Real Sets Research Articles

Development of a virtual reality application for the Old Harran School

The objective of this project that is described in this paper was to introduce the little-known Harran School of the medieval Islamic period located in Southeast Türkiye to a broader audience using virtual reality technologies. 3D models of the medieval Inner Castle in the Harran archaeological site and one of its saloons were created using photogrammetric methods. In this main saloon, 3D models of artifacts found during excavation works on this site that were used in medieval and earlies times were placed in a museum-like style. Additional objects, illumination and paintings of great Islamic scholars were appended to create a more realistic setting in the saloon. All these 3D models were transferred to the game engine Unity to generate a virtual environment, in which the visitor can immerse. Currently, this virtual museum can be experienced by PC-based head-mounted devices. While different implementations of virtual museums are already available in Türkiye with this project, a museum deploying virtual reality technology has been realized for the first time.

Interventions to alleviate anxiety and pain during venipuncture in children with chronic gastrointestinal and/or liver disease: A single-center prospective observational study.

The goal of this longitudinal study was to reduce anxiety and pain in children with chronic conditions from the gastrointestinal tract during venipuncture. These children undergo regular venipuncture as part of their medical management and the procedure is often accompanied with anxiety and pain. In addition, children as well as their parents and health care professionals (HCPs) often suffer "compassionate pain" because of emotional interference. In a realistic clinical setting, different psychological and medical interventions were examined: (1) Psychoeducational brochures and (2) four different medical-technical interventions during venipuncture. In a large hospital in Germany, 169 children, their parents, and HCPs were asked to rate anxiety and pain during venipuncture before and after the intervention. Children showed a clear preference for some of the medical-technical interventions. Using Linear Mixed Models anxiety and pain rated by the children themselves showed no significant reduction. However, parents and HCPs reported a significant reduction. Age, gender, and status of liver transplantation were associated with a reduction in anxiety and pain in most of the analyses. Both psychoeducational brochures and medical-technical interventions had a positive impact on anxiety and pain. However, effectivity for the medical-technical interventions was lower than in previous studies utilizing individual interventions. Reasons for this difference as well as possibilities to improve the intervention are discussed. In addition, this study provides practical day-to-day information about the implementation of interventions for the work in pediatric units such as when and how to provide psychoeducational materials.

Similar social attention, physiological arousal, and familiarity effect in autistic and neurotypical children: A real-life recreational eye-tracking paradigm.

Social attention is reported to be crucial for the development of social skills, and, according to the social cognitive developmental theory, is fostered by social interactions. Autism is of central importance to the study of social attention, as autism is characterized by atypical social interactions and low social attention, both linked according to the social motivation theory to diminished social interest. Much evidence for positing low social interest in autism comes from eye-tracking studies, which, however, lack ecological validity. Our study documents social attention and physiological arousal, within close to real-life settings, in autistic children, as well as in their neurotypical peers, matched on gender and mental or chronological age. To explore the potential influence of partner familiarity or of the conversational topic, children gaze and electrodermal activity were recorded while they engaged in watercolor activities with, first a familiar and, next, an unfamiliar adult experimenter, who both introduced various topics. Autistic and neurotypical children exhibited comparable attention to their partners' eyes. Notably, across all groups, heightened visual attention was directed to familiar rather than unfamiliar partners. Moreover, parallel arousal patterns emerged, with all children displaying increased skin conductance responses during more engaging topics and when looking at their interactional partner's eyes. These findings underscore the task- and context-dependent nature of social attention and highlight the role of familiarity in an ecologically valid context. The absence of group differences challenges the universality of the social cognitive developmental theory and questions the scope of the social motivation theory of autism. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

A151 THRESHOLD TRANSFERABILITY STUDY FOR 3 QUANTITATIVE FITS IN AN ACTIVE COLORECTAL CANCER SCREENING PROGRAM (CRCSP)

Abstract Background Three quantitative fecal immunochemical tests (FITs) were evaluated in the context of a possible supplier transition of the Quebec CRCSP. FITs from different suppliers have different sensitivities, specificities and positivity rates at preset thresholds (PT) which impacts the number of colonoscopies generated. Our hypothesis was that the determination of an equivalent threshold (ET) by a positivity rate equivalent strategy (PRES) or with regression analysis would lead to similar clinical performances. Aims Determine if ETs can be obtained and what their impact is on the sensitivity and specificity of each supplier Methods Kits containing a collection tube for each supplier were distributed to 6600 patients. Sampling was done on a fresh stool by patients with specific instructions for each supplier. Patients with FIT results ≥35 ug/g of stool hemoglobin were considered positive and referred for a colonoscopy within the CRCSP and those between 20 and 35 ug/g were invited to consent to a colonoscopy. Colonoscopy results were considered positive if advanced neoplasia (AN) was found. ETs were determined by PRES and regression analysis comparing each pair of suppliers. Median sensitivity and specificities were computed for each supplier at current and ETs. Results Kits from 5513 patients were included in the study. Colonoscopy results were obtained for 262/342 FIT positive patients and 65/155 patients with results between 20 and 35 ug/g who consented to a colonoscopy. 69 patients were positive for AN at colonoscopy. ETs were obtained for the PRES but not for the regression method due to failed linearity tests. The ETs yielding the same positivity rate as Eiken’s 35.0 ug/g were determined to be 21.6 ug/g and 37.4 ug/g for Sentinel and Alfresa respectively. At the PT, the sensitivities of Alfresa (0.71, CI :0.58-0.80) and Eiken (0.71, CI:0.58-0.80) did not differ significantly but were higher than Sentinel (0.52, CI:0,39-0,64, p= 0.0059 and 0.016). The specificity for Sentinel (0.64, CI: 0.57-0.69) was significantly higher than Alfresa (0.45, CI: 0.38-0.50, p= 3.3E-6) and Eiken (0.54, CI: 0.47-0.59, p= 0.0059), as was the one from Eiken when compared to Alfresa (p =0.038). After adjustment using PRES, no significant differences for sensitivity (Alfresa (0.68, CI:0.55-0.77), Eiken (0.71, CI :0.58-0.80), Sentinel (0.52, CI:0.46-0.70)) and specificity (Alfresa (0.51 CI: 0.45-0.57), Eiken (0.54, CI: 0.47-0.59), Sentinel (0.57, CI: 0.50-0.62)) were observed between suppliers. Conclusions Our study demonstrates that a PRES can be used to adjust the threshold in a supplier transition in order to achieve similar clinical performances. To our knowledge this is the first study to evaluate this in a real life active CRCSP setting where the stool is sampled by the patient and non-homogenized. Funding Agencies Ministère de la Santé et des Services Sociaux du Québec and private funds (Somagen-Eiken, Quidel-Sentinel, Abbott-Alfresa)

Assessment framework for deepfake detection in real-world situations

Detecting digital face manipulation in images and video has attracted extensive attention due to the potential risk to public trust. To counteract the malicious usage of such techniques, deep learning-based deepfake detection methods have been employed and have exhibited remarkable performance. However, the performance of such detectors is often assessed on related benchmarks that hardly reflect real-world situations. For example, the impact of various image and video processing operations and typical workflow distortions on detection accuracy has not been systematically measured. In this paper, a more reliable assessment framework is proposed to evaluate the performance of learning-based deepfake detectors in more realistic settings. To the best of our acknowledgment, it is the first systematic assessment approach for deepfake detectors that not only reports the general performance under real-world conditions but also quantitatively measures their robustness toward different processing operations. To demonstrate the effectiveness and usage of the framework, extensive experiments and detailed analysis of four popular deepfake detection methods are further presented in this paper. In addition, a stochastic degradation-based data augmentation method driven by realistic processing operations is designed, which significantly improves the robustness of deepfake detectors.

A supply-chain perspective on producing and upscaling bioplastic from cultivated brown seaweed

Plastic pollution is an environmental emergency and finding sustainable alternatives to traditional plastics has become a pressing need. Seaweed-based bioplastic has emerged as a promising solution, as it is biodegradable and made from renewable biomass, while seaweed cultivation itself provides various environmental benefits. However, the feasibility of implementing a brown seaweed-based bioplastic supply chain in a realistic setting remains unclear, as previous research focused either on single processing steps or on virtual supply chains aggregating data from different studies. This study describes a case study for seaweed-based bioplastic within the PlastiSea research project: from seaweed cultivation to biomass processing and bioplastic and composite material development at the lab and pilot scale, thus providing insights into its feasibility. Adopting a multidisciplinary approach, the study employs multiple methods to characterize each stage in the supply chain and provides an overall life cycle assessment (LCA) as well as lessons learned throughout the process. The analysis showed potential for producing and utilizing multiple co-products from the same seaweed source, including biopolymer extracts with varying degrees of refinement for use in low-cost (bioplastic films) and high-cost (microfiber composites) applications. The use of residual biomass as a source of alginates for producing bioplastics offers a low-cost and sustainable biomass supply currently not competing with other markets. The LCA results indicate the potential for reducing the environmental impact of seaweed-based bioplastic production through upscaling and increasing process efficiency.

Feasible Peer-Mediated Intervention for Autistic Children Using Minimal Speech: A Qualitative Intervention Development Process.

Qualitative engagement with stakeholders in the development of interventions can provide insight into strategies to maximize feasibility in real-life settings. We engaged stakeholders (autistic adults, early childhood educators, early childhood sector leaders and policy influencers, parents of autistic children, and speech-language pathologists) to inform the development of an educator-led peer-mediated intervention (PMI) for autistic preschoolers who use minimal speech that is feasible to implement in inclusive early childhood education and care (ECEC) settings. A qualitative iterative intervention design process was utilized. Stakeholders (N = 15) attended an online workshop and completed a document review exploring the acceptability and feasibility of the proposed embedded PMI. A two-step analysis procedure using the Theoretical Domains Framework and template analysis was conducted to identify the barriers, enablers, and supports to the implementation of embedded PMI in early childhood settings. While embedded PMI was unanimously acceptable to stakeholders, several participants expressed concerns regarding feasibility. Barriers to the successful integration and implementation of PMI in inclusive preschool contexts included access to skills, knowledge, and resources. Participants identified strategies to overcome modifiable barriers and to enhance the existing enablers. These strategies are reflected in the following themes: build on the familiar, build capacity in augmentative and alternative communication, adopt a whole center approach, adapt to meet the needs of the ECEC setting, and engage in proactive implementation. To address barriers to the implementation of embedded PMI, action is needed at various levels: macro (national/policy), meso (organization/setting), and micro (individual). https://doi.org/10.23641/asha.25155770.

Older adults’ awareness, motivation, and behavior changes by wearable activity trackers before and during the COVID-19 pandemic

Background: Advanced sensor technology, such as commercial wearable activity trackers, is expected to be useful for improving public health in an aging society. However, its effects on psychological and behavioral changes in older adults have not been thoroughly investigated. Objective: This study aimed to clarify the effects of wearable activity trackers on psychological and behavioral changes in older adults before the pandemic and to confirm whether the effects were maintained during the pandemic. Method: Thirty Japanese older adults wearing Fitbit trackers for 12 weeks (i.e., six weeks before the pandemic and six weeks during the onset of the pandemic) in a real-life setting participated in this study. Participants’ perceptions of physical activity (PA) and monitoring were measured weekly using a questionnaire in addition to the pre-monitoring survey (baseline). Results: Significant positive effects of activity trackers were observed on participants’ awareness of the benefits of PA and motivation toward monitoring from the first week, and high levels were maintained until the end of the study, regardless of the pandemic. Accordingly, an approximately 3,000-steps/day increase was observed in the first week and maintained before the pandemic; although PA levels decreased during the onset of the pandemic, they remained higher than at baseline. The decrease in the number of steps and activity calories in COVID-19-non-affected participants was 2%. In the affected participants, an 11% decrease in steps and an 8% decrease in activity calories were observed, indicating that 3% was compensated by other activities. Conclusion: Positive effects of activity trackers were observed on older adults’ awareness of PA benefits. Thus, future studies are required to verify the effects of increased PA benefits awareness on older adults’ long-term activity levels for their health.

A Cell-Based Nasal Model for Screening the Deposition, Biocompatibility, and Transport of Aerosolized PLGA Nanoparticles.

The olfactory region of the nasal cavity directly links the brain to the external environment, presenting a potential direct route to the central nervous system (CNS). However, targeting drugs to the olfactory region is challenging and relies on a combination of drug formulation, delivery device, and administration technique to navigate human nasal anatomy. In addition, in vitro and in vivo models utilized to evaluate the performance of nasal formulations do not accurately reflect deposition and uptake in the human nasal cavity. The current study describes the development of a respirable poly(lactic-co-glycolic acid) nanoparticle (PLGA NP) formulation, delivered via a pressurized metered dose inhaler (pMDI), and a cell-containing three-dimensional (3D) human nasal cast model for deposition assessment of nasal formulations in the olfactory region. Fluorescent PLGA NPs (193 ± 3 nm by dynamic light scattering) were successfully formulated in an HFA134a-based pMDI and were collected intact following aerosolization. RPMI 2650 cells, widely employed as a nasal epithelial model, were grown at the air-liquid interface (ALI) for 14 days to develop a suitable barrier function prior to exposure to the aerosolized PLGA NPs in a glass deposition apparatus. Direct aerosol exposure was shown to have little effect on cell viability. Compared to an aqueous NP suspension, the transport rate of the aerosolized NPs across the RPMI 2650 barrier was higher at all time points indicating the potential advantages of delivery via aerosolization and the importance of employing ALI cellular models for testing respirable formulations. The PLGA NPs were then aerosolized into a 3D-printed human nasal cavity model with an insert of ALI RPMI 2650 cells positioned in the olfactory region. Cells remained highly viable, and there was significant deposition of the fluorescent NPs on the ALI cultures. This study is a proof of concept that pMDI delivery of NPs is a viable means of targeting the olfactory region for nose-to-brain drug delivery (NTBDD). The cell-based model allows not only maintenance under ALI culture conditions but also sampling from the basal chamber compartment; hence, this model could be adapted to assess drug deposition, uptake, and transport kinetics in parallel under real-life settings.

Dupilumab Efficacy on Asthma Functional, Inflammatory, and Patient-Reported Outcomes across Different Disease Phenotypes and Severity: A Real-Life Perspective.

Dupilumab is currently approved for the treatment of Type 2 severe asthma and chronic rhinosinusitis with nasal polyps (CRSwNP). Few studies have specifically reported on dupilumab efficacy on asthma outcomes as a primary objective in a real-life setting, in patients with and without CRSwNP. Our study aimed to explore the efficacy of dupilumab on functional, inflammatory, and patient-reported outcomes in asthma patients across different disease phenotypes and severity, including mild-to-moderate asthma coexisting with CRSwNP. Data from 3, 6, and 12 months follow-up were analyzed. Asthma (FEV1%, Tiffeneau%, ACT, FeNO, oral steroid use, exacerbation rate, and blood eosinophilia) and polyposis (SNOT22, VAS, NPS) outcomes showed a rapid (3 months) and sustained (6 and 12 months) significant change from baseline, despite most of the patients achieving oral steroid withdrawal. According to the sensitivity analysis, the improvement was not conditioned by either the presence of polyposis or severity of asthma at baseline. Of note, even in the case of milder asthma forms, a significant further improvement was recorded during dupilumab treatment course. Our report provides short-, medium-, and long-term follow-up data on asthma outcomes across different diseases phenotypes and severity, contributing to the real-world evidence related to dupilumab efficacy on upper and lower airways T2 inflammation.

A new bio imagery user-friendly tool for automatic morphometry measurement on muscle cell cultures and histological sections

TRUEFAD (TRUE Fiber Atrophy Distinction) is a bioimagery user-friendly tool developed to allow consistent and automatic measurement of myotube diameter in vitro, muscle fiber size and type using rodents and human muscle biopsies. This TRUEFAD package was set up to standardize and dynamize muscle research via easy-to-obtain images run on an open-source plugin for FIJI. We showed here both the robustness and the performance of our pipelines to correctly segment muscle cells and fibers. We evaluated our pipeline on real experiment image sets and showed consistent reliability across images and conditions. TRUEFAD development makes possible systematical and rapid screening of substances impacting muscle morphology for helping scientists focus on their hypothesis rather than image analysis.

In-label, off-label prescription, efficacy and tolerability of dalbavancin: report from a National Registry.

Although dalbavancin is currently approved for the treatment of ABSSIs, several studies suggest its efficacy and tolerance as long-term therapy for other off-label indications requiring prolonged intravenous antibiotic administration. We conducted a prospective nationwide study of dalbavancin use in real-life settings for both approved and off-label indications analysing for each case the clinical and microbiological characteristics of infection the efficacy and safety of treatments. During the study period (from December 2018 to July 2021), the ID specialists from 14 different centres enrolled 223 patients treated with dalbavancin [141 males (63%) and 82 females (37%); male/female ratio 1.72; mean age 59 (SD 17.2) years, (range 15-96). Most patients in the study population (136/223; 61.0%) came from community rather than health care facilities and most of them were visited in Infectious Diseases wards (93/223; 41.7%) and clinics (55/223; 24.7%) even though some patients were cured in other settings, such as surgery wards (18/223; 8.1%), orthopaedic wards (11/223; 4.9%), Emergency Rooms (7/223; 3.1%) and non-surgical other than ID wards (6/223; 2.7%). The most common ID diagnoses were osteomyelitis (44 cases/223; 19.7%; of which 29 acute and 15 chronic osteomyelitis), cellulitis (28/223; 12.5%), cutaneous abscess (23/223; 10.3%), orthopaedic prosthesis-associated infection (22/223; 9.9%), surgical site infection (20/223; 9.0%) and septic arthritis (15/223; 6.7%). In conclusion, by virtue of its PK/PD properties, dalbavancin represents a valuable option to daily in-hospital intravenous or outpatient antimicrobial regimens also for off-label indications requiring a long-term treatment of Gram-positive infections.

Predicting subgroup treatment effects for a new study: Motivations, results and learnings from running a data challenge in a pharmaceutical corporation.

We present the motivation, experience, and learnings from a data challenge conducted at a large pharmaceutical corporation on the topic of subgroup identification. The data challenge aimed at exploring approaches to subgroup identification for future clinical trials. To mimic a realistic setting, participants had access to 4 Phase III clinical trials to derive a subgroup and predict its treatment effect on a future study not accessible to challenge participants. A total of 30 teams registered for the challenge with around 100 participants, primarily from Biostatistics organization. We outline the motivation for running the challenge, the challenge rules, and logistics. Finally, we present the results of the challenge, the participant feedback as well as the learnings. We also present our view on the implications of the results on exploratory analyses related to treatment effect heterogeneity.

Efficacy of Lamivudine and Dolutegravir simplification therapy compared with triple therapy in Northeast Brazil (LAMDO Study)

Background: Modern antiretroviral therapy provides numerous effective and well-tolerated treatment options for individuals living with HIV. However, due to medication tolerability, toxicity, and cost optimization associated with the emergence of highly potent drugs, dual therapy has emerged as a new therapeutic alternative for patients with viral suppression. Observational studies worldwide are being conducted to assess the effectiveness of dual therapy in people living with HIV/AIDS. A real-world study is important to validate the findings obtained in controlled studies. Objective: Assess the effectiveness of dual therapy with lamivudine and dolutegravir compared to triple therapy in real-life settings. Methods: The study was conducted at São José Infectious Diseases Hospital, a tertiary referral hospital in the state of Ceará, northeast Brazil, for the treatment of PLWHA. Results: A total of 521 patients were taking double therapy with lamivudine plus dolutegravir and 450 patients were in triple therapy, mostly in the use of association with dolutegravir, were analyzed. Patients on dual therapy had a higher median age compared to those on triple therapy. A statistically significant higher viral suppression was observed in patients on dual therapy compared to triple therapy (p < 0,001). Viral suppression on dual therapy under 200 copies was 97.2%. There was a statistically significant higher percentage of patients with higher CD4/CD8 ratio using triple therapy compared to dual therapy. Conclusion: The current study suggests a higher effective response to dual therapy compared to triple therapy in PLWHA in the real-world, supporting therapy simplification as a sustainable option to maintain virological suppression in patients experiencing toxicity or comorbidities.

Integrative Medicine for Cancer-Related Pain: A Narrative Review

Cancer-related pain (C-RP) is a prevalent and debilitating concern among patients with cancer, with conventional treatments limited in their ability to provide adequate relief, and by the adverse effects associated with their use. Complementary and integrative medicine (CIM) modalities have been shown to be potentially effective and safe for the treatment of pain and related symptoms, when used in conjunction with conventional medications and under medical supervision. An increasing number of oncology centers provide CIM within their conventional supportive and palliative care service, in an “Integrative Oncology” (IO) setting. A large body of clinical research, including systematic reviews and guidelines such as those published in 2022 by the Society for Integrative Oncology (SIO), in collaboration with the American Society for Clinical Oncology (ASCO), support the use of some CIM modalities for C-RP and related concerns. These include acupuncture for general and peri-operative/procedural pain, as well as aromatase inhibitor-associated arthralgia (AIA); reflexology or acupressure for pain during systemic therapy for cancer; hypnosis for procedural pain or pain due to diagnostic workup; and massage for pain experienced by patients during palliative and hospice care. Further research is needed, within both randomized control trials and pragmatic non-controlled studies which are more reflective of the real-life IO setting. This review summarizes the evidence supporting the use of CIM for C-RP; the analgesic mechanism of the modalities presented; and the challenges facing IO researchers, as well as the implementation of the 2022 SIO-ASCO guideline recommendations.

The impacts of active and self-supervised learning on efficient annotation of single-cell expression data

A crucial step in the analysis of single-cell data is annotating cells to cell types and states. While a myriad of approaches has been proposed, manual labeling of cells to create training datasets remains tedious and time-consuming. In the field of machine learning, active and self-supervised learning methods have been proposed to improve the performance of a classifier while reducing both annotation time and label budget. However, the benefits of such strategies for single-cell annotation have yet to be evaluated in realistic settings. Here, we perform a comprehensive benchmarking of active and self-supervised labeling strategies across a range of single-cell technologies and cell type annotation algorithms. We quantify the benefits of active learning and self-supervised strategies in the presence of cell type imbalance and variable similarity. We introduce adaptive reweighting, a heuristic procedure tailored to single-cell data—including a marker-aware version—that shows competitive performance with existing approaches. In addition, we demonstrate that having prior knowledge of cell type markers improves annotation accuracy. Finally, we summarize our findings into a set of recommendations for those implementing cell type annotation procedures or platforms. An R package implementing the heuristic approaches introduced in this work may be found at https://github.com/camlab-bioml/leader.

Effectiveness and Safety of Direct-acting Antivirals for Treatment of Adolescents With HCV/HIV Coinfection: Real-world Data From Europe.

We evaluated the effectiveness and safety of direct-acting antivirals in adolescents with hepatitis C (HCV)/HIV coinfection using pooled individual patient-level data from 5 European cohorts. Of 122 participants in follow-up from November 2013 to August 2021, 19 were treated <18 years of age; of 15 with HCV RNA available at/after 12 weeks post-treatment, all had sustained virologic response with acceptable safety. This evidence addresses an important gap in knowledge of treatment outcomes in adolescents with HCV/HIV coinfection in real-life settings.

Ventilator-Associated Pneumonia Prediction Models Based on AI: Scoping Review.

Ventilator-associated pneumonia (VAP) is a serious complication of mechanical ventilation therapy that affects patients' treatments and prognoses. Owing to its excellent data mining capabilities, artificial intelligence (AI) has been increasingly used to predict VAP. This paper reviews VAP prediction models that are based on AI, providing a reference for the early identification of high-risk groups in future clinical practice. A scoping review was conducted in accordance with the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews) guidelines. The Wanfang database, the Chinese Biomedical Literature Database, Cochrane Library, Web of Science, PubMed, MEDLINE, and Embase were searched to identify relevant articles. Study selection and data extraction were independently conducted by 2 reviewers. The data extracted from the included studies were synthesized narratively. Of the 137 publications retrieved, 11 were included in this scoping review. The included studies reported the use of AI for predicting VAP. All 11 studies predicted VAP occurrence, and studies on VAP prognosis were excluded. Further, these studies used text data, and none of them involved imaging data. Public databases were the primary sources of data for model building (studies: 6/11, 55%), and 5 studies had sample sizes of <1000. Machine learning was the primary algorithm for studying the VAP prediction models. However, deep learning and large language models were not used to construct VAP prediction models. The random forest model was the most commonly used model (studies: 5/11, 45%). All studies only performed internal validations, and none of them addressed how to implement and apply the final model in real-life clinical settings. This review presents an overview of studies that used AI to predict and diagnose VAP. AI models have better predictive performance than traditional methods and are expected to provide indispensable tools for VAP risk prediction in the future. However, the current research is in the model construction and validation stage, and the implementation of and guidance for clinical VAP prediction require further research.

Multitask Deep Learning for Joint Detection of Necrotizing Viral and Noninfectious Retinitis From Common Blood and Serology Test Data.

Necrotizing viral retinitis is a serious eye infection that requires immediate treatment to prevent permanent vision loss. Uncertain clinical suspicion can result in delayed diagnosis, inappropriate administration of corticosteroids, or repeated intraocular sampling. To quickly and accurately distinguish between viral and noninfectious retinitis, we aimed to develop deep learning (DL) models solely using noninvasive blood test data. This cross-sectional study trained DL models using common blood and serology test data from 3080 patients (noninfectious uveitis of the posterior segment [NIU-PS] = 2858, acute retinal necrosis [ARN] = 66, cytomegalovirus [CMV], retinitis = 156). Following the development of separate base DL models for ARN and CMV retinitis, multitask learning (MTL) was employed to enable simultaneous discrimination. Advanced MTL models incorporating adversarial training were used to enhance DL feature extraction from the small, imbalanced data. We evaluated model performance, disease-specific important features, and the causal relationship between DL features and detection results. The presented models all achieved excellent detection performances, with the adversarial MTL model achieving the highest receiver operating characteristic curves (0.932 for ARN and 0.982 for CMV retinitis). Significant features for ARN detection included varicella-zoster virus (VZV) immunoglobulin M (IgM), herpes simplex virus immunoglobulin G, and neutrophil count, while for CMV retinitis, they encompassed VZV IgM, CMV IgM, and lymphocyte count. The adversarial MTL model exhibited substantial changes in detection outcomes when the key features were contaminated, indicating stronger causality between DL features and detection results. The adversarial MTL model, using blood test data, may serve as a reliable adjunct for the expedited diagnosis of ARN, CMV retinitis, and NIU-PS simultaneously in real clinical settings.

Pulsed field ablation for atrial fibrillation in real-life settings: Efficacy, safety, and lesion durability in patients with recurrences

Pulsed field ablation for atrial fibrillation in real-life settings: Efficacy, safety, and lesion durability in patients with recurrences