Digital Microscopy Augmented by Artificial Intelligence to Interpret Bone Marrow Samples for Hematological Diseases.

Analysis of bone marrow aspirates (BMAs) is an essential step in the diagnosis of hematological disorders. This analysis is usually performed based on a visual examination of samples under a conventional optical microscope, which involves a labor-intensive process, limited by clinical experience and subject to high observer variability. In this work, we present a comprehensive digital microscopy system that enables BMA analysis for cell type counting and differentiation in an efficient and objective manner. This system not only provides an accessible and simple method to digitize, store, and analyze BMA samples remotely but is also supported by an Artificial Intelligence (AI) pipeline that accelerates the differential cell counting process and reduces interobserver variability. It has been designed to integrate AI algorithms with the daily clinical routine and can be used in any regular hospital workflow.

A Smartphone App for Real-Time Assessment of Malaria Prophylaxis Adverse Events.

Background: Growth of international travel to malarial areas over the last decades has contributed to more travelers taking malaria prophylaxis. Travel-related symptoms may be wrongly attributed to malaria prophylaxis and hinder compliance. Here, we aimed to assess the frequency of real-time reporting of symptoms by travelers following malaria prophylaxis using a smartphone app. Method: Adult international travelers included in this single-center study (Barcelona, Spain) used the smartphone Trip Doctor® app developed by our group for real-time tracking of symptoms and adherence to prophylaxis. Results: Six hundred four (n = 604) international travelers were included in the study; 74.3% (449) used the app daily, and for one-quarter of travelers, malaria prophylaxis was prescribed. Participants from the prophylaxis group traveled more to Africa (86.7% vs. 4.3%; p < 0.01) and to high travel medical risk countries (60.8% vs. 18%; p < 0.01) and reported more immunosuppression (30.8% vs. 23.1% p < 0.01). Regarding symptoms, no significant intergroup differences were observed, and no relationship was found between the total number of malarial pills taken and reported symptoms. Conclusions: In our cohort, the number of symptoms due to malaria prophylaxis was not significantly higher than in participants for whom prophylaxis was not prescribed, and the overall proportion of symptoms is higher compared with other studies.

Exploring YouTube's Recommendation System in the Context of COVID-19 Vaccines: Computational and Comparative Analysis of Video Trajectories.

BACKGROUND: Throughout the COVID-19 pandemic, there has been a concern that social media may contribute to vaccine hesitancy due to the wide availability of antivaccine content on social media platforms. YouTube has stated its commitment to removing content that contains misinformation on vaccination. Nevertheless, such claims are difficult to audit. There is a need for more empirical research to evaluate the actual prevalence of antivaccine sentiment on the internet. OBJECTIVE: This study examines recommendations made by YouTube's algorithms in order to investigate whether the platform may facilitate the spread of antivaccine sentiment on the internet. We assess the prevalence of antivaccine sentiment in recommended videos and evaluate how real-world users' experiences are different from the personalized recommendations obtained by using synthetic data collection methods, which are often used to study YouTube's recommendation systems. METHODS: We trace trajectories from a credible seed video posted by the World Health Organization to antivaccine videos, following only video links suggested by YouTube's recommendation system. First, we gamify the process by asking real-world participants to intentionally find an antivaccine video with as few clicks as possible. Having collected crowdsourced trajectory data from respondents from (1) the World Health Organization and United Nations system (nWHO/UN=33) and (2) Amazon Mechanical Turk (nAMT=80), we next compare the recommendations seen by these users to recommended videos that are obtained from (3) the YouTube application programming interface's RelatedToVideoID parameter (nRTV=40) and (4) from clean browsers without any identifying cookies (nCB=40), which serve as reference points. We develop machine learning methods to classify antivaccine content at scale, enabling us to automatically evaluate 27,074 video recommendations made by YouTube. RESULTS: We found no evidence that YouTube promotes antivaccine content; the average share of antivaccine videos remained well below 6% at all steps in users' recommendation trajectories. However, the watch histories of users significantly affect video recommendations, suggesting that data from the application programming interface or from a clean browser do not offer an accurate picture of the recommendations that real users are seeing. Real users saw slightly more provaccine content as they advanced through their recommendation trajectories, whereas synthetic users were drawn toward irrelevant recommendations as they advanced. Rather than antivaccine content, videos recommended by YouTube are likely to contain health-related content that is not specifically related to vaccination. These videos are usually longer and contain more popular content. CONCLUSIONS: Our findings suggest that the common perception that YouTube's recommendation system acts as a "rabbit hole" may be inaccurate and that YouTube may instead be following a "blockbuster" strategy that attempts to engage users by promoting other content that has been reliably successful across the platform.

Operational response simulation tool for epidemics within refugee and IDP settlements: A scenario-based case study of the Cox's Bazar settlement.

The spread of infectious diseases such as COVID-19 presents many challenges to healthcare systems and infrastructures across the world, exacerbating inequalities and leaving the world's most vulnerable populations most affected. Given their density and available infrastructure, refugee and internally displaced person (IDP) settlements can be particularly susceptible to disease spread. In this paper we present an agent-based modeling approach to simulating the spread of disease in refugee and IDP settlements under various non-pharmaceutical intervention strategies. The model, based on the June open-source framework, is informed by data on geography, demographics, comorbidities, physical infrastructure and other parameters obtained from real-world observations and previous literature. The development and testing of this approach focuses on the Cox's Bazar refugee settlement in Bangladesh, although our model is designed to be generalizable to other informal settings. Our findings suggest the encouraging self-isolation at home of mild to severe symptomatic patients, as opposed to the isolation of all positive cases in purpose-built isolation and treatment centers, does not increase the risk of secondary infection meaning the centers can be used to provide hospital support to the most intense cases of COVID-19. Secondly we find that mask wearing in all indoor communal areas can be effective at dampening viral spread, even with low mask efficacy and compliance rates. Finally, we model the effects of reopening learning centers in the settlement under various mitigation strategies. For example, a combination of mask wearing in the classroom, halving attendance regularity to enable physical distancing, and better ventilation can almost completely mitigate the increased risk of infection which keeping the learning centers open may cause. These modeling efforts are being incorporated into decision making processes to inform future planning, and further exercises should be carried out in similar geographies to help protect those most vulnerable.

Collaborative intelligence and gamification for on-line malaria species differentiation.

BACKGROUND: Current World Health Organization recommendations for the management of malaria include the need for a parasitological confirmation prior to triggering appropriate treatment. The use of rapid diagnostic tests (RDTs) for malaria has contributed to a better infection recognition and a more targeted treatment. Nevertheless, low-density infections and parasites that fail to produce HRP2 can cause false-negative RDT results. Microscopy has traditionally been the methodology most commonly used to quantify malaria and characterize the infecting species, but the wider use of this technique remains challenging, as it requires trained personnel and processing capacity. OBJECTIVE: In this study, the feasibility of an on-line system for remote malaria species identification and differentiation has been investigated by crowdsourcing the analysis of digitalized infected thin blood smears by non-expert observers using a mobile app. METHODS: An on-line videogame in which players learned how to differentiate the young trophozoite stage of the five Plasmodium species has been designed. Images were digitalized with a smartphone camera adapted to the ocular of a conventional light microscope. Images from infected red blood cells were cropped and puzzled into an on-line game. During the game, players had to decide the malaria species (Plasmodium falciparum, Plasmodium malariae, Plasmodium vivax, Plasmodium ovale, Plasmodium knowlesi) of the infected cells that were shown in the screen. After 2 months, each player's decisions were analysed individually and collectively. RESULTS: On-line volunteers playing the game made more than 500,000 assessments for species differentiation. Statistically, when the choice of several players was combined (n > 25), they were able to significantly discriminate Plasmodium species, reaching a level of accuracy of 99% for all species combinations, except for P. knowlesi (80%). Non-expert decisions on which Plasmodium species was shown in the screen were made in less than 3 s. CONCLUSION: These findings show that it is possible to train malaria-naïve non-experts to identify and differentiate malaria species in digitalized thin blood samples. Although the accuracy of a single player is not perfect, the combination of the responses of multiple casual gamers can achieve an accuracy that is within the range of the diagnostic accuracy made by a trained microscopist.

Plasmodium species differentiation by non-expert on-line volunteers for remote malaria field diagnosis.

BACKGROUND: Routine field diagnosis of malaria is a considerable challenge in rural and low resources endemic areas mainly due to lack of personnel, training and sample processing capacity. In addition, differential diagnosis of Plasmodium species has a high level of misdiagnosis. Real time remote microscopical diagnosis through on-line crowdsourcing platforms could be converted into an agile network to support diagnosis-based treatment and malaria control in low resources areas. This study explores whether accurate Plasmodium species identification-a critical step during the diagnosis protocol in order to choose the appropriate medication-is possible through the information provided by non-trained on-line volunteers. METHODS: 88 volunteers have performed a series of questionnaires over 110 images to differentiate species (Plasmodium falciparum, Plasmodium ovale, Plasmodium vivax, Plasmodium malariae, Plasmodium knowlesi) and parasite staging from thin blood smear images digitalized with a smartphone camera adapted to the ocular of a conventional light microscope. Visual cues evaluated in the surveys include texture and colour, parasite shape and red blood size. RESULTS: On-line volunteers are able to discriminate Plasmodium species (P. falciparum, P. malariae, P. vivax, P. ovale, P. knowlesi) and stages in thin-blood smears according to visual cues observed on digitalized images of parasitized red blood cells. Friendly textual descriptions of the visual cues and specialized malaria terminology is key for volunteers learning and efficiency. CONCLUSIONS: On-line volunteers with short-training are able to differentiate malaria parasite species and parasite stages from digitalized thin smears based on simple visual cues (shape, size, texture and colour). While the accuracy of a single on-line expert is far from perfect, a single parasite classification obtained by combining the opinions of multiple on-line volunteers over the same smear, could improve accuracy and reliability of Plasmodium species identification in remote malaria diagnosis.

Humanitarian applications of machine learning with remote-sensing data: review and case study in refugee settlement mapping.

The coordination of humanitarian relief, e.g. in a natural disaster or a conflict situation, is often complicated by a scarcity of data to inform planning. Remote sensing imagery, from satellites or drones, can give important insights into conditions on the ground, including in areas which are difficult to access. Applications include situation awareness after natural disasters, structural damage assessment in conflict, monitoring human rights violations or population estimation in settlements. We review machine learning approaches for automating these problems, and discuss their potential and limitations. We also provide a case study of experiments using deep learning methods to count the numbers of structures in multiple refugee settlements in Africa and the Middle East. We find that while high levels of accuracy are possible, there is considerable variation in the characteristics of imagery collected from different sensors and regions. In this, as in the other applications discussed in the paper, critical inferences must be made from a relatively small amount of pixel data. We, therefore, consider that using machine learning systems as an augmentation of human analysts is a reasonable strategy to transition from current fully manual operational pipelines to ones which are both more efficient and have the necessary levels of quality control.This article is part of a discussion meeting issue 'The growing ubiquity of algorithms in society: implications, impacts and innovations'.

Social Media Monitoring of Discrimination and HIV Testing in Brazil, 2014-2015.

Big data can be used to assess perceptions about public health issues. This study assessed social media data from Twitter to inform communication campaigns to promote HIV testing and reduce discrimination related to HIV/AIDS or towards key populations to the HIV epidemic, and its potential utility to evaluate such campaigns through HIV testing uptake. Tweets from Brazil were collected from January 2014 to March 2015 and filtered by four categories of keywords including discrimination, HIV prevention, HIV testing, and HIV campaigns. In total over 100,000 geo-located tweets were extracted and analyzed. A dynamic online dashboard updated daily allowed mapping trends, anomalies and influencers, and enabled its use for feedback to campaigns, including correcting misconceptions. These results encourage the use of social networking data for improved messaging in campaigns. Clinical HIV test data was collected monthly from the city of Curitiba and compared to the number of tweets mapped to the city showing a moderate positive correlation (r = 0.39). Results are limited due to the availability of the HIV testing data. The potential of social media as a proxy for HIV testing uptake needs further validation, which can only be done with higher frequency and higher spatial granularity of service delivery data, enabling comparisons with the social media data. Such timely information could empower early response immediate media messaging to support programmatic efforts, such as HIV prevention, testing, and treatment scale up.

A digital framework to build, visualize and analyze a gene expression atlas with cellular resolution in zebrafish early embryogenesis.

A gene expression atlas is an essential resource to quantify and understand the multiscale processes of embryogenesis in time and space. The automated reconstruction of a prototypic 4D atlas for vertebrate early embryos, using multicolor fluorescence in situ hybridization with nuclear counterstain, requires dedicated computational strategies. To this goal, we designed an original methodological framework implemented in a software tool called Match-IT. With only minimal human supervision, our system is able to gather gene expression patterns observed in different analyzed embryos with phenotypic variability and map them onto a series of common 3D templates over time, creating a 4D atlas. This framework was used to construct an atlas composed of 6 gene expression templates from a cohort of zebrafish early embryos spanning 6 developmental stages from 4 to 6.3 hpf (hours post fertilization). They included 53 specimens, 181,415 detected cell nuclei and the segmentation of 98 gene expression patterns observed in 3D for 9 different genes. In addition, an interactive visualization software, Atlas-IT, was developed to inspect, supervise and analyze the atlas. Match-IT and Atlas-IT, including user manuals, representative datasets and video tutorials, are publicly and freely available online. We also propose computational methods and tools for the quantitative assessment of the gene expression templates at the cellular scale, with the identification, visualization and analysis of coexpression patterns, synexpression groups and their dynamics through developmental stages.

Edge Artificial Intelligence (AI) for real-time automatic quantification of filariasis in mobile microscopy.

Filariasis, a neglected tropical disease caused by roundworms, is a significant public health concern in many tropical countries. Microscopic examination of blood samples can detect and differentiate parasite species, but it is time consuming and requires expert microscopists, a resource that is not always available. In this context, artificial intelligence (AI) can assist in the diagnosis of this disease by automatically detecting and differentiating microfilariae. In line with the target product profile for lymphatic filariasis as defined by the World Health Organization, we developed an edge AI system running on a smartphone whose camera is aligned with the ocular of an optical microscope that detects and differentiates filarias species in real time without the internet connection. Our object detection algorithm that uses the Single-Shot Detection (SSD) MobileNet V2 detection model was developed with 115 cases, 85 cases with 1903 fields of view and 3342 labels for model training, and 30 cases with 484 fields of view and 873 labels for model validation before clinical validation, is able to detect microfilariae at 10x magnification and distinguishes four species of them at 40x magnification: Loa loa, Mansonella perstans, Wuchereria bancrofti, and Brugia malayi. We validated our augmented microscopy system in the clinical environment by replicating the diagnostic workflow encompassed examinations at 10x and 40x with the assistance of the AI models analyzing 18 samples with the AI running on a middle range smartphone. It achieved an overall precision of 94.14%, recall of 91.90% and F1 score of 93.01% for the screening algorithm and 95.46%, 97.81% and 96.62% for the species differentiation algorithm respectively. This innovative solution has the potential to support filariasis diagnosis and monitoring, particularly in resource-limited settings where access to expert technicians and laboratory equipment is scarce.

Wavelet-based image fusion in multi-view three-dimensional microscopy.

MOTIVATION: Multi-view microscopy techniques such as Light-Sheet Fluorescence Microscopy (LSFM) are powerful tools for 3D + time studies of live embryos in developmental biology. The sample is imaged from several points of view, acquiring a set of 3D views that are then combined or fused in order to overcome their individual limitations. Views fusion is still an open problem despite recent contributions in the field. RESULTS: We developed a wavelet-based multi-view fusion method that, due to wavelet decomposition properties, is able to combine the complementary directional information from all available views into a single volume. Our method is demonstrated on LSFM acquisitions from live sea urchin and zebrafish embryos. The fusion results show improved overall contrast and details when compared with any of the acquired volumes. The proposed method does not need knowledge of the system's point spread function (PSF) and performs better than other existing PSF independent fusion methods. AVAILABILITY AND IMPLEMENTATION: The described method was implemented in Matlab (The Mathworks, Inc., USA) and a graphic user interface was developed in Java. The software, together with two sample datasets, is available at http://www.die.upm.es/im/software/SPIMFusionGUI.zip A public release, free of charge for non-commercial use, is planned after the publication of this article.

Digital Platform for Automatic Qualitative and Quantitative Reading of a Cryptococcal Antigen Point-of-Care Assay Leveraging Smartphones and Artificial Intelligence.

Cryptococcosis is a fungal infection that causes serious illness, particularly in immunocompromised individuals such as people living with HIV. Point of care tests (POCT) can help identify and diagnose patients with several advantages including rapid results and ease of use. The cryptococcal antigen (CrAg) lateral flow assay (LFA) has demonstrated excellent performance in diagnosing cryptococcosis, and it is particularly useful in resource-limited settings where laboratory-based tests may not be readily available. The use of artificial intelligence (AI) for the interpretation of rapid diagnostic tests can improve the accuracy and speed of test results, as well as reduce the cost and workload of healthcare professionals, reducing subjectivity associated with its interpretation. In this work, we analyze a smartphone-based digital system assisted by AI to automatically interpret CrAg LFA as well as to estimate the antigen concentration in the strip. The system showed excellent performance for predicting LFA qualitative interpretation with an area under the receiver operating characteristic curve of 0.997. On the other hand, its potential to predict antigen concentration based solely on a photograph of the LFA has also been demonstrated, finding a strong correlation between band intensity and antigen concentration, with a Pearson correlation coefficient of 0.953. The system, which is connected to a cloud web platform, allows for case identification, quality control, and real-time monitoring.

Evaluation of the Performance of a 3D-Printed Smartphone-Based Retinal Imaging Device as a Screening Tool for Retinal Pathology in Mozambique.

Low-income countries carry approximately 90% of the global burden of visual impairment, and up to 80% of this could be prevented or cured. However, there are only a few studies on the prevalence of retinal disease in these countries. Easier access to retinal information would allow differential diagnosis and promote strategies to improve eye health, which are currently scarce. This pilot study aims to evaluate the functionality and usability of a tele-retinography system for the detection of retinal pathology, based on a low-cost portable retinal scanner, manufactured with 3D printing and controlled by a mobile phone with an application designed ad hoc. The study was conducted at the Manhiça Rural Hospital in Mozambique. General practitioners, with no specific knowledge of ophthalmology or previous use of retinography, performed digital retinographies on 104 hospitalized patients. The retinographies were acquired in video format, uploaded to a web platform, and reviewed centrally by two ophthalmologists, analyzing the image quality and the presence of retinal lesions. In our sample there was a high proportion of exudates and hemorrhages-8% and 4%, respectively. In addition, the presence of lesions was studied in patients with known underlying risk factors for retinal disease, such as HIV, diabetes, and/or hypertension. Our tele-retinography system based on a smartphone coupled with a simple and low-cost 3D printed device is easy to use by healthcare personnel without specialized ophthalmological knowledge and could be applied for the screening and initial diagnosis of retinal pathology.

Crowdsourcing malaria parasite quantification: an online game for analyzing images of infected thick blood smears.

BACKGROUND: There are 600,000 new malaria cases daily worldwide. The gold standard for estimating the parasite burden and the corresponding severity of the disease consists in manually counting the number of parasites in blood smears through a microscope, a process that can take more than 20 minutes of an expert microscopist's time. OBJECTIVE: This research tests the feasibility of a crowdsourced approach to malaria image analysis. In particular, we investigated whether anonymous volunteers with no prior experience would be able to count malaria parasites in digitized images of thick blood smears by playing a Web-based game. METHODS: The experimental system consisted of a Web-based game where online volunteers were tasked with detecting parasites in digitized blood sample images coupled with a decision algorithm that combined the analyses from several players to produce an improved collective detection outcome. Data were collected through the MalariaSpot website. Random images of thick blood films containing Plasmodium falciparum at medium to low parasitemias, acquired by conventional optical microscopy, were presented to players. In the game, players had to find and tag as many parasites as possible in 1 minute. In the event that players found all the parasites present in the image, they were presented with a new image. In order to combine the choices of different players into a single crowd decision, we implemented an image processing pipeline and a quorum algorithm that judged a parasite tagged when a group of players agreed on its position. RESULTS: Over 1 month, anonymous players from 95 countries played more than 12,000 games and generated a database of more than 270,000 clicks on the test images. Results revealed that combining 22 games from nonexpert players achieved a parasite counting accuracy higher than 99%. This performance could be obtained also by combining 13 games from players trained for 1 minute. Exhaustive computations measured the parasite counting accuracy for all players as a function of the number of games considered and the experience of the players. In addition, we propose a mathematical equation that accurately models the collective parasite counting performance. CONCLUSIONS: This research validates the online gaming approach for crowdsourced counting of malaria parasites in images of thick blood films. The findings support the conclusion that nonexperts are able to rapidly learn how to identify the typical features of malaria parasites in digitized thick blood samples and that combining the analyses of several users provides similar parasite counting accuracy rates as those of expert microscopists. This experiment illustrates the potential of the crowdsourced gaming approach for performing routine malaria parasite quantification, and more generally for solving biomedical image analysis problems, with future potential for telediagnosis related to global health challenges.

Comparing sources of mobility for modelling the epidemic spread of Zika virus in Colombia.

Timely, accurate, and comparative data on human mobility is of paramount importance for epidemic preparedness and response, but generally not available or easily accessible. Mobile phone metadata, typically in the form of Call Detail Records (CDRs), represents a powerful source of information on human movements at an unprecedented scale. In this work, we investigate the potential benefits of harnessing aggregated CDR-derived mobility to predict the 2015-2016 Zika virus (ZIKV) outbreak in Colombia, when compared to other traditional data sources. To simulate the spread of ZIKV at sub-national level in Colombia, we employ a stochastic metapopulation epidemic model for vector-borne diseases. Our model integrates detailed data on the key drivers of ZIKV spread, including the spatial heterogeneity of the mosquito abundance, and the exposure of the population to the virus due to environmental and socio-economic factors. Given the same modelling settings (i.e. initial conditions and epidemiological parameters), we perform in-silico simulations for each mobility network and assess their ability in reproducing the local outbreak as reported by the official surveillance data. We assess the performance of our epidemic modelling approach in capturing the ZIKV outbreak both nationally and sub-nationally. Our model estimates are strongly correlated with the surveillance data at the country level (Pearson's r = 0.92 for the CDR-informed network). Moreover, we found strong performance of the model estimates generated by the CDR-informed mobility networks in reproducing the local outbreak observed at the sub-national level. Compared to the CDR-informed networks, the performance of the other mobility networks is either comparatively similar or substantially lower, with no added value in predicting the local epidemic. This suggests that mobile phone data captures a better picture of human mobility patterns. This work contributes to the ongoing discussion on the value of aggregated mobility estimates from CDRs data that, with appropriate data protection and privacy safeguards, can be used for social impact applications and humanitarian action.

Deep learning-based lesion subtyping and prediction of clinical outcomes in COVID-19 pneumonia using chest CT.

The main objective of this work is to develop and evaluate an artificial intelligence system based on deep learning capable of automatically identifying, quantifying, and characterizing COVID-19 pneumonia patterns in order to assess disease severity and predict clinical outcomes, and to compare the prediction performance with respect to human reader severity assessment and whole lung radiomics. We propose a deep learning based scheme to automatically segment the different lesion subtypes in nonenhanced CT scans. The automatic lesion quantification was used to predict clinical outcomes. The proposed technique has been independently tested in a multicentric cohort of 103 patients, retrospectively collected between March and July of 2020. Segmentation of lesion subtypes was evaluated using both overlapping (Dice) and distance-based (Hausdorff and average surface) metrics, while the proposed system to predict clinically relevant outcomes was assessed using the area under the curve (AUC). Additionally, other metrics including sensitivity, specificity, positive predictive value and negative predictive value were estimated. 95% confidence intervals were properly calculated. The agreement between the automatic estimate of parenchymal damage (%) and the radiologists' severity scoring was strong, with a Spearman correlation coefficient (R) of 0.83. The automatic quantification of lesion subtypes was able to predict patient mortality, admission to the Intensive Care Units (ICU) and need for mechanical ventilation with an AUC of 0.87, 0.73 and 0.68 respectively. The proposed artificial intelligence system enabled a better prediction of those clinically relevant outcomes when compared to the radiologists' interpretation and to whole lung radiomics. In conclusion, deep learning lesion subtyping in COVID-19 pneumonia from noncontrast chest CT enables quantitative assessment of disease severity and better prediction of clinical outcomes with respect to whole lung radiomics or radiologists' severity score.

Remote analysis of sputum smears for mycobacterium tuberculosis quantification using digital crowdsourcing.

Worldwide, TB is one of the top 10 causes of death and the leading cause from a single infectious agent. Although the development and roll out of Xpert MTB/RIF has recently become a major breakthrough in the field of TB diagnosis, smear microscopy remains the most widely used method for TB diagnosis, especially in low- and middle-income countries. This research tests the feasibility of a crowdsourced approach to tuberculosis image analysis. In particular, we investigated whether anonymous volunteers with no prior experience would be able to count acid-fast bacilli in digitized images of sputum smears by playing an online game. Following this approach 1790 people identified the acid-fast bacilli present in 60 digitized images, the best overall performance was obtained with a specific number of combined analysis from different players and the performance was evaluated with the F1 score, sensitivity and positive predictive value, reaching values of 0.933, 0.968 and 0.91, respectively.

A Smartphone-Based Platform Assisted by Artificial Intelligence for Reading and Reporting Rapid Diagnostic Tests: Evaluation Study in SARS-CoV-2 Lateral Flow Immunoassays.

BACKGROUND: Rapid diagnostic tests (RDTs) are being widely used to manage COVID-19 pandemic. However, many results remain unreported or unconfirmed, altering a correct epidemiological surveillance. OBJECTIVE: Our aim was to evaluate an artificial intelligence-based smartphone app, connected to a cloud web platform, to automatically and objectively read RDT results and assess its impact on COVID-19 pandemic management. METHODS: Overall, 252 human sera were used to inoculate a total of 1165 RDTs for training and validation purposes. We then conducted two field studies to assess the performance on real-world scenarios by testing 172 antibody RDTs at two nursing homes and 96 antigen RDTs at one hospital emergency department. RESULTS: Field studies demonstrated high levels of sensitivity (100%) and specificity (94.4%, CI 92.8%-96.1%) for reading IgG band of COVID-19 antibody RDTs compared to visual readings from health workers. Sensitivity of detecting IgM test bands was 100%, and specificity was 95.8% (CI 94.3%-97.3%). All COVID-19 antigen RDTs were correctly read by the app. CONCLUSIONS: The proposed reading system is automatic, reducing variability and uncertainty associated with RDTs interpretation and can be used to read different RDT brands. The web platform serves as a real-time epidemiological tracking tool and facilitates reporting of positive RDTs to relevant health authorities.

Epidemiological modelling in refugee and internally displaced people settlements: challenges and ways forward.

The spread of infectious diseases such as COVID-19 presents many challenges to healthcare systems and infrastructures across the world, exacerbating inequalities and leaving the world's most vulnerable populations at risk. Epidemiological modelling is vital to guiding evidence-informed or data-driven decision making. In forced displacement contexts, and in particular refugee and internally displaced people (IDP) settlements, it meets several challenges including data availability and quality, the applicability of existing models to those contexts, the accurate modelling of cultural differences or specificities of those operational settings, the communication of results and uncertainties, as well as the alignment of strategic goals between diverse partners in complex situations. In this paper, we systematically review the limited epidemiological modelling work applied to refugee and IDP settlements so far, and discuss challenges and identify lessons learnt from the process. With the likelihood of disease outbreaks expected to increase in the future as more people are displaced due to conflict and climate change, we call for the development of more approaches and models specifically designed to include the unique features and populations of refugee and IDP settlements. To strengthen collaboration between the modelling and the humanitarian public health communities, we propose a roadmap to encourage the development of systems and frameworks to share needs, build tools and coordinate responses in an efficient and scalable manner, both for this pandemic and for future outbreaks.

Combining collective and artificial intelligence for global health diseases diagnosis using crowdsourced annotated medical images.

Visual inspection of microscopic samples is still the gold standard diagnostic methodology for many global health diseases. Soil-transmitted helminth infection affects 1.5 billion people worldwide, and is the most prevalent disease among the Neglected Tropical Diseases. It is diagnosed by manual examination of stool samples by microscopy, which is a time-consuming task and requires trained personnel and high specialization. Artificial intelligence could automate this task making the diagnosis more accessible. Still, it needs a large amount of annotated training data coming from experts.In this work, we proposed the use of crowdsourced annotated medical images to train AI models (neural networks) for the detection of soil-transmitted helminthiasis in microscopy images from stool samples leveraging non-expert knowledge collected through playing a video game. We collected annotations made by both school-age children and adults, and we showed that, although the quality of crowdsourced annotations made by school-age children are sightly inferior than the ones made by adults, AI models trained on these crowdsourced annotations perform similarly (AUC of 0.928 and 0.939 respectively), and reach similar performance to the AI model trained on expert annotations (AUC of 0.932). We also showed the impact of the training sample size and continuous training on the performance of the AI models.In conclusion, the workflow proposed in this work combined collective and artificial intelligence for detecting soil-transmitted helminthiasis. Embedded within a digital health platform can be applied to any other medical image analysis task and contribute to reduce the burden of disease.