have resulted in a saving of 6675.8 €. In conclusion, the data presented supports the potential utility of FA-PP for early EAT adjustment in patients with ventilator-associated lower respiratory tract infection.

Enhanced Security and Privacy via Fragmented Federated Learning.

In federated learning (FL), a set of participants share updates computed on their local data with an aggregator server that combines updates into a global model. However, reconciling accuracy with privacy and security is a challenge to FL. On the one hand, good updates sent by honest participants may reveal their private local information, whereas poisoned updates sent by malicious participants may compromise the model's availability and/or integrity. On the other hand, enhancing privacy via update distortion damages accuracy, whereas doing so via update aggregation damages security because it does not allow the server to filter out individual poisoned updates. To tackle the accuracy-privacy-security conflict, we propose fragmented FL (FFL), in which participants randomly exchange and mix fragments of their updates before sending them to the server. To achieve privacy, we design a lightweight protocol that allows participants to privately exchange and mix encrypted fragments of their updates so that the server can neither obtain individual updates nor link them to their originators. To achieve security, we design a reputation-based defense tailored for FFL that builds trust in participants and their mixed updates based on the quality of the fragments they exchange and the mixed updates they send. Since the exchanged fragments' parameters keep their original coordinates and attackers can be neutralized, the server can correctly reconstruct a global model from the received mixed updates without accuracy loss. Experiments on four real data sets show that FFL can prevent semi-honest servers from mounting privacy attacks, can effectively counter-poisoning attacks, and can keep the accuracy of the global model.

BD MAX Enteric Bacterial, Bacterial Plus, and Virus Panels for Diagnosis of Acute Infectious Gastroenteritis: a Cost-Benefit Analysis.

Economic assessment is required to gauge the value of implementing PCR syndromic platforms in the microbiology laboratory for the diagnosis of community-acquired acute gastroenteritis (AGE) in pediatric and adult in- and outpatients. A cost-benefit analysis was conducted from a health care system perspective using BD MAX Enteric Bacterial, Bacterial Plus, and Virus panels. Two 6-month periods were selected, in which either conventional procedures (in 2017) or BD MAX PCR multiplex panels (in 2018) were used. We retrospectively reviewed medical records of all patients with positive results and a representative sample of negative ones. A Markov model was used to represent transition probabilities between different health care states from time of stool microbiological study until completion of AGE-episode-associated health care. A total of 1,336 medical records were reviewed (829 in 2018 and 507 in 2017), showing overall a significantly higher positivity rate in 2018 than in 2017 (26% versus 6%, P < 0.001). The total cost per individual associated with health care for AGE was €314 in 2018 and €341 in 2017; when we only considered the pediatric cohort, the figures were €271 and €456, respectively. Using Tornado sensitivity analyses, we found that the three variables that most influenced the model in descending order of weight were the probability of longer hospital stays, the probability of returning to the emergency room (ER), and the probability of hospitalization from the ER. Use of BD MAX enteric PCR platforms for the diagnosis of community-acquired AGE instead of a non-PCR-based conventional approach results in an incremental benefit from a health care perspective in the general population, particularly children. IMPORTANCE The implementation of multiplex molecular panels allows microbiological laboratories to quickly, sensitively, and accurately diagnose acute infectious gastroenteritis. This methodology therefore allows faster decisions regarding treatment and infection control measures. Economic evaluations are required to gauge the value of implementing these syndromic PCR platforms in a community-based acute gastroenteritis setting. We studied the potential clinical and cost benefits, in terms of both their impact on laboratory costs and the subsequent costs of managing patients.

"CAPA in Progress": A New Real-Life Approach for the Management of Critically Ill COVID-19 Patients.

(1) Background: COVID-19-associated pulmonary aspergillosis (CAPA) has worsened the prognosis of patients with pneumonia and acute respiratory distress syndrome admitted to the intensive care unit (ICU). The lack of specific diagnosis criteria is an obstacle to the timely initiation of appropriate antifungal therapy. Tracheal aspirate (TA) has been employed under special pandemic conditions. Galactomannan (GM) antigens are released during active fungal growth. (2) Methods: We proposed the term "CAPA in progress" (CAPA-IP) for diagnosis at an earlier stage by GM testing on TA in a specific population admitted to ICU presenting with clinical deterioration. A GM threshold ≥0.5 was set as the mycological inclusion criterion. This was followed by a pre-emptive short-course antifungal. (3) Results: We prospectively enrolled 200 ICU patients with COVID-19. Of these, 164 patients (82%) initially required invasive mechanical ventilation and GM was tested in TA in 93 patients. A subset of 19 patients (11.5%) fulfilled the CAPA-IP criteria at a median of 9 days after ICU admittance. The median GM value was 3.25 ± 2.82. CAPA-IP cases showed significantly higher ICU mortality [52.6% (10/19) vs. 34.5% (50/145), p = 0.036], as well as a much longer median ICU stay than those with a normal GM index [27 (7-64) vs. 11 (9-81) days, p = 0.008]. All cases were treated with a pre-emptive systemic antifungal for a median time of 19 (3-39) days. (4) Conclusions: CAPA-IP highlights a new real-life early approach in the field of fungal stewardship in ICU programs.

Photoluminescence Enhancement by Band Alignment Engineering in MoS2/FePS3 van der Waals Heterostructures.

Single-layer semiconducting transition metal dichalcogenides (2H-TMDs) display robust excitonic photoluminescence emission, which can be improved by controlled changes to the environment and the chemical potential of the material. However, a drastic emission quench has been generally observed when TMDs are stacked in van der Waals heterostructures, which often favor the nonradiative recombination of photocarriers. Herein, we achieve an enhancement of the photoluminescence of single-layer MoS2 on top of van der Waals FePS3. The optimal energy band alignment of this heterostructure preserves light emission of MoS2 against nonradiative interlayer recombination processes and favors the charge transfer from MoS2, an n-type semiconductor, to FePS3, a p-type narrow-gap semiconductor. The strong depletion of carriers in the MoS2 layer is evidenced by a dramatic increase in the spectral weight of neutral excitons, which is strongly modulated by the thickness of the FePS3 underneath, leading to the increase of photoluminescence intensity. The present results demonstrate the potential for the rational design of van der Waals heterostructures with advanced optoelectronic properties.

Spin-crossover nanoparticles anchored on MoS2 layers for heterostructures with tunable strain driven by thermal or light-induced spin switching.

In the past few years, the effect of strain on the optical and electronic properties of MoS2 layers has attracted particular attention as it can improve the performance of optoelectronic and spintronic devices. Although several approaches have been explored, strain is typically externally applied on the two-dimensional material. In this work, we describe the preparation of a reversible 'self-strainable' system in which the strain is generated at the molecular level by one component of a MoS2-based composite material. Spin-crossover nanoparticles were covalently grafted onto functionalized layers of semiconducting MoS2 to form a hybrid heterostructure. Their ability to switch between two spin states on applying an external stimulus (light irradiation or temperature change) serves to generate strain over the MoS2 layer. A volume change accompanies this spin crossover, and the created strain induces a substantial and reversible change of the electrical and optical properties of the heterostructure.

Behavioural Improvements in Children with Autism Spectrum Disorder after Participation in an Adapted Judo Programme Followed by Deleterious Effects during the COVID-19 Lockdown.

The public health lockdown prompted by the novel coronavirus (COVID-19) pandemic, which included school closures that may have potentially serious consequences for people with disabilities or special educational needs, disrupted an ongoing adapted judo training intervention in children with Autism Spectrum Disorder (ASD). The purpose of this study was to compare repetitive behaviours, social interaction, social communication, emotional responses, cognitive style and maladaptive speech scores across four time-points: baseline, after an eight-week control period, after an eight-week judo intervention and after an eight-week lockdown period due to COVID-19. The sample consisted of 11 children diagnosed with ASD according to the criteria of the Diagnostic and Statistical Manual of Mental Disorders-Fifth Edition (DSM-V), with an intelligence quotient (IQ) range between 60 and 70. Significant improvements were shown following the judo intervention period compared to the baseline and control periods. However, the same values significantly declined during the COVID-19 lockdown period resulting in values lower than those recorded at baseline, and following the control period and the judo intervention. The decline in psychosocial and behavioural scores are likely due to the stress caused by the sudden halt in activity and the increase in sedentary practices associated with the lockdown.

Give more data, awareness and control to individual citizens, and they will help COVID-19 containment.

The rapid dynamics of COVID-19 calls for quick and effective tracking of virus transmission chains and early detection of outbreaks, especially in the "phase 2" of the pandemic, when lockdown and other restriction measures are progressively withdrawn, in order to avoid or minimize contagion resurgence. For this purpose, contact-tracing apps are being proposed for large scale adoption by many countries. A centralized approach, where data sensed by the app are all sent to a nation-wide server, raises concerns about citizens' privacy and needlessly strong digital surveillance, thus alerting us to the need to minimize personal data collection and avoiding location tracking. We advocate the conceptual advantage of a decentralized approach, where both contact and location data are collected exclusively in individual citizens' "personal data stores", to be shared separately and selectively (e.g., with a backend system, but possibly also with other citizens), voluntarily, only when the citizen has tested positive for COVID-19, and with a privacy preserving level of granularity. This approach better protects the personal sphere of citizens and affords multiple benefits: it allows for detailed information gathering for infected people in a privacy-preserving fashion; and, in turn this enables both contact tracing, and, the early detection of outbreak hotspots on more finely-granulated geographic scale. The decentralized approach is also scalable to large populations, in that only the data of positive patients need be handled at a central level. Our recommendation is two-fold. First to extend existing decentralized architectures with a light touch, in order to manage the collection of location data locally on the device, and allow the user to share spatio-temporal aggregates-if and when they want and for specific aims-with health authorities, for instance. Second, we favour a longer-term pursuit of realizing a Personal Data Store vision, giving users the opportunity to contribute to collective good in the measure they want, enhancing self-awareness, and cultivating collective efforts for rebuilding society.

Field evaluation of a rapid antigen test (Panbio™ COVID-19 Ag Rapid Test Device) for COVID-19 diagnosis in primary healthcare centres.

OBJECTIVES: To our knowledge no previous study has assessed the performance of a rapid antigen diagnostic immunoassay (RAD) conducted at the point of care (POC). We evaluated the Panbio™ COVID-19 Ag Rapid Test Device for diagnosis of coronavirus 2019 disease (COVID-19) in symptomatic patients (n = 412) attending primary healthcare centres. METHODS: RAD was performed immediately after sampling following the manufacturer's instructions (reading at 15 min). RT-PCRs were carried out within 24 h of specimen collection. Samples displaying discordant results were processed for culture in Vero E6 cells. Presence of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) in cell cultures was confirmed by RT-PCR. RESULTS: Out of 412 patients, 43 (10.4%) tested positive by RT-PCR and RAD, and 358 (86.9%) tested negative by both methods; discordant results (RT-PCR+/RAD-) were obtained in 11 patients (2.7%). Overall specificity and sensitivity of rapid antigen detection (RAD) was 100% (95%CI 98.7-100%) and 79.6% (95%CI 67.0-88.8%), respectively, taking RT-PCR as the reference. Overall RAD negative predictive value for an estimated prevalence of 5% and 10% was 99% (95%CI 97.4-99.6%) and 97.9% (95%CI 95.9-98.9), respectively. SARS-CoV-2 could not be cultured from specimens yielding RT-PCR+/RAD- results (n = 11). CONCLUSION: The Panbio™ COVID-19 Ag Rapid Test Device performed well as a POC test for early diagnosis of COVID-19 in primary healthcare centres. More crucially, the data suggested that patients with RT-PCR-proven COVID-19 testing negative by RAD are unlikely to be infectious.

Qualitative assessment of SARS-CoV-2-specific antibody avidity by lateral flow immunochromatographic IgG/IgM antibody assay.

Knowledge of the precise timing of SARS-CoV-2 infection may be of clinical and epidemiological relevance. The presence of low-avidity IgGs has conventionally been considered an indicator of recent infection. Here, we carried out qualitative assessment of SARS-CoV-2-specific antibody avidity using an urea (6M) dissociation test performed on a lateral flow immunochromatographic IgG/IgM device. We included a total of 76 serum specimens collected from 57 COVID-19 patients, of which 39 tested positive for both IgG and IgM and 37 only for IgG. Sera losing IgG reactivity after urea treatment (n = 28) were drawn significantly earlier (P = .04) after onset of symptoms than those which preserved it (n = 48). This assay may be helpful to estimate the time of acquisition of infection in patients with mild to severe COVID-19.