Abstract Purpose To develop a pan-fungal lateral flow device (LFD); evaluating device performance with samples obtained from ex vivo porcine cornea infection models. Methods Fungal β-glucan (PF1), human CLEC7A/Dectin-1/CLECSF12 protein (Fc Tag; PF2), and fungal melanin (PF3) antibodies were evaluated for binding to clinically relevant fungal and bacterial species ( Aspergillus flavus, Fusarium keratoplasticum, Candida albicans, Pseudomonas aeruginosa, Staphylococcus aureus ) by immunofluorescence staining. PF1 and PF2 were evaluated in proof-of-concept, in-house LFD strips using cultured pathogens and ex vivo porcine corneal infection samples. The lead antibody (PF1) was validated in a commercially-developed prototype LFD. Results PF1 and PF2 discriminated target fungi from bacteria by immunofluorescence microscopy and in-house LFD strips. The lead candidate PF1 demonstrated good sensitivity (0.75) and specificity (0.94) with cultured fungal hyphae. Samples obtained from infected ex vivo cornea by clinically relevant methods confirmed excellent sensitivity (scrapes: 1.00, swabs: 0.94) and specificity (scrapes: 1.00, swabs: 0.83). The commercially-developed PF1-LFD prototype achieved perfect sensitivity (1.00) and specificity (1.00) when detecting and discriminating fungi from non-fungal ex vivo corneal swab samples. Conclusions Feasibility of pan-fungal LFD application in microbial keratitis diagnosis was demonstrated – using β-glucan as a pan-fungal target and a clinically relevant microbial keratitis ex vivo model. LFDs were able to differentiate fungal from bacterial samples, detect antigen present in corneal swabs, and provide a read-out within 20 minutes. Sensitivity and specificity values are comparable to currently used diagnostic tests. Translational Relevance Early discrimination of fungal keratitis cases is important to adapt treatment, and improve patient outcome.

Microbial keratitis (MK) is a major global cause of blindness. Yet, treatment is heavily dependent on antimicrobials with limited options for immunomodulators - despite the critical role of dysregulated immune responses in disease pathogenesis. This gap reflects a critical unmet clinical need and is compounded by the lack of model systems capable of real-time high-resolution immune dynamics analysis. To address this, we developed a zebrafish larvae MK model utilising transgenic zebrafish lines with fluorescently labelled neutrophils, macrophages and basal epithelial cells. Corneal injury triggered rapid immune cell recruitment which was amplified by exposure to pro-inflammatory mediators such as N-formylmethionine-leucyl-phenylalanine (fMLF) and leukotriene B4 (LTB4). Infection with live bacteria induced robust, sustained neutrophil and macrophage recruitment, marked by increased neutrophil speed and migratory distance. This model enables dynamic in vivo visualization of immune cell dynamics, offering a powerful and scalable platform to accelerate the discovery and screening of novel immunomodulators for MK.

The rise of multi-drug resistant bacteria constitutes a global challenge with direct impact on millions of patients worldwide. The identification of bacterial species is essential to avoid unnecessary use of antibiotics and to minimize the emergence of additional resistance; however, there are few chemical strategies to identify resistant species in a rapid and unbiased manner. Cross-reactive arrays combine multiple sensor components to produce distinctive fingerprints for similar targets; herein, we present a cross-reactive sensing array built with long lifetime organic fluorophores. To the best of our knowledge, we synthesize the largest collection to date of bioconjugatable triangulenium fluorophores by including heteroatom and side-chain diversification for spectral and lifetime diversity, as well as water-soluble and orthogonal moieties for peptide tagging and compatibility with biological samples. After optimization, we evolve a four-fluorophore sensor array that correctly assigns all seven ESKAPEE bacterial species by combining their optical signatures. Lifetime chemical sensor arrays will open avenues for concentration-independent, reliable and sensitive optical detection without the need for prior knowledge of molecular targets.

Pneumonia, a respiratory disease often caused by bacterial infection in the distal lung, necessitates prompt and precise diagnosis, particularly in critical care settings. Optical endomicroscopy (OEM) facilitates real-time acquisition of in vivo and in situ optical biopsies, thus expediting bacterial detection. Nonetheless, visually analyzing the vast number of images generated by the OEM in real time can be challenging, potentially impeding timely intervention. In this regard, to rapidly segment and detect the bacteria, we propose EmiNet, a novel dual-stream network that integrates the capabilities of Transformer and Convolutional Neural Networks (CNN) within an encoder-decoder architecture that simultaneously captures local-global appearance and motion features. Within EmiNet, we introduce a multi-modal cross-channel attention module that facilitates the integration of motion features with appearance features. Furthermore, to compensate for the lack of annotated training data, we developed a synthetic dataset by simulating bacterial motion and integrating these models onto real backgrounds devoid of bacteria. The authenticity of this dataset was confirmed through a Visual Turing Test, where medical experts assessed a mixture of synthetic and real bacterial images. The results indicate that the synthetic images are almost indistinguishable from the real ones. EmiNet's performance is evaluated on both real and synthetic datasets. Experiments show that EmiNet surpasses state-of-the-art segmentation models and leads to a 6.8% improvement in detection correlation over the state-of-the-art bacteria detection algorithms.

To measure aetiology and antibiotic resistance (AMR) trends of bacteria cultured from corneal scrapings from patients with infectious keratitis at a tertiary referral hospital in South India. In this retrospective study, bacterial aetiology and antimicrobial resistance profiles were identified from the microbiology records of patients undergoing microbial keratitis diagnosis at the Aravind Eye Hospital, Madurai, India from 2013-2024. Statistical analyses were performed by Spearman’s rank correlation coefficient to identify significant trends. P. aeruginosa (n=1047) was the most frequently isolated bacteria, followed by S. pneumoniae (n=987). There were significant increases in the number of P. aeruginosa (rs: 0.66; P=0.0219) and S. aureus isolates cultured (r s : 0.70; P=0.0130). S. aureus demonstrated increasing resistance to cefazolin (rs: 0.76; P=0.0015), gatifloxacin (rs: 0.75; P=0.0071), levofloxacin (r s : 0.60; P=0.0442), moxifloxacin (r s : 0.59; P=0.0437) and chloramphenicol (r s : 0.78; P=0.0049) over time. S. pneumoniae resistance towards tetracycline significantly increased (r s : 0.80; P=0.0029). P. aeruginosa isolates remained largely susceptible to all antibiotics screened, with significant decreasing resistance rates to ceftazidime (r s : −0.71; P=0.019), amikacin (r s : −0.59; P=0.0489), gentamicin (r s : −0.66; P=0.0219) and tobramycin (r s : −0.69; P=0.017) identified. No significant trends in resistance patterns were identified for Nocardia spp .. Bacterial aetiology and antibiotic resistance rates shifted over time for key pathogens causing keratitis. Understanding these in the local context is important. For instance, while others in India have reported increasing P. aeruginosa AMR, we did not find this to be true in our patient population. This may have implications for local prescribing guidelines. Microbial keratitis remains a highly prevalent, sight-limiting condition in India. Treatment options are limited, and are confounded by emergent antibiotic resistance (AMR) of causative pathogens. This study demonstrates shifting bacterial aetiology, and antimicrobial resistance patterns. Our data on P. aeruginosa resistance patterns contradict other reporting within India, where our resistance rates remained low. Fluoroquinolones are the most common first-line treatment for bacterial keratitis, yet over 80% of our S. aureus isolates demonstrated resistance towards fluoroquinolones in 2024. Understanding local and temporal bacterial aetiology and resistance rates are imperative for designing local guidelines for patient treatment practices.

Fungal keratitis (FK) is a severe eye infection mainly caused by Aspergillus flavus and Fusarium solani. We examined the changes in bacterial and fungal microbiome profiles over a week of disease progression, treatment, and clinical status using targeted next-generation sequencing (NGS). Samples were collected from infected and healthy contralateral eyes of 25 FK patients and one eye of 10 healthy, non-infected cataract controls. QIIME (Quantitative Insights into Microbial Ecology) and MicrobiomeAnalyst were utilised for the data analysis. There was a reduction in beneficial bacteria like Prevotella, Lactobacillus, and Leuconostoc in FK patients compared to the control samples. On the other hand, opportunistic bacteria including Clostridium, Bifidobacterium, and Pseudomonas increased in FK patients. Aspergillus, Colletotrichum, and Basidiobolus were more abundant in keratitis patients, whereas Malassezia and Trichoderma were less abundant. This dysbiosis was also evident in the uninfected contralateral eyes of FK patients. Treatment resulted in significant changes in bacterial genera like Dolosigranulum, Sutterella, and Akkermansia, and fungal genera such as Myrothecium, Corynespora, and Penicillium. Further, treatment returned them to the control group levels, except for Akkermansia and Corynespora. Among the treated patients, a large subset remains nonresponsive to treatment. This treatment outcome, responder versus non-responder, was reflected in the abundance of bacterial genera such as Tannerella, Sutterella, Odoribacter, and fungal genera such as Coprinellus and Volutella. This study highlights the clinical relevance of microbiome signatures in FK, demonstrating bilateral dysbiosis, integrated bacterial-fungal profiling, and correlations with treatment outcomes. These findings suggest potential for microbiome-informed diagnostics, prognostic biomarkers, and risk stratification.

Far-ultraviolet C (Far-UVC) radiation, with a wavelength range from 200 to 235 nm, is germicidal and holds potential for clinical applications. However, its use against deep-seated and internal infections, such as those affecting the lungs, remains less well established. The safety profile of Far-UVC irradiation requires further investigation across different human tissues. In this study, we utilised a krypton-chloride excimer lamp and a pulsed laser system to examine the effects of Far-UVC irradiation on human lung cells in vitro and primary human tracheal tissue. Primary human tracheal tissue and cells exposed to continuous wave (222 nm) and pulsed 206 nm and 222 nm light at doses of 5, 25, and 50 mJ/cm 2 exhibited DNA damage, including phosphorylation of γH2AX (Ser139). The continuous wave and pulsed 222 nm irradiation caused the formation of pyrimidine-pyrimidone (6-4) photoproducts. Irradiated human lung cells demonstrated reduced viability in vitro, and increased lactate dehydrogenase release into the culture medium 48 h post-irradiation. Our findings reveal that even low doses of Far-UVC (206 nm, 222 nm) light can penetrate monolayers of human lung epithelial cells, causing direct DNA damage in the form of (6 -4) photoproducts and DNA double-strand breaks, ultimately leading to cell death.

Pneumonia, a respiratory disease often caused by bacterial infection in the distal lung, requires rapid and accurate identification, especially in settings such as critical care. Initiating or de-escalating antimicrobials should ideally be guided by the quantification of pathogenic bacteria for effective treatment. Optical endomicroscopy is an emerging technology with the potential to expedite bacterial detection in the distal lung by enabling in vivo and in situ optical tissue characterisation. With advancements in detector technology, optical endomicroscopy can utilize fluorescence lifetime imaging (FLIM) to help detect events that were previously challenging or impossible to identify using fluorescence intensity imaging. In this paper, we propose an iterative Bayesian approach for bacterial detection in FLIM. We model the FLIM image as a linear combination of background intensity, Gaussian noise, and additive outliers (labelled bacteria). While previous bacteria detection methods model anomalous pixels as bacteria, here the FLIM outliers are modelled as circularly symmetric Gaussian-shaped objects, based on their discrete shape observed through visual analysis and the physical nature of the imaging modality. A Hierarchical Bayesian model is used to solve the bacterial detection problem where prior distributions are assigned to unknown parameters. A Metropolis-Hastings within Gibbs sampler draws samples from the posterior distribution. The proposed method's detection performance is initially measured using synthetic images, and shows significant improvement over existing approaches. Further analysis is conducted on real optical endomicroscopy FLIM images annotated by trained personnel. The experiments show the proposed approach outperforms existing methods by a margin of +16.85% ( F1 ) for detection accuracy.

Pneumonia, a respiratory disease often caused by bacterial infection in the distal lung, necessitates prompt and precise diagnosis, particularly in critical care settings. Optical endomicroscopy (OEM) facilitates realtime acquisition of in vivo and in situ optical biopsies, thus expediting bacterial detection. Nonetheless, visually analysing the vast number of images generated by the OEM in real time can be challenging, potentially impeding timely intervention. In this regard, to rapidly segment and detect the bacteria, we propose EmiNet, a novel dual-stream network that integrates the capabilities of Transformer and Convolutional Neural Networks (CNN) within an encoder-decoder architecture that simultaneously captures local-global appearance and motion features. Within EmiNet, we introduce a multimodal cross-channel attention module that facilitates the integration of motion features with appearance features. Furthermore, to compensate for the lack of annotated training data, we developed a synthetic dataset by simulating bacterial motion and integrating these models onto real backgrounds devoid of bacteria. The authenticity of this dataset was confirmed through a Visual Turing Test, where medical experts assessed a mixture of synthetic and real bacterial images. The results indicate that the synthetic images are almost indistinguishable from the real ones. EmiNet's performance is evaluated on both real and synthetic datasets. Experiments show that EmiNet surpasses state-of-the-art segmentation models and leads to a 6.8% improvement in detection correlation over the state-of-the-art bacteria detection algorithms.

The emergence of multidrug resistant (MDR) pathogens and the scarcity of new potent antibiotics and antifungals are one of the biggest threats to human health. Antimicrobial photodynamic therapy (aPDT) combines light and photosensitizers to kill drug-resistant pathogens; however, there are limited materials that can effectively ablate different classes of infective pathogens. In the present work, a new class of benzodiazole-paired materials is designed as highly potent PDT agents with broad-spectrum antimicrobial activity upon illumination with nontoxic light. The results mechanistically demonstrate that the energy transfer and electron transfer between nonphotosensitive and photosensitive benzodiazole moieties embedded within pathogen-binding peptide sequences result in increased singlet oxygen generation and enhanced phototoxicity. Chemical optimization renders PEP3 as a novel PDT agent with remarkable activity against MDR bacteria and fungi as well as pathogens at different stages of development (e.g., biofilms, spores, and fungal hyphae), which also prove effective in an ex vivo porcine model of microbial keratitis. The chemical modularity of this strategy and its general compatibility with peptide-based targeting agents will accelerate the design of highly photosensitive materials for antimicrobial PDT.

Pneumonia, a lung infection typically caused by bacteria, requires swift and accurate diagnosis, especially in critical care. Optical endomicroscopy (OEM) facilitates real-time acquisition of in vivo and in situ optical biopsies, aiding in the quick identification of bacteria. However, the challenge of visually analyzing the vast number of images generated by the OEM in real-time can lead to delays in necessary treatments. To address this, we introduce Back2Seg, a novel approach for the segmentation of bacteria in OEM image sequences. Prior research mainly focused on exploiting bacteria motion or relied on less accurate unsupervised background estimation methods. In this regard, to enhance the background estimation and thus bacteria segmentation, Back2Seg employs a two-stage architecture with one sub-network dedicated to estimating the background using a Convolutional Neural Network (CNN)-Transformer architecture and the other is a dual-input network, processing both the original and the estimated background sequences to accurately segment the bacteria. Our experiments demonstrate that Back2Seg effectively integrates the advantages of both supervised and unsupervised learning techniques, showing a 4.62% increase in correlation with annotations over unsupervised models and a 1.05 reduction in root mean squared error (RMSE), outperforming the top supervised approach.

Microbial keratitis (MK) is an infection of the cornea, caused by bacteria, fungi, parasites, or viruses. MK leads to significant morbidity, being the fifth leading cause of blindness worldwide. There is an urgent requirement to better understand pathogenesis in order to develop novel diagnostic and therapeutic approaches to improve patient outcomes. Many in vitro, ex vivo and in vivo MK models have been developed and implemented to meet this aim. Here, we present current in vitro and ex vivo MK model systems, examining their varied design, outputs, reporting standards, and strengths and limitations. Major limitations include their relative simplicity and the perceived inability to study the immune response in these MK models, an aspect widely accepted to play a significant role in MK pathogenesis. Consequently, there remains a dependence on in vivo models to study this aspect of MK.However, looking to the future, we draw from the broader field of corneal disease modelling, which utilises, for example, three-dimensional co-culture models and dynamic environments observed in bioreactors and organ-on-a-chip scenarios. These remain unexplored in MK research, but incorporation of these approaches will offer further advances in the field of MK corneal modelling, in particular with the focus of incorporation of immune components which we anticipate will better recapitulate pathogenesis and yield novel findings, therefore contributing to the enhancement of MK outcomes.

Pneumonia, a lung-related illness often resulting from bacterial infection, requires quick and accurate identification, especially in intensive care situations. Optical endomicroscopy (OEM) offers a solution by providing real-time acquisition of in vivo and in situ optical biopsies, enhancing the speed of bacterial identification. However, the sheer volume of images produced by OEM for real-time analysis poses a significant challenge, potentially delaying critical treatments. Prior approaches to bacteria detection in OEM imagery have relied on unsupervised models. These models are hindered by either the need for manual threshold setting or high computational demands, making real-time analysis unfeasible. To address these challenges, supervised learning methods can be considered, as they have shown superior performance and efficiency in various medical applications. However, supervised learning approaches heavily depend on the availability of vast quantities of accurately labeled data, which is scarce for OEM images. To this end, in this paper we introduce a novel approach to generate synthetic bacteria within OEM image sequences, enabling the use of deep learning techniques. We developed two models to simulate bacterial movement and embedded these models into real, bacteria-free background images. To assess the efficacy of synthetic image sequences, we employed a 3D U-Net for training. The results revealed that the 3D U-Net, when trained on these synthetic sequences, exhibited a 3.86% enhancement in correlation with real annotations over state-of-the-art bacteria detection models.

Time-resolved fluorescence imaging techniques, like confocal fluorescence lifetime imaging microscopy, are powerful photonic instrumentation tools of modern science with diverse applications, including: biology, medicine, and chemistry. However, complexities of the systems, both at specimen and device levels, cause difficulties in quantifying soft biomarkers. To address the problems, we first aim to understand and model the underlying photophysics of fluorescence decay curves. For this purpose, we provide a set of mathematical functions, called "life models", fittable with the real temporal recordings of histogram of photon counts. For each model, an equivalent electrical circuit, called a "life circuit", is derived for explaining the whole process. In confocal endomicroscopy, the components of excitation laser, specimen, and fluorescence-emission signal as the histogram of photon counts are modelled by a power source, network of resistor-inductor-capacitor circuitry, and multimetre, respectively. We then design a novel pixel-level temporal classification algorithm, called a "fit-flexible approach", where qualities of "intensity", "fall-time", and "life profile" are identified for each point. A model selection mechanism is used at each pixel to flexibly choose the best representative life model based on a proposed Misfit-percent metric. A two-dimensional arrangement of the quantified information detects some kind of structural information. This approach showed a potential of separating microbeads from lung tissue, distinguishing the tri-sensing from conventional methods. We alleviated by 7% the error of the Misfit-percent for recovering the histograms on real samples than the best state-of-the-art competitor. Codes are available online.

IntroductionPulmonary-resident memory T cells (TRM) and B cells (BRM) orchestrate protective immunity to reinfection with respiratory pathogens. Developing methods for the in situ detection of these populations would benefit both research and clinical settings.MethodsTo address this need, we developed a novel in situ immunolabelling approach combined with clinic-ready fibre-based optical endomicroscopy (OEM) to detect canonical markers of lymphocyte tissue residency in situ in human lungs undergoing ex vivo lung ventilation (EVLV).ResultsInitially, cells from human lung digests (confirmed to contain TRM/BRM populations using flow cytometry) were stained with CD69 and CD103/CD20 fluorescent antibodies and imaged in vitro using KronoScan, demonstrating it’s ability to detect antibody labelled cells. We next instilled these pre-labelled cells into human lungs undergoing EVLV and confirmed they could still be visualised using both fluorescence intensity and lifetime imaging against background lung architecture. Finally, we instilled fluorescent CD69 and CD103/CD20 antibodies directly into the lung and were able to detect TRM/BRM following in situ labelling within seconds of direct intra-alveolar delivery of microdoses of fluorescently labelled antibodies.DiscussionIn situ, no wash, immunolabelling with intra-alveolar OEM imaging is a novel methodology with the potential to expand the experimental utility of EVLV and pre-clinical models.

Decades of antibiotic misuse have led to alarming levels of antimicrobial resistance, and the development of alternative diagnostic and therapeutic strategies to delineate and treat infections is a global priority. In particular, the nosocomial, multi-drug resistant “ESKAPE” pathogens such as Gram-positive methicillin resistant Staphylococcus aureus (MRSA) and vancomycin resistant Enterococcus spp (VRE) urgently require alternative treatments. Here, we developed light-activated molecules, based on conjugation of the FDA-approved photosensitizer riboflavin to the Gram-positive specific ligand vancomycin, to enable targeted antimicrobial photodynamic therapy. The riboflavin-vancomycin conjugate proved to be a potent and versatile antibacterial agent, enabling the rapid, light-mediated, killing of MRSA and VRE with no significant off-target effects. The attachment of riboflavin on vancomycin also led to an increased in antibiotic activity against S. aureus and VRE. Simultaneously, we evidenced for the first time that the flavin sub-unit undergoes an efficient photo-induced bond cleavage reaction to release vancomycin, thereby acting as a photo-removable protecting group for drug-delivery.

Rationale: High circulating galectin-3 is associated with poor outcomes in patients with coronavirus disease (COVID-19). We hypothesized that GB0139, a potent inhaled thiodigalactoside galectin-3 inhibitor with antiinflammatory and antifibrotic actions, would be safely and effectively delivered in COVID-19 pneumonitis. Objectives: Primary outcomes were safety and tolerability of inhaled GB0139 as an add-on therapy for patients hospitalized with COVID-19 pneumonitis. Methods: We present the findings of two arms of a phase Ib/IIa randomized controlled platform trial in hospitalized patients with confirmed COVID-19 pneumonitis. Patients received standard of care (SoC) or SoC plus 10 mg inhaled GB0139 twice daily for 48 hours, then once daily for up to 14 days or discharge. Measurements and Main Results: Data are reported from 41 patients, 20 of which were assigned randomly to receive GB0139. Primary outcomes: the GB0139 group experienced no treatmentrelated serious adverse events. Incidences of adverse events were similar between treatment arms (40 with GB01391SoC vs. 35 with SoC). Secondary outcomes: plasma GB0139 was measurable in all patients after inhaled exposure and demonstrated target engagement with decreased circulating galectin (overall treatment effect post-hoc analysis of covariance [ANCOVA] over days 2-7; P = 0.0099 vs. SoC). Plasma biomarkers associated with inflammation, fibrosis, coagulopathy, and major organ function were evaluated. Conclusions: In COVID-19 pneumonitis, inhaled GB0139 was well-tolerated and achieved clinically relevant plasma concentrations with target engagement. The data support larger clinical trials to determine clinical efficacy. Clinical trial registered with ClinicalTrials.gov (NCT04473053) and EudraCT (2020-002230-32).

Bacterial infections remain among the biggest challenges to human health, leading to high antibiotic usage, morbidity, hospitalizations, and accounting for approximately 8 million deaths worldwide every year.The overuse of antibiotics and paucity of antimicrobial innovation has led to antimicrobial resistant pathogens that threaten to reverse key advances of modern medicine.Photodynamic therapeutics can kill bacteria but there are few agents that can ablate pathogens with minimal off-target effects.Methods: We describe nitrobenzoselenadiazoles as some of the first environmentally sensitive organic photosensitizers, and their adaptation to produce theranostics with optical detection and light-controlled antimicrobial activity.We combined nitrobenzoselenadiazoles with bacteria-targeting moieties (i.e., glucose-6-phosphate, amoxicillin, vancomycin) producing environmentally sensitive photodynamic agents.Results: The labelled vancomycin conjugate was able to both visualize and eradicate multidrug resistant Gram-positive ESKAPE pathogens at nanomolar concentrations, including clinical isolates and those that form biofilms.Conclusion: Nitrobenzoselenadiazole conjugates are easily synthesized and display strong environment dependent ROS production.Due to their small size and non-invasive character, they unobtrusively label antimicrobial targeting moieties.We envisage that the simplicity and modularity of this chemical strategy will accelerate the rational design of new antimicrobial therapies for refractory bacterial infections.