Robust Collaborative Visual-Inertial SLAM for Mobile Augmented Reality.

Achieving precise real-time localization and ensuring robustness are critical challenges in multi-user mobile AR applications. Leveraging collaborative information to augment tracking accuracy on lightweight devices and fortify overall system robustness emerges as a crucial necessity. In this paper, we propose a robust centralized collaborative multi-agent VI-SLAM system for mobile AR interaction and server-side efficient consistent mapping. The system deploys a lightweight VIO frontend on mobile devices for real-time tracking, and a backend running on a remote server to update multiple submaps. When overlapping areas between submaps across agents are detected, the system performs submap fusion to establish a globally consistent map. Additionally, we propose a map registration and fusion strategy based on covisibility areas for online registration and fusion in multi-agent scenarios. To improve the tracking accuracy of the frontend on agent, we introduce a strategy for updating the global map to the local map at a moderate frequency between the camera-rate pose estimation of the frontend VIO and the low-frequency global map optimization, using a tightly coupled strategy to achieve consistency of the multi-agent frontend poses estimation in the global map. The effectiveness of the proposed method is further confirmed by executing backend mapping on the server and deploying VIO frontends on multiple mobile devices for AR demostration. Additionally, we discuss the scalability of the proposed system by analyzing network traffic, synchronization frequency, and other factors at both the agent and server ends.

Migration of naphthalene in a biochar-amended bioretention facility based on HYDRUS-1D analysis.

Biochar has been proved as a promising and efficient filler in bioretention facilities for enhancing the stormwater pollutants removal. However, the migration behaviors of stormwater pollutants in biochar filled bioretention facilities is unclear. In this study, as one of the most typical stormwater pollutants, naphthalene was selected as an example and a HYDRUS-1D model was first used to understand the migration behavior of naphthalene in a bioretention facility. In comparison with the conventional bioretention soil media (sandy loam), the amended biochar filled bioretention cell showed that the naphthalene removal rate was enhanced by up to 10.1%. Meanwhile, the experimental data was well-fitted by the "two-site sorption model" in HYDRUS-1D model. Another, the effect of rainfall intensity on the naphthalene migration in both bioretention columns was further investigated. The HYDRUS-1D model fitting indicated that the increase in rainfall intensity promoted naphthalene migration by increasing hydraulic conductivity and water flux. In addition, static batch experiments revealed that the biochar filled fillers achieved about 50% higher adsorption capacity than sandy loam. The sensitivity analysis from the HYDRUS-1D model data verified adsorption coefficient Kd and longitudinal dispersivity λ are the main factors affecting naphthalene migration. Finally, the model simulation displays that the proportion of naphthalene retained by the fillers was highest during high rainfall intensities, indicating that the fillers remain the most important fate for naphthalene. This study presents research on the behavior and mechanisms of stormwater pollutant transport through improved bioretention facilities.

The mode detangler.

Astrophotonics aims to transfer photonic technology to the development of compact astronomical instruments. However, light coupling from a multimode fiber, typically adopted in modern observatories, to a single-mode photonic device still poses a challenge. Though a photonic lantern can enable this transition in a low-loss way, it requires that the number of single-mode fibers (SMFs) at the output is the same as the number of guided modes in the multimode fiber, resulting in a cumbersome fan-out of many single-mode devices to be connected. Herein, we invent an active device in a waveguide form called "the mode detangler" (MD). We show that it can adaptively transform a complex light field from a multimode fiber to a single-mode-like spot. In this way, only one single-mode device is required at the end. The path leading to the idea and the theory behind the mode detangling effect is explained, followed by numerical simulations and experimental demonstrations using a few-mode fiber as proof of concept. We believe this device has the potential to address the multimode-to-single-mode conversion challenge in astrophotonics but also sheds light on (de)multiplexing applications regarding spatial mode technology in optical communications.

Endoplasmic reticulum-mitochondrial encounter structure regulates the mitochondrial morphology, DON biosynthesis and toxisome formation in Fusarium graminearum.

The endoplasmic reticulum-mitochondrial encounter structure (ERMES) complex is known to play crucial roles in various cellular processes. However, its functional significance in filamentous fungi, particularly its impact on deoxynivalenol (DON) biosynthesis in Fusarium graminearum, remains inadequately understood. In this study, we aimed to investigate the regulatory function of the ERMES complex in F. graminearum. Our findings indicate significant changes in mitochondrial morphology of ERMES mutants, accompanied by decreased ATP content and ergosterol production. Notably, the toxisome formation in the ERMES mutant ΔFgMDM10 was defective, resulting in a substantial reduction in DON biosynthesis. This suggests a pivotal role of ERMES in toxisome formation, as evidenced by the pronounced inhibition of toxisome formation when ERMES was disrupted by boscalid. Furthermore, ERMES deficiencies were shown to diminish the virulence of F. graminearum towards host plants significantly. In conclusion, our results suggest ERMES is an important regulator of mitochondrial morphology, DON biosynthesis, and toxisome formation in F. graminearum.

Network model with internal complexity bridges artificial intelligence and neuroscience.

Artificial intelligence (AI) researchers currently believe that the main approach to building more general model problems is the big AI model, where existing neural networks are becoming deeper, larger and wider. We term this the big model with external complexity approach. In this work we argue that there is another approach called small model with internal complexity, which can be used to find a suitable path of incorporating rich properties into neurons to construct larger and more efficient AI models. We uncover that one has to increase the scale of the network externally to stimulate the same dynamical properties. To illustrate this, we build a Hodgkin-Huxley (HH) network with rich internal complexity, where each neuron is an HH model, and prove that the dynamical properties and performance of the HH network can be equivalent to a bigger leaky integrate-and-fire (LIF) network, where each neuron is a LIF neuron with simple internal complexity.

Design and performance analysis of low damage anti-skid crescent drills for bone drilling.

BACKGROUND: With orthopedic surgery increasing year on year, the main challenges in bone drilling are thermal damage, mechanical damage, and drill skid. The need for new orthopedic drills that improve the quality of surgery is becoming more and more urgent. METHODS: Here, we report the skidding mechanism of drills at a wide range of inclination angle and propose two crescent drills (CDTI and CDTII). The anti-skid performance and drilling damage of the crescent drills were analyzed for the first time. Inclined bone drilling experiments were carried out with crescent drills and twist drills and real-time drilling forces and temperatures were collected. RESULTS: The crescent drills are significantly better than the twist drill in terms of anti-skid, reducing skidding forces, thrust forces and temperature. The highest temperature is generated close to the upper surface of the workpiece rather than at the hole exit. Finally, the longer crescent edge with a small and negative polar angle increases the rake angle of the cutting edge and reduces thrust forces but increases skidding force and temperature. This study can promote the development of high-quality orthopedic surgery and the development of new bone drilling tools. CONCLUSION: The crescent drills did not skid and caused little drilling damage. In comparison, the CDTI performs better in reducing the skidding force, while the CDTII performs better in reducing the thrust force.

Assessing the importance of risk factors for diabetic retinopathy in patients with type 2 diabetes mellitus: Results from the classification and regression tree models.

BACKGROUND: Diabetic retinopathy (DR) is one of the serious complications of diabetes mellitus (DM). Many studies have identified the risk factors associated with DR, but there is not much evidence on the importance of these factors for DR. This study aimed to investigate the associated factors for patients with type 2 DM (T2DM) and calculate the importance of the identified factors. MATERIALS AND METHODS: Using probability proportionate to size sampling method in this community-based cross-sectional study, 22 community health service centers were selected from 10 administrative districts in Shenzhen, China. Approximately 60 T2DM patients were recruited from each center. The participants completed a structural questionnaire, had their venous blood collected, and underwent medical examinations and fundus photography. Logistic regression models were used to identify the risk factors of DR. The classification and regression tree (CART) model was used to calculate the importance of the identified risk factors. RESULTS: This study recruited 1097 T2DM patients, 266 of whom were identified as having DR, yielding a prevalence rate of 24.3% (95% confidence interval [CI]: 21.7%-26.9%). Results showed that a longer duration of DM, indoor-type lifestyle, and higher levels of hemoglobin A1c (HbA1c) or urea increased the risk of DR. Patients with HbA1c values ≥7% were about 2.45 times (odds ratio: 2.45; 95% CI: 1.83-3.29) more likely to have DR than their counterparts. The CART model found that the values of variable importance for HbA1c, DM duration, lifestyle (i.e., indoor type), and urea were 48%, 37%, 10%, and 4%, respectively. CONCLUSION: The prevalence of DR is high for T2DM patients who receive DM health management services from the primary healthcare system. HbA1c is the most important risk factor for DR. Integration of DR screening and HbA1c testing into the healthcare services for T2DM to reduce vision impairment and blindness is urgently warranted.

Deciphering the therapeutic potential of Myeloid-Specific JAK2 inhibition in acute respiratory distress syndrome.

Acute respiratory distress syndrome (ARDS) is a life-threatening condition characterized by severe inflammation and pulmonary dysfunction. Despite advancements in critical care, effective pharmacological interventions for ARDS remain elusive. While Janus kinase 2 (JAK2) inhibitors have emerged as an innovative treatment for numerous autoinflammatory diseases, their therapeutic potential in ARDS remains unexplored. In this study, we investigated the contribution of JAK2 and its underlying mechanisms in ARDS utilizing myeloid-specific JAK2 knockout murine models alongside a pharmacological JAK2 inhibitor. Notably, myeloid-specific JAK2 knockout led to a notable attenuation of ARDS induced by intratracheal administration of LPS, accompanied by reduced levels of neutrophils and inflammatory cytokines in bronchoalveolar lavage fluid (BALF) and lung tissue. Intriguingly, the ameliorative effects were abolished upon the depletion of monocyte-derived alveolar macrophages (Mo-AMs) rather than tissue-resident alveolar macrophages (TR-AMs). JAK2 deficiency markedly reversed LPS-induced activation of STAT5 in macrophages. Remarkably, pharmacological JAK2 inhibition using baricitinib failed to substantially alleviate neutrophils infiltration, implying that specific inhibition of JAK2 in Mo-AMs is imperative for ARDS amelioration. Collectively, our data suggest that JAK2 may mitigate ARDS progression through the JAK2 pathway in Mo-AMs, underscoring JAK2 in alveolar macrophages, particularly Mo-AMs, as a promising therapeutic target for ARDS treatment.

Single-crystalline metal-oxide dielectrics for top-gate 2D transistors.

Two-dimensional (2D) structures composed of atomically thin materials with high carrier mobility have been studied as candidates for future transistors1-4. However, owing to the unavailability of suitable high-quality dielectrics, 2D field-effect transistors (FETs) cannot attain the full theoretical potential and advantages despite their superior physical and electrical properties3,5,6. Here we demonstrate the fabrication of atomically thin single-crystalline Al2O3 (c-Al2O3) as a high-quality top-gate dielectric in 2D FETs. By using intercalative oxidation techniques, a stable, stoichiometric and atomically thin c-Al2O3 layer with a thickness of 1.25 nm is formed on the single-crystalline Al surface at room temperature. Owing to the favourable crystalline structure and well-defined interfaces, the gate leakage current, interface state density and dielectric strength of c-Al2O3 meet the International Roadmap for Devices and Systems requirements3,5,7. Through a one-step transfer process consisting of the source, drain, dielectric materials and gate, we achieve top-gate MoS2 FETs characterized by a steep subthreshold swing of 61 mV dec-1, high on/off current ratio of 108 and very small hysteresis of 10 mV. This technique and material demonstrate the possibility of producing high-quality single-crystalline oxides suitable for integration into fully scalable advanced 2D FETs, including negative capacitance transistors and spin transistors.

Alleviation of volatile fatty acids inhibition in anaerobic digestion of swine manure with nano-bubble water supplementation.

Nano-bubble water (NBW) was applied to anaerobic digestion (AD) to alleviate volatile fatty acids (VFAs) inhibition, improve the buffering capacity and CH4 production in this work. Results indicated that NBW accelerated the consumption of VFAs and prevented inhibition due to VFAs accumulation. Additionally, NBW facilitated a rapid increase in partial alkalinity (PA) and total alkalinity (TA) as well as a corresponding rapid decrease in intermediate alkalinity (IA)/PA and VFA/TA, thereby improving buffering capacity and alleviating VFAs inhibition. Moreover, CH4 production improved by more than 12.2% by NBW. Similarly, the activities of the extracellular hydrolases and coenzyme F420 increased. Besides, NBW increased the abundance of microbial community and strengthened the metabolic pathways of hydrogenotrophic methanogens, which could be the intrinsic mechanism by which NBW alleviated VFAs inhibition, improved system stability, and increased CH4 production. This study demonstrates that NBW supplementation can be an effective method for mitigating frequent VFAs inhibition.

Pd Nanoparticle-Modified BiOBr/CdS S-Scheme Photocatalyst for Enhanced Conversion of CO2.

S-scheme heterojunction photocatalyst-coupled plasma-resonance effect can enhance the response range and absorption of light and charge transfer, and, at the same time, obtain strong redox ability, which is an effective way to improve CO2 conversion. In this work, plasma S-scheme heterojunctions of Pd/BiOBr/CdS with heterogeneous interfaces have been successfully constructed by a simple hydrothermal method. The possible reaction mechanism was proposed by in situ infrared, ultraviolet-visible spectroscopy (UV-vis), electron paramagnetic resonance (ESR), density functional theory (DFT), and electrochemical techniques. It was proved that the plasma S-scheme heterojunction can enhance the charge separation efficiency and improve the photocatalytic activity. When the loading ratio is Pd0.6-10%-BiOBr/CdS, it has the best performance, and the CO yield is 30.24 µmol/g, which is 15 and 30 times that of pure BiOBr and CdS, respectively. The results show that with the strong absorption of photon energy and the special electron transfer mode of S-scheme heterojunction, the charge can be effectively separated and transferred, and the photocatalytic activity is significantly improved. This study provides a useful strategy for charge transfer kinetics of plasma S-scheme heterojunction photocatalysts.

An Improved Fire and Smoke Detection Method Based on YOLOv8n for Smart Factories.

Factories play a crucial role in economic and social development. However, fire disasters in factories greatly threaten both human lives and properties. Previous studies about fire detection using deep learning mostly focused on wildfire detection and ignored the fires that happened in factories. In addition, lots of studies focus on fire detection, while smoke, the important derivative of a fire disaster, is not detected by such algorithms. To better help smart factories monitor fire disasters, this paper proposes an improved fire and smoke detection method based on YOLOv8n. To ensure the quality of the algorithm and training process, a self-made dataset including more than 5000 images and their corresponding labels is created. Then, nine advanced algorithms are selected and tested on the dataset. YOLOv8n exhibits the best detection results in terms of accuracy and detection speed. ConNeXtV2 is then inserted into the backbone to enhance inter-channel feature competition. RepBlock and SimConv are selected to replace the original Conv and improve computational ability and memory bandwidth. For the loss function, CIoU is replaced by MPDIoU to ensure an efficient and accurate bounding box. Ablation tests show that our improved algorithm achieves better performance in all four metrics reflecting accuracy: precision, recall, F1, and mAP@50. Compared with the original model, whose four metrics are approximately 90%, the modified algorithm achieves above 95%. mAP@50 in particular reaches 95.6%, exhibiting an improvement of approximately 4.5%. Although complexity improves, the requirements of real-time fire and smoke monitoring are satisfied.

One-Step Spraying Strategy for Fabricating Bioinspired, Graphene-Based, and Multifunctional-Integrated Coatings on Structural Steel with Good Water Repellency, Fireproofing, Anticorrosion, and Durability.

Traditionally, many coatings were merely concentrated on settling the inherent fire protection problem of steel structures, while surface contamination and corrosion susceptibility should also be considered. Concurrently addressing these problems in fireproof efficiency and surface multifunctionality has become an issue of great significance in further expanding the application value in industrial and daily scenarios. Based on this condition, ecofriendly, graphene-based, and superhydrophobic coatings with multifunctional integration were constructed on steel via a one-step spraying method. The as-prepared coatings mainly consist of epoxy resin (EP), silicone resin (SR), a cyclodextrin-based flame retardant (MCDPM), expandable graphite (EG), and multilayered graphene (MG). The results demonstrate that the water contact angle (WCA) and water sliding angle (WSA) of as-prepared coatings can reach 156.8 ± 1.6 and 5.8 ± 0.7°, respectively, revealing good water repellency and self-cleaning properties. The coatings can also exhibit adequate adaptability for various substrates including wood, polyurethane foam, and cotton fabrics. Besides, good durability and robustness of coatings have been also verified via acid/alkali immersion, outdoor exposure, O2/plasma etching, and linear abrasion tests. Simultaneously, the coatings can exhibit excellent anticorrosion capacity for steel materials via a double barrier effect. Most importantly, the coatings have exhibited the lowest backside temperature (234.5 °C) during fire impact tests, suggesting excellent fireproof and heat insulation performance. This fact can be ascribed to the conjunct action between the physical/chemical charring process of flame retardants and the remarkable thermal stability of graphene. Consequently, this article can be expected to further promote the development and application of multifunctional-integrated coatings for steel structures in more fields.

The effect of family physician team processes on chronic disease management performance from a structural network perspective.

OBJECTIVE: The family physician team has become the core carrier for delivery primary health care in China. This study aimed to measure the effect of the network structural characteristics of family physician team processes on health performance. Strategic recommendations for optimizing the family physician team processes with a view to improving performance were presented. METHODS: A cross-sectional survey was conducted from October to December 2021 in Qianjiang in Hubei Province and Changsha in Hunan Province. Task performance, contextual performance, social networks, and sociodemographic characteristics were collected. Social network analysis was conducted to calculate density and centralization, then hierarchical linear regression analysis was employed to explore the relationship between the network structural characteristics of family physician team processes and performance. RESULTS: In total, 88 family physician teams attended in this investigation. The transition processes of family physician team showed a distinctive low density (0.272 ± 0.112), high centralization (0.866 ± 0.197) network structure. For family physician team, the density of action processes significantly and positively affected task performance (B = 0.600, P < 0.05); the centralization of action processes positively affected task performance (B = 0.604, P < 0.01); the density of action processes positively affected contextual performance (B = 0.545, P < 0.01); the density of interpersonal processes significantly and positively affected contextual performance (B = 0.326, P < 0.05). CONCLUSION: The network density and centralization of family physician team processes have positive effects on chronic disease management performance. The results from this study help to enhance our conceptual understanding of social network and its implications for team-dynamics. Optimizing family physician team processes is an effective way to strengthen the construction of family physician team and promote the quality and efficiency of family physician-contracted service. It is recommended to strengthen the management of team processes, enhance the internal collaboration mechanism, and optimize the centralized network structure of family physician team.

The phytoplankton community affects the energy metabolism and immunomodulation strategy of oyster during breeding seasons.

The mass mortality of Pacific oyster Crassostrea gigas has become a severe ecological and economic concern to Chinese aquaculture, which is proposed to be linked to the phytoplankton community in the farming waters. In the present study, both field and laboratory experiments were conducted to identify the phytoplankton taxa associated with oyster mortality and explore the molecular mechanism by which they affect the physiological health of oysters. The field experiment showed that more serious mortality of oysters was observed in the North Yellow Sea from July to September in 2018 (average survival rate of 75.11 %) than in 2019 (average survival rate of 85.78 %), with the proportion of Bacillariophyta (diatoms) in the phytoplankton community in 2018 lower than that in 2019. In comparison to 2019, reduced dry weight, lower glycogen and triglyceride contents in hepatopancreas, lower 17ß-estradiol and testosterone concentrations in gonad, as well as a generally weaker immune response against Vibrio splendidus stimulation were detected in the oysters sampled in 2018. The treatment of oysters with either starvation (starvation group) or Nitzschia closterium f. minutissima feeding (N. closterium group) was conducted to verify the field findings, with individuals reared in natural seawater as control. After 40 days of N. closterium feeding, dry weight, glycogen and triglyceride contents in hepatopancreas significantly increased, as well as the biosynthesis of sex hormones and gonadal maturation were promoted compared to the control and starvation groups. Moreover, a much stronger immune response against V. splendidus stimulation was observed in the oysters of N. closterium group, with the fold-changes of norepinephrine content in serum, SOD activity in hepatopancreas, and the mRNA expression level of IL17-5 and HSP70 in haemocytes higher than those in the control and starvation groups. Collectively, these results suggested that lack of diatoms in the farming waters suppressed the energy storage and gonadal maturation of adult oysters, and also resulted in a compromised immune response against bacterial infection, which may be a leading cause of the mass mortality of oysters living in diatom-deficient waters during breeding seasons.

Association of handgrip strength with health care utilisation among older adults: A longitudinal study in China.

Background: Evidence on the association between handgrip strength (HGS) and health care utilisation among Chinese older adults is scarce. In this study, we aimed to investigate the association of HGS with health care utilisation and to identify whether these associations varied by gender. Methods: The analytic sample of this prospective cohort study included 6007 Chinese older adults (≥60 years) from 2011 to 2018 waves of the China Health and Retirement Longitudinal Study. A handgrip dynamometer was used to measure HGS. We measured health care utilisation by outpatient visits, inpatient visits, and unmet hospitalisation needs. We used covariates-adjusted general estimating equations for the analyses. Results: Longitudinal results showed that participants with weakness increased the likelihood of outpatient visits (odds ratio (OR) = 1.13; 95% confidence interval (CI) = 1.01-1.27), inpatient visits (OR = 1.51; 95% CI = 1.32-1.73), and unmet hospitalisation needs (OR = 1.44; 95% CI = 1.19-1.79) than their counterparts. Participants with weakness increased the number of outpatient visits (incidence rate ratio (IRR) = 1.29; 95% CI = 1.11-1.51) and the number of inpatient visits (IRR = 1.39; 95% CI = 1.10-1.61). Participants with HGS asymmetry increased the likelihood of unmet hospitalisation needs (OR = 1.19; 95% CI = 1.03-1.43) than their counterparts. The results of the impact of every one-kilogramme (kg) increase in HGS on health care utilisation indicated consistent results. The associations were similarly observed irrespective of gender. Conclusions: Chinese older adults with weakness or HGS asymmetry used more health care. Interventions for improving muscle strength and correcting strength asymmetry are highly recommended, with the potential to considerably save households and health care systems.

Overview and Evolution of Insect Fibroin Heavy Chain (FibH).

The FibH gene, crucial for silk spinning in insects, encodes a protein that significantly influences silk fiber mechanics. Due to its large size and repetitive sequences, limited known sequences of insect FibH impede comprehensive understanding. Here, we analyzed 114 complete FibH gene sequences from Lepidoptera (71 moths, 24 butterflies) and 13 Trichoptera, revealing single-copy FibH in most species, with 2-3 copies in Hesperinae and Heteropterinae (subfamily of skippers). All FibH genes are structured with two exons and one intron (39-45 bp), with the second exon being notably longer. Moths exhibit higher GC content in FibH compared to butterflies and Trichoptera. The FibH composition varies among species, with moths and butterflies favoring Ala, Gly, Ser, Pro, Gln, and Asn, while Trichoptera FibH is enriched in Gly, Ser, and Arg, and has less Ala. Unique to Trichoptera FibH are Tyr, Val, Arg, and Trp, whereas Lepidoptera FibH is marked by polyAla (polyalanine), polySer (polyserine), and the hexapeptide GAGSGA. A phylogenetic analysis suggests that Lepidoptera FibH evolved from Trichoptera, with skipper FibH evolving from Papilionoidea. This study substantially expands the FibH repertoire, providing a foundation for the development of artificial silk.

Hepatic encephalopathy post-TIPS: Current status and prospects in predictive assessment.

Transjugular intrahepatic portosystemic shunt (TIPS) is an essential procedure for the treatment of portal hypertension but can result in hepatic encephalopathy (HE), a serious complication that worsens patient outcomes. Investigating predictors of HE after TIPS is essential to improve prognosis. This review analyzes risk factors and compares predictive models, weighing traditional scores such as Child-Pugh, Model for End-Stage Liver Disease (MELD), and albumin-bilirubin (ALBI) against emerging artificial intelligence (AI) techniques. While traditional scores provide initial insights into HE risk, they have limitations in dealing with clinical complexity. Advances in machine learning (ML), particularly when integrated with imaging and clinical data, offer refined assessments. These innovations suggest the potential for AI to significantly improve the prediction of post-TIPS HE. The study provides clinicians with a comprehensive overview of current prediction methods, while advocating for the integration of AI to increase the accuracy of post-TIPS HE assessments. By harnessing the power of AI, clinicians can better manage the risks associated with TIPS and tailor interventions to individual patient needs. Future research should therefore prioritize the development of advanced AI frameworks that can assimilate diverse data streams to support clinical decision-making. The goal is not only to more accurately predict HE, but also to improve overall patient care and quality of life.

Synthetic DNA barcodes identify singlets in scRNA-seq datasets and evaluate doublet algorithms.

Single-cell RNA sequencing (scRNA-seq) datasets contain true single cells, or singlets, in addition to cells that coalesce during the protocol, or doublets. Identifying singlets with high fidelity in scRNA-seq is necessary to avoid false negative and false positive discoveries. Although several methodologies have been proposed, they are typically tested on highly heterogeneous datasets and lack a priori knowledge of true singlets. Here, we leveraged datasets with synthetically introduced DNA barcodes for a hitherto unexplored application: to extract ground-truth singlets. We demonstrated the feasibility of our framework, "singletCode," to evaluate existing doublet detection methods across a range of contexts. We also leveraged our ground-truth singlets to train a proof-of-concept machine learning classifier, which outperformed other doublet detection algorithms. Our integrative framework can identify ground-truth singlets and enable robust doublet detection in non-barcoded datasets.