Symptom Checkers (SCs) provide medical information tailored to user symptoms. A critical challenge in SC development is preventing unexpected performance degradation for individual diseases, especially rare diseases, when updating algorithms. This risk stems from the lack of practical pre-deployment evaluation methods. For rare diseases, obtaining sufficient evaluation data from user feedback is difficult. To evaluate the impact of algorithm updates on the diagnostic performance for individual rare diseases before deployment, this study proposes and validates a novel Synthetic Vignette Simulation Approach. This approach aims to enable this essential evaluation efficiently and at a low cost. To estimate the impact of algorithm updates, we generated synthetic vignettes from disease-phenotype annotations in the Human Phenotype Ontology (HPO), a publicly available knowledge base for rare diseases curated by experts. Using these vignettes, we simulated SC interviews to predict changes in diagnostic performance. The effectiveness of this approach was validated retrospectively by comparing the predicted changes with actual performance metrics using the R-squared ($R^2$) coefficient. Our experiment, covering eight past algorithm updates for rare diseases, showed that the proposed method accurately predicted performance changes for diseases with phenotype frequency information in HPO (n=5). For these updates, we found a strong correlation for both Recall@8 change ($R^2$ = 0.83,$p$ = 0.031) and Precision@8 change ($R^2$ = 0.78,$p$ = 0.047). Our proposed method enables the pre-deployment evaluation of SC algorithm changes for individual rare diseases. This evaluation is based on a publicly available medical knowledge database created by experts, ensuring transparency and explainability for stakeholders. Additionally, SC developers can efficiently improve diagnostic performance at a low cost.
Joy Kitson, Prescott C. Alexander, Joseph Tuccillo
et al.
Human cognitive responses, behavioral responses, and disease dynamics co-evolve over the course of any disease outbreak, and can result in complex feedbacks. We present a dynamic agent-based model that explicitly couples the spread of disease with the spread of fear surrounding the disease, implemented within the EpiCast simulation framework. EpiCast models transmission across a realistic synthetic population, capturing individual-level interactions. In our model, fear propagates through both in-person contact and broadcast media, prompting individuals to adopt protective behaviors that reduce disease spread. In order to better understand these coupled dynamics, we create and compare a range of compartmental surrogate models to analyze the impact of including various disease states. Additionally, we compare a range of behavioral scenarios within EpiCast, varying the level and intensity of fear and behavioral change. Our results show that the addition of asymptomatic, exposed, and pre-symptomatic disease states can impact both the rate at which an outbreak progresses and its overall trajectory. Moreover, the combination of non-local fear spread via broadcasters and strong behavioral responses by fearful individuals generally leads to multiple epidemic waves, an outcome that occurs only within a narrow parameter range when fear spreads purely through local contact. Accounting for the coupled spread of fear and disease is critical for understanding disease dynamics and designing timely, targeted responses to emerging infectious threats.
Zilal Eiz AlDin, John Wu, Jeffrey Paul Fung
et al.
Despite rare diseases affecting 1 in 10 Americans, their differential diagnosis remains challenging. Due to their impressive recall abilities, large language models (LLMs) have been recently explored for differential diagnosis. Existing approaches to evaluating LLM-based rare disease diagnosis suffer from two critical limitations: they rely on idealized clinical case studies that fail to capture real-world clinical complexity, or they use ICD codes as disease labels, which significantly undercounts rare diseases since many lack direct mappings to comprehensive rare disease databases like Orphanet. To address these limitations, we explore MIMIC-RD, a rare disease differential diagnosis benchmark constructed by directly mapping clinical text entities to Orphanet. Our methodology involved an initial LLM-based mining process followed by validation from four medical annotators to confirm identified entities were genuine rare diseases. We evaluated various models on our dataset of 145 patients and found that current state-of-the-art LLMs perform poorly on rare disease differential diagnosis, highlighting the substantial gap between existing capabilities and clinical needs. From our findings, we outline several future steps towards improving differential diagnosis of rare diseases.
Plant disease is a critical factor affecting agricultural production. Traditional manual recognition methods face significant drawbacks, including low accuracy, high costs, and inefficiency. Deep learning techniques have demonstrated significant benefits in identifying plant diseases, but they still face challenges such as inference delays and high energy consumption. Deep learning algorithms are difficult to run on resource-limited embedded devices. Offloading these models to cloud servers is confronted with the restriction of communication bandwidth, and all of these factors will influence the inference's efficiency. We propose a collaborative inference framework for recognizing plant diseases between edge devices and cloud servers to enhance inference speed. The DNN model for plant disease recognition is pruned through deep reinforcement learning to improve the inference speed and reduce energy consumption. Then the optimal split point is determined by a greedy strategy to achieve the best collaborated inference acceleration. Finally, the system for collaborative inference acceleration in plant disease recognition has been implemented using Gradio to facilitate friendly human-machine interaction. Experiments indicate that the proposed collaborative inference framework significantly increases inference speed while maintaining acceptable recognition accuracy, offering a novel solution for rapidly diagnosing and preventing plant diseases.
Kelvin Moore Jr., Noelle Le Tourneau, Jasmin Alvarez
et al.
Abstract Background Point-of-care HIV viral load testing may enhance patient care and improve HIV health services. We aimed to evaluate the feasibility and acceptability of implementing such testing in a high-volume community sexual health clinic in the United States. Methods We conducted a cross-sectional, mixed-methods study. Remnant venipuncture specimens from clients undergoing HIV and other sexual health screenings were analyzed using the Xpert® HIV-1 Viral Load assay. Results were compared to COBAS® AmpliPrep/COBAS® TaqMan® HIV-1 Test. Clinical staff observations, study meeting notes, and two semi-structured in-depth interviews with clinical staff were used to understand perspectives on incorporating this testing into clinical practice. Results We analyzed 113 samples from 111 clients. The Xpert assay showed excellent agreement with COBAS, with no clinically significant difference in viral load measurements. Clinical staff found Xpert testing acceptable, based on its ability to provide rapid, accurate test results and potential for bridging patient care gaps. Respondents noted that this testing would be particularly beneficial for individuals in whom barriers to care engagement may complicate follow-up. Challenges in implementation included machine errors as well as concerns related to staff workload, testing logistics, and the need for comprehensive equipment training. Conclusions This study represents the first effort in the United States to describe the feasibility of HIV viral load point-of-care testing in routine care. While the Xpert demonstrated comparable results to standard-of-care testing and staff found it acceptable, further work is needed to develop the workflow and implementation strategies that would enable real-time use and improved patient care. Clinical trial Not applicable.
Jerson Andrés Cuéllar-Sáenz, Alfonso J. Rodríguez-Morales, Álvaro A. Faccini-Martínez
Yellow fever, a zoonotic arboviral disease, has reemerged in Colombia, triggering a major outbreak in the country. During 2024 through mid-2025, a total of 132 human cases and 68 infections in nonhuman primates were confirmed, primarily in the department of Tolima, historically considered a low-risk area. We analyzed the historical and current epidemiology of yellow fever in Colombia, highlighting ecologic, social, and surveillance factors that contributed to the outbreak. Low vaccination coverage, insufficient epizootic and entomological surveillance, deforestation, habitat fragmentation, and limited application of One Health approaches have all exacerbated the situation. The high mortality rate of nonhuman primate species indicated a more profound ecologic crisis. Immediate, comprehensive measures, including mass vaccination, genomic surveillance, and integrated One Health frameworks, are urgently needed. Colombia’s experience underscores the need to reevaluate risk stratification and preparedness strategies across the Americas to prevent future yellow fever outbreaks in previously unaffected regions.
Darlyn Buenaño Vera, Byron Oviedo, Washington Chiriboga Casanova
et al.
The early identification of diseases in cocoa pods is an important task to guarantee the production of high-quality cocoa. The use of artificial intelligence techniques such as machine learning, computer vision and deep learning are promising solutions to help identify and classify diseases in cocoa pods. In this paper we introduce the development and evaluation of a deep learning computational model applied to the identification of diseases in cocoa pods, focusing on "monilia" and "black pod" diseases. An exhaustive review of state-of-the-art of computational models was carried out, based on scientific articles related to the identification of plant diseases using computer vision and deep learning techniques. As a result of the search, EfficientDet-Lite4, an efficient and lightweight model for object detection, was selected. A dataset, including images of both healthy and diseased cocoa pods, has been utilized to train the model to detect and pinpoint disease manifestations with considerable accuracy. Significant enhancements in the model training and evaluation demonstrate the capability of recognizing and classifying diseases through image analysis. Furthermore, the functionalities of the model were integrated into an Android native mobile with an user-friendly interface, allowing to younger or inexperienced farmers a fast and accuracy identification of health status of cocoa pods
Rare diseases pose significant challenges in diagnosis and treatment due to their low prevalence and heterogeneous clinical presentations. Unstructured clinical notes contain valuable information for identifying rare diseases, but manual curation is time-consuming and prone to subjectivity. This study aims to develop a hybrid approach combining dictionary-based natural language processing (NLP) tools with large language models (LLMs) to improve rare disease identification from unstructured clinical reports. We propose a novel hybrid framework that integrates the Orphanet Rare Disease Ontology (ORDO) and the Unified Medical Language System (UMLS) to create a comprehensive rare disease vocabulary. The proposed hybrid approach demonstrates superior performance compared to traditional NLP systems and standalone LLMs. Notably, the approach uncovers a significant number of potential rare disease cases not documented in structured diagnostic records, highlighting its ability to identify previously unrecognized patients.
Rachael A. Turner, Roseann M. Johnson, Yasmin Yazdani-Farsad
et al.
Abstract
Introduction:
Periprosthetic joint infection (PJI) may result from pathogen-to-patient transmission within the environment. High-touch surfaces (HTS) areas near the operative field from previous studies had been identified as the least likely to be thoroughly cleaned between operative cases and were utilized for this study. The purpose of this study was to assess the impact of a handheld ultraviolet-c (UV-C) light-emitting diode (LED) disinfection device on the decontamination of HTS in the operating room.
Methods:
This prospective study was conducted between 03/02/2021 and 04/20/2021. Tryptic soy agar contact plates were used to determine the bacterial load of the selected surfaces before the initiation of the case, after the case was complete, before manual cleaning, and after disinfection of the LED device. The plates were then incubated for 48 hours at 36º +/–1° C. Colony forming units (CFU) were recorded 48 hours after incubation. Mean, median, and range of CFU were recorded.
Results:
Average CFU per surface before and after the surgical case were 14.1 (range 0–200) and 13.5 (range 0–200) respectively, these were not significantly different (P = 0.9397). Manual cleaning reduced average CFU by 74% to 3.35 (range 0–200) per surface (P = 0.0162). Disinfection with the handheld LED unit further reduced the average CFU by 92% to 0.28 (range 0–4) per surface (P < 0.0001).
Conclusions:
A handheld UV-C LED disinfection device may decrease environmental contamination near the operative field in HTS areas. Further research is warranted with this technology to determine if this correlates with a decrease in PJI.
Infectious and parasitic diseases, Public aspects of medicine
Nooshin Mojahed, M. A. Mohammadkhani, A. Mohamadkhani
Background: Climate change based on temperature, humidity and wind can improve many characteristics of the arthropod carrier life cycle, including survival, arthropod population, pathogen communication, and the spread of infectious agents from vectors. This study aimed to find association between content of disease followed climate change we demonstrate in humans. Methods: All the articles from 2016 to 2021 associated with global climate change and the effect of vector-borne disease were selected form databases including PubMed and the Global Biodiversity information facility database. All the articles selected for this short review were English. Results: Due to the high burden of infectious diseases and the growing evidence of the possible effects of climate change on the incidence of these diseases, these climate changes can potentially be involved with the COVID-19 epidemic. We highlighted the evidence of vector-borne diseases and the possible effects of climate change on these communicable diseases. Conclusion: Climate change, specifically in rising temperature system is one of the world’s greatest concerns already affected pathogen-vector and host relation. Lice parasitic, fleas, mites, ticks, and mosquitos are the prime public health importance in the transmission of virus to human hosts.
Scrub typhus is a poorly studied but life-threatening disease caused by the intracellular bacteriumOrientia tsutsugamushi(Ot). Cellular and humoral immunity inOt-infected patients is not long-lasting, waning as early as one-year post-infection; however, its underlying mechanisms remain unclear. To date, no studies have examined germinal center (GC) or B cell responses inOt-infected humans or experimental animals. This study was aimed at evaluating humoral immune responses at acute stages of severeOtinfection and possible mechanisms underlying B cell dysfunction. Following inoculation withOtKarp, a clinically dominant strain known to cause lethal infection in C57BL/6 mice, we measured antigen-specific antibody titers, revealing IgG2c as the dominant isotype induced by infection. Splenic GC responses were evaluated by immunohistology, co-staining for B cells (B220), T cells (CD3), and GCs (GL-7). Organized GCs were evident at day 4 post-infection (D4), but they were nearly absent at D8, accompanied by scattered T cells throughout splenic tissues. Flow cytometry revealed comparable numbers of GC B cells and T follicular helper (Tfh) cells at D4 and D8, indicating that GC collapse was not due to excessive death of these cell subtypes at D8. B cell RNAseq analysis revealed significant differences in expression of genes associated with B cell adhesion and co-stimulation at D8 versus D4. The significant downregulation ofS1PR2(a GC-specific adhesion gene) was most evident at D8, correlating with disrupted GC formation. Signaling pathway analysis uncovered downregulation of 71% of B cell activation genes at D8, suggesting attenuation of B cell activation during severe infection. This is the first study showing the disruption of B/T cell microenvironment and dysregulation of B cell responses duringOtinfection, which may help understand the transient immunity associated with scrub typhus.
The number of microbes on Earth may be 1030, exceeding all other diversity. A small number of these can infect people and cause disease. The diversity of parasitic organisms likely correlates with the hosts they live in and the number mammal hosts for zoonotic infections increases with species richness among mammalian orders. Thus, while habitat loss and fragmentation may reduce species diversity, the habitat encroachment by people into species-rich areas may increase the exposure of people to novel infectious agents from wildlife. Here, we present a theoretical framework that exploits the species–area relationship to link the exposure of people to novel infections with habitat biodiversity. We model changes in human exposure to microbes through defined classes of habitat fragmentation and predict that increased habitat division intrinsically increases the hazard from microbes for all modelled biological systems. We apply our model to African tropical forests as an example. Our results suggest that it is possible to identify high-risk areas for the mitigation and surveillance of novel disease emergence and that mitigation measures may reduce this risk while conserving biodiversity.
Christopher J. Banks, Katherine Simpson, Nicholas Hanley
et al.
Financial incentives encourage the plantation of new woodland to increase habitat, biodiversity, carbon sequestration, as a contribution to meeting climate change and biodiversity conservation targets. Whilst these are largely positive effects, it is worth considering that this expansion of woodland can lead to increased presence of wildlife species in proximity to agricultural holdings that may pose an enhanced risk of disease transmission between wildlife and livestock. Wildlife and the provision of a reservoir for infectious disease is particularly important in the transmission dynamics of bovine tuberculosis, the case studied here. In this paper we develop an economic model for predicting changes in land use resulting from subsidies for woodland planting. We use this to assess the consequent impact on wild deer populations in the newly created woodland areas, and thus the emergent infectious disease risk arising from the proximity of new and existing wild deer populations and existing cattle holdings. We consider an area in the South-West of Scotland, having existing woodland, deer populations, and extensive and diverse cattle farm holdings. In this area we find that, with a varying level of subsidy and plausible new woodland creation scenarios, the contact risk between areas of wild deer and cattle increases between 26% and 35% over the risk present with a zero subsidy. This provides a foundation for extending to larger regions and for examining potential risk mitigation strategies, for example the targeting of subsidy in low disease risk areas, or provisioning for buffer zones between woodland and agricultural holdings.
The last century has witnessed an increasing rate of new disease emergence across the world leading to permanent loss of biodiversity. Perkinsea is a microeukaryotic parasitic phylum composed of four main lineages of parasitic protists with broad host ranges. Some of them represent major ecological and economical threats because of their geographically invasive ability and pathogenicity (leading to mortality events). In marine environments, three lineages are currently described, the Parviluciferaceae, the Perkinsidae, and the Xcellidae, infecting, respectively, dinoflagellates, mollusks, and fish. In contrast, only one lineage is officially described in freshwater environments: the severe Perkinsea infectious agent infecting frog tadpoles. The advent of high-throughput sequencing methods, mainly based on 18S rRNA assays, showed that Perkinsea is far more diverse than the previously four described lineages especially in freshwater environments. Indeed, some lineages could be parasites of green microalgae, but a formal nature of the interaction needs to be explored. Hence, to date, most of the newly described aquatic clusters are only defined by their environmental sequences and are still not (yet) associated with any host. The unveiling of this microbial black box presents a multitude of research challenges to understand their ecological roles and ultimately to prevent their most negative impacts. This review summarizes the biological and ecological traits of Perkinsea—their diversity, life cycle, host preferences, pathogenicity, and highlights their diversity and ubiquity in association with a wide range of hosts.
Understanding the interplay between human behavioral phenomena and infectious disease dynamics has been one of the central challenges of mathematical epidemiology. However, socio-cognitive processes critical for the initiation of desired behavioral responses during an outbreak have often been neglected or oversimplified in earlier models. Combining the microscopic Markov chain approach with the law of total probability, we herein institute a mathematical model describing the dynamic interplay between stage-based progression of awareness diffusion and endemic disease transmission in multiplex networks. We analytically derived the epidemic thresholds for both discrete-time and continuous-time versions of our model, and we numerically demonstrated the accuracy of our analytic arguments in capturing the time course and the steady-state of the coupled disease-awareness dynamics. We found that our model is exact for arbitrary unclustered multiplex networks, outperforming a widely adopted probability-tree-based method, both in the prediction of the time-evolution of a contagion and in the final epidemic size. Our findings show that informing the unaware individuals about the circulating disease will not be sufficient for the prevention of an outbreak unless the distributed information triggers strong awareness of infection risks with adequate protective measures, and that the immunity of highly-aware individuals can elevate the epidemic threshold, but only if the rate of transition from weak to strong awareness is sufficiently high. Our study thus reveals that awareness diffusion and other behavioral parameters can nontrivially interact when producing their effects on epidemiological dynamics of an infectious disease, suggesting that future public health measures should not ignore this complex behavioral interplay and its influence on contagion transmission in multilayered networked systems.
Fakrul Islam Tushar, Vincent M. D'Anniballe, Geoffrey D. Rubin
et al.
Despite the potential of weakly supervised learning to automatically annotate massive amounts of data, little is known about its limitations for use in computer-aided diagnosis (CAD). For CT specifically, interpreting the performance of CAD algorithms can be challenging given the large number of co-occurring diseases. This paper examines the effect of co-occurring diseases when training classification models by weakly supervised learning, specifically by comparing multi-label and multiple binary classifiers using the same training data. Our results demonstrated that the binary model outperformed the multi-label classification in every disease category in terms of AUC. However, this performance was heavily influenced by co-occurring diseases in the binary model, suggesting it did not always learn the correct appearance of the specific disease. For example, binary classification of lung nodules resulted in an AUC of < 0.65 when there were no other co-occurring diseases, but when lung nodules co-occurred with emphysema, the performance reached AUC > 0.80. We hope this paper revealed the complexity of interpreting disease classification performance in weakly supervised models and will encourage researchers to examine the effect of co-occurring diseases on classification performance in the future.
Understanding the spread of diseases through complex networks is of great interest where realistic, heterogeneous contact patterns play a crucial role in the spread. Most works have focused on mean-field behavior -- quantifying how contact patterns affect the emergence and stability of (meta)stable endemic states in networks. On the other hand, much less is known about longer time scale dynamics, such as disease extinction, whereby inherent process stochasticity and contact heterogeneity interact to produce large fluctuations that result in the spontaneous clearance of infection. Here we show that heterogeneity in both susceptibility and infectiousness (incoming and outgoing degree, respectively) has a non-trivial effect on extinction in directed contact networks, both speeding-up and slowing-down extinction rates depending on the relative proportion of such edges in a network, and on whether the heterogeneities in the incoming and outgoing degrees are correlated or anticorrelated. In particular, we show that weak anticorrelated heterogeneity can increase the disease stability, whereas strong heterogeneity gives rise to markedly different results for correlated and anticorrelated heterogeneous networks. All analytical results are corroborated through various numerical schemes including network Monte-Carlo simulations.
Background: After Neisseria gonorrhoeae FC428 was first found in Japan, ceftriaxone-resistant strains disseminated globally, and the gonococcal resistance rate increased remarkably. Epidemiological investigations are greatly significant for the analysis of antimicrobial resistance (AMR) trends, molecular features and evolution. Objectives: To clarify the AMR trend from 2016–2019 and reveal the molecular characteristics and evolution of ceftriaxone-resistant penA 60.001 isolates. Methods: The minimum inhibitory concentrations (MICs) of antibiotics against 4113 isolates were detected by the agar dilution method. N. gonorrhoeae multiantigen sequence typing (NG-MAST), multilocus sequence typing (MLST) and N.gonorrhoeae sequence typing for antimicrobial resistance (NG-STAR) were used to identify the sequence types. Genome analysis was conducted to analyze resistance genes, virulence factors, and evolutionary sources. Results: Isolates with decreased ceftriaxone susceptibility have increased from 2.05% (2016) to 16.18% (2019). Six ceftriaxone-resistant isolates possessing penA 60.001 appeared in Guangdong Province, and were resistant to ceftriaxone, penicillin, tetracycline, ciprofloxacin and cefixime, but susceptible to azithromycin and spectinomycin. Single-nucleotide polymorphisms (SNPs) in the porB gene were the major cause of different NG-MAST types. ST1903 was the main NG-STAR genotype and only strain-ZH545 was ST7365, with molecular features consistent with the MICs. Furthermore, different MLSTs suggested diverse evolutionary sources. Genome analysis revealed a set of virulence factors along with the resistance genes “penA” and “blaTEM-1B”. Half of penA 60.001 strains were fully mixed with global FC428-related strains. Conclusions: Global FC428-related clones have disseminated across Guangdong, possibly causing decreased ceftriaxone susceptibility. Enhanced gonococcal surveillance will help elucidate the trajectory of transmission and curb further dissemination.
Milan Radovanovic, Brodie R. Marthaler, Charles W. Nordstrom
et al.
Background: Cardiobacterum hominis (C. hominis) is the part of the HACEK group (Haemophilus spp, Actinobacillus spp, C. hominis, Eikenella, and Kingella spp) that accounts for the majority of the Gram-negative infective endocarditis cases. Historically, the fastidious characteristics of these microorganisms proved challenging to many clinicians. Advances in microbiological identification of culture-negative endocarditis; however, may be the reason for the rising incidence of these infections. Here, we report an incidentally diagnosed C. hominis endocarditis following an aortic valve replacement. Case report: A healthy 54-year-old gentleman presented after several months of generalized weakness and exertional intolerance. He was found to have a bicuspid aortic valve with regurgitation and underwent aortic valve replacement surgery. Intraoperatively, the patient was found to have a large perforation of the fused leaflet associated with abnormal pink tissue in the aortic valve area. The aortic valve tissue was cultured. Gram-negative rods were isolated 48 h later and were ultimately identified as C. hominis. He was successfully treated with 6 weeks of intravenous ceftriaxone with sterile blood cultures throughout the hospital stay. In retrospect, the patient’s valve failure was likely secondary to subacute endocarditis from C. hominis complicated by leaflet perforation. Conclusion: C. hominis is a rare cause of infective endocarditis with an excellent prognosis when timely diagnosed and managed. By reporting this case, we wish to raise awareness of potential asymptomatic infection, particularly amongst patients with underlying native valve abnormalities, new leaflet perforation, and valve insufficiency.