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 microfilarias 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.

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.

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.

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.

Mobile microscopy and telemedicine platform assisted by deep learning for the quantification of Trichuris trichiura infection

Soil-transmitted helminths (STH) are the most prevalent pathogens among the group of neglected tropical diseases (NTDs). The Kato-Katz technique is the diagnosis method recommended by the World Health Organization (WHO) although it often presents a decreased sensitivity in low transmission settings and it is labour intensive. Visual reading of Kato-Katz preparations requires the samples to be analyzed in a short period of time since its preparation. Digitizing the samples could provide a solution which allows to store the samples in a digital database and perform remote analysis. Artificial intelligence (AI) methods based on digitized samples can support diagnosis by performing an objective and automatic quantification of disease infection. In this work, we propose an end-to-end pipeline for microscopy image digitization and automatic analysis of digitized images of STH. Our solution includes (a) a digitization system based on a mobile app that digitizes microscope samples using a 3D printed microscope adapter, (b) a telemedicine platform for remote analysis and labelling, and (c) novel deep learning algorithms for automatic assessment and quantification of parasitological infections by STH. The deep learning algorithm has been trained and tested on 51 slides of stool samples containing 949 Trichuris spp. eggs from 6 different subjects. The algorithm evaluation was performed using a cross-validation strategy, obtaining a mean precision of 98.44% and a mean recall of 80.94%. The results also proved the potential of generalization capability of the method at identifying different types of helminth eggs. Additionally, the AI-assisted quantification of STH based on digitized samples has been compared to the one performed using conventional microscopy, showing a good agreement between measurements. In conclusion, this work has presented a comprehensive pipeline using smartphone-assisted microscopy. It is integrated with a telemedicine platform for automatic image analysis and quantification of STH infection using AI models.

Diagnóstico molecular de parasitosis intestinales

Infections causes by parasites of the gastrointestinal tract are a global public health problem. In industrialised countries, their particular epidemiological (low general prevalence of enteroparasites), economic (high labour costs) and clinical characteristics (constant increase in the number of samples and diagnostic determinations to be performed) have led molecular techniques to progressively replace conventional microscopy as the first-line diagnostic method of these pathogens in modern clinical laboratories. PCR-based techniques, particularly those developed for the simultaneous detection of the various agents that can cause the same infectious disease (syndromic diagnosis), already represent a cost-effective option that allow process automisation, workflow optimisation, and comparison of results among different laboratories, and facilitate accreditation of diagnostic procedures. This review clearly and concisely discusses the current situation of the molecular diagnosis of the main species of intestinal parasites in humans, particularly the enteric protozoans causing diarrhoea (Cryptosporidium spp., Giardia duodenalis, Entamoeba histolytica), the most important members the Microsporidia phyla (Enterocytozoon bieneusi) and Stramenopiles phyla (Blastocystis sp.), as well as the helminths transmitted by soil (Ancylostoma spp., Ascaris lumbricoides, Necator americanus, Strongyloides stercoralis and Trichuris trichiura) and food (Anisakis spp., Clonorchis sinensis, Fasciola spp., Taenia solium, and Trichinella spiralis). Special attention is paid to the description of available techniques and formats, to their diagnostic benefits and the most widely used genetic markers for their detection, both in clinical laboratories and genotyping in referral and research centres.

Epidemiology of intestinal helminthiases in a rural community of Ethiopia: Is it time to expand control programs to include Strongyloides stercoralis and the entire community?

BACKGROUND

Soil transmitted helminths are highly prevalent worldwide. Globally, approximately 1.5 billion people are infected with Ascaris lumbricoides, Trichuris trichiura or hookworm. Endemic countries carry out periodic mass treatment of at-risk populations with albendazole or mebendazole as a control measure. Most prevalence studies have focused on school aged children and therefore control programs are implemented at school level, not at community level. In this study, the prevalence of intestinal helminths, including Strongyloides stercoralis, was examined using a comprehensive laboratory approach in a community in north-western Ethiopia.

METHODS

A cross-sectional survey was conducted on 792 individuals ≥5 years old in randomly selected houses in a rural district. Stools were examined using three techniques: a formol-ether concentration, the Baermann technique and a real time polymerase chain reaction test (these last two specific for S. stercoralis). Statistical analyses were performed between two large age groups, children (≤14 years old) and adults (≥15 years old).

RESULTS

The prevalence of helminths was 91.3%; (95% CI: 89.3-93.3%). Hookworm was the most prevalent, 78.7% (95% CI 75.6-81.4%), followed by S. stercoralis 55.7% (95% CI 52.2-59.1%). Co-infection with both was detected in 45.4% (95% CI 42.0-49.0%) of the participants. The mean age of hookworm-infected individuals was significantly higher than non-infected ones (p = 0.003). Also, S. stercoralis infection was significantly associated with age, being more prevalent in adults (p = 0.002).

CONCLUSIONS

This is the highest prevalence of S. stercoralis detected in Ethiopia so far. Our results highlight the need of searching specifically for infection by this parasite since it usually goes unnoticed if helminth studies rely only on conventional diagnostic techniques, i.e. Kato-Katz. Moreover, the focus of these programs on children undermines the actual prevalence of hookworm. The adult population acts as a reservoir for both hookworm and S. stercoralis and this fact may negatively impact the current control programs in Ethiopia which only target treatment of school aged children. This reservoir, together with a lack of adequate water, sanitation and hygiene, increases the probability of re-infection in children. Finally, the high prevalence of S. stercoralis found calls for a comprehensive diagnostic approach in endemic areas in addition to a revision of control measures that is, adding ivermectin to current albendazole/mebendazole, since it is the drug of choice for S. stercoralis.

Risk associations for intestinal parasites in symptomatic and asymptomatic schoolchildren in central Mozambique

OBJECTIVES

Chronic infections by enteric parasites including protist and helminthic species produce long-term sequelae on the health status of infected children. This study assesses potential associations linked with enteric parasite infections in symptomatic and asymptomatic children in Zambézia province, Mozambique.

METHODS

In this prospective cross-sectional study, stool samples and epidemiological questionnaires on demographics and risk associations were collected from symptomatic children (n = 286) from clinical settings and asymptomatic (n = 807) children from 17 schools and creches aged 3‒14 years. We detected enteric parasites using PCR-based methods. We calculated prevalence (adjusted for age, sex, house construction, drinking water, and latrine use) and odds ratios (ORs) for risk associations with logistic regression, after adjusting for district, neighbourhood and symptoms.

RESULTS

Numbers and adjusted prevalence (95% confidence intervals in parentheses) for the symptomatic and asymptomatic populations were Giardia duodenalis 120, 52% (22-82), 339, 42% (25-59); followed by Strongyloides stercoralis 52, 14% (9‒20), 180, 20% (15-25). Risk associations for G. duodenalis included drinking untreated river/spring water, OR 2.91 (1.80-4.70); contact with ducks, OR 14.96 (2.93‒76.31); dogs, OR 1.92 (1.04-3.52); cats, OR 1.73 (1.16-2.59), and a relative with diarrhoea, OR 2.59 (1.54‒4.37). Risk associations for S. stercoralis included having no latrine, OR 2.41 (1.44-4.02); drinking well water, OR 1.82 (1.02-3.25), and increasing age, OR 1.11 (1.04-1.20).

CONCLUSIONS

We found a high prevalence of intestinal parasites regardless of the children's symptoms. Drinking well or river water, domestic animals, and latrine absence were contributing factors of human infections.

Protist enteroparasites in wild boar (Sus scrofa ferus) and black Iberian pig (Sus scrofa domesticus) in southern Spain: a protective effect on hepatitis E acquisition?

BACKGROUND

Several studies have independently evaluated the occurrence of hepatitis E virus (HEV) and enteroparasites in swine, but no surveys have been conducted to jointly assess the prevalence and genetic diversity of enteroparasites in pigs and wild boars, their sympatric transmission between hosts, and their potential interaction with HEV.

METHODS

We prospectively collected serum and faecal samples from black Iberian domestic pigs and wild boars from southern Spain between 2015‒2016. We evaluated for HEV in serum and faeces, and for the presence of enteroparasites (Giardia duodenalis, Cryptosporidium spp., Blastocystis sp., Neobalantidium coli and Strongyloides spp.) in the same faecal samples. The prevalence of each intestinal parasite species was calculated.

RESULTS

A total of 328 animals (56.7% black Iberian pigs and 43.3% wild boars) were included in the study. The overall global prevalence of HEV in serum was 16.8%. The overall global prevalence of each enteroparasite species was 19.5% for G. duodenalis, 8.2% for Cryptosporidium spp., 41.8% for Blastocystis sp., 31.4% for N. coli, and 8.8% for Strongyloides spp. HEV-infected animals showed a significantly lower prevalence of G. duodenalis (3.2 vs 20%; P = 0.002) and Blastocystis sp. (38.7 vs 80%; P < 0.001) than those uninfected by HEV. Animals carrying G. duodenalis and Blastocystis sp. infections showed a significantly lower rate of HEV infection than those not harbouring these enteroparasites (P < 0.001).

CONCLUSIONS

Our study found a high prevalence of enteroparasites in black Iberian pigs and wild boars in southern Spain, suggesting a sympatric co-transmission of some of the species investigated. It is suggested that extracellular G. duodenalis and Blastocystis sp. might have a protective effect on HEV acquisition in swine.

Exploring interactions between Blastocystis sp., Strongyloides spp. and the gut microbiomes of wild chimpanzees in Senegal

BACKGROUND

Gut parasites exert an important influence on the gut microbiome, with many studies focusing on the human gut microbiome. It has, however, undergone severe richness depletion. Hygienic lifestyle, antimicrobial treatments and altered gut homeostasis (e.g., chronic inflammation) reduce gut microbiome richness and also parasite prevalence; which may confound results. Studying species closely related to humans could help overcome this problem by providing insights into the ancestral relationship between humans, their gut microbiome and their gut parasites. Chimpanzees are a particularly promising model as they have similar gut microbiomes to humans and many parasites infect both species.

AIMS

We study the interaction between gut microbiome and enteric parasites in chimpanzees. Investigating what novel insights a closely related species can reveal when compared to studies on humans.

METHODS

Using eighty-seven faecal samples from wild western chimpanzees (Pan troglodytes verus) in Senegal, we combine 16S rRNA gene amplicon sequencing for gut microbiome characterization with PCR detection of parasite taxa (Blastocystis sp., Strongyloides spp., Giardia duodenalis, Cryptosporidium spp., Plasmodium spp., Filariae and Trypanosomatidae). We test for differences in gut microbiota ecosystem traits and taxonomical composition between Blastocystis and Strongyloides bearing and non-bearing samples.

RESULTS

For Blastocystis, twelve differentially abundant taxa (e.g., Methanobrevibacter), including Prevotella and Ruminococcus-Methanobrevibacter enterotype markers, replicate findings in humans. However, several richness indices are lower in Blastocystis carriers, contradicting human studies. This indicates Blastocystis, unlike Strongyloides, is associated to a "poor health" gut microbiome, as does the fact that Faecalibacterium, a bacterium with gut protective traits, is absent in Blastocystis-positive samples. Strongyloides was associated to Alloprevotella and five other taxonomic groups. Each parasite had its unique impact on the gut microbiota indicating parasite-specific niches. Our results suggest that studying the gut microbiomes of wild chimpanzees could help disentangle biological from artefactual associations between gut microbiomes and parasites.

Field and laboratory comparative evaluation of a LAMP assay for the diagnosis of urogenital schistosomiasis in Cubal, Central Angola

OBJECTIVE

To evaluate the performance of Rapid-Heat LAMPellet assay in field conditions for diagnosis of urogenital schistosomiasis in an endemic area in Cubal, Angola, and to assess the reproducibility in a reference laboratory.

METHODS

A total of 172 urine samples from school-age children were tested for microhaematuria, microscopic detection of Schistosoma haematobium eggs and LAMP for DNA detection. Urine samples were stored in a basic equipped laboratory. Field-LAMP tests were performed with and without prior DNA extraction from urine samples, and the results were read by turbidity and by colour change. When field procedures were finished, samples were sent to a reference laboratory to be reanalysed by LAMP.

RESULTS

A total of 83 of 172 (48.3%) were positive for microhaematuria, 87/172 (50.6%) were microscopy-positive for S. haematobium eggs detection, and 127/172 (73.8%) showed LAMP-positive results for detecting S. haematobium using purified DNA and 109/172 (63.4%) without prior DNA extraction. MacNemar's test showed a statistical significant relation between LAMP results and microscopy-detected S. haematobium infections and microhaematuria (P < 0.001 in both cases), respectively. When samples of purified DNA were reanalysed in a reference laboratory in Spain using the same LAMP methodology, the overall reproducibility achieved 72.1%.

CONCLUSIONS

The ease of use, simplicity and feasibility demonstrated by LAMP assay in field conditions together with the acceptable level of reproducibility achieved in a reference laboratory support the use of LAMP assay as an effective test for molecular diagnosis of urogenital schistosomiasis in endemic remote areas.

Prevalence and molecular characterization of Strongyloides stercoralis, Giardia duodenalis, Cryptosporidium spp., and Blastocystis spp. isolates in school children in Cubal, Western Angola

BACKGROUND

Human infections by the gastrointestinal helminth Strongyloides stercoralis and the enteric protozoans Giardia duodenalis, Cryptosporidium spp. and Blastocystis spp. are not formally included in the list of 20 neglected tropical diseases prioritised by the World Health Organization. Although largely underdiagnosed and considered of lower public health relevance, these infections have been increasingly demonstrated to cause significant morbidity and even mortality globally, particularly among children living in resource-poor settings.

METHODS

In this cross-sectional survey the prevalence, frequency and molecular diversity of S. stercoralis, G. duodenalis, Cryptosporidium spp. and Blastocystis spp. were investigated in a school children population in the province of Benguela (Angola). A total of 351 stool samples were collected during January to June 2015. The presence of S. stercoralis and G. duodenalis was confirmed by qPCR methods. Giardia duodenalis assemblages and sub-assemblages were determined by multilocus sequence-based genotyping of the glutamate dehydrogenase and β-giardin genes of the parasite. Detection and identification of Cryptosporidium and Blastocystis species and subtypes was carried out by amplification and sequencing of a partial fragment of the small-subunit ribosomal RNA gene of both protozoan. Analyses of risk factors potentially associated with the transmission of these pathogens were also conducted.

RESULTS

Prevalences of S. stercoralis, G. duodenalis, Cryptosporidium spp., and Blastocystis spp. were estimated at 21.4% (95% CI: 17.1-25.7%), 37.9% (95% CI: 32.8-43.0%), 2.9% (95% CI: 1.1-4.5%) and 25.6% (95% CI: 21.18-30.2%), respectively. Overall, 64.1% (225/351) of the children were infected by at least one of the pathogens investigated. Sequence analyses of the 28 G. duodenalis isolates that were successfully genotyped allowed the identification of sub-assemblages AI (14.3%), AII (14.3%), BIII (7.1%) and BIV (25.0%). Discordant typing results AII/AIII and BIII/BIV were identified in 7.1% and 14.3% of the isolates, respectively. A total of five additional isolates (17.9%) were identified as assemblage B. Three Cryptosporidium species including C. hominis (70%), C. parvum (20%) and C. canis (10%) were found circulating in the children population under study. A total of 75 Blastocystis isolates were assigned to the subtypes ST1 (30.7%), ST2 (30.7%), ST3 (36.0%), ST5 (1.3%) and ST7 (1.3%), respectively. Children younger than seven years of age had significantly higher risk of infections by protozoan enteropathogens (PRR: 1.35, P < 0.01), whereas being underweight seemed to have a protective effect against these infections (PRR: 0.74, P = 0.005).

CONCLUSIONS

The burden of disease attributable to human strongyloidiasis, giardiosis, cryptosporidiosis and blastocystosis in Angola is considerably higher than initially estimated in previous surveys. Surveillance and control of these infections should be jointly tackled with formally considered neglected tropical diseases in order to maximize effort and available resources. Our data also demonstrate the added value of using molecular diagnostic methods in high transmission areas.