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

Harnessing artificial intelligence microscopy to improve diagnostics for soil-transmitted helminthiasis and schistosomiasis: a review of recent advances and future pathways

PURPOSE OF REVIEW

This opinion piece aims to explore the transformative potential of integrating artificial intelligence with digital microscopy to enhance diagnostics for soil-transmitted helminthiasis (STH) and schistosomiasis (SCH), two pervasive neglected tropical diseases (NTDs). By aligning innovative artificial intelligence-driven solutions with WHO's strategic objectives and calls for better, more accessible, and more integrated diagnostics, we highlight the latest advancements that may support improved health outcomes in affected communities.

RECENT FINDINGS

The review covers recent advancements in artificial intelligence-based diagnostic technologies, emphasizing automated egg detection and quantification. These technologies promise to mitigate challenges such as human error and the need for skilled technicians.

SUMMARY

The findings have significant implications for public health, ethical considerations and regulatory pathways, particularly in resource-limited settings. The authors advocate for interdisciplinary collaboration and a strategic focus on meeting WHO target product profiles to ensure uptake, ultimately to support reaching WHO NTD targets.

Innovative strategies for the surveillance, prevention, and management of pediatric infections applied to low-income settings

INTRODUCTION

Infectious diseases still cause a significant burden of morbidity and mortality among children in low- and middle-income countries (LMICs). There are ample opportunities for innovation in surveillance, prevention, and management, with the ultimate goal of improving survival.

AREAS COVERED

This review discusses the current status in the use and development of innovative strategies for pediatric infectious diseases in LMICs by focusing on surveillance, diagnosis, prevention, and management. Topics covered are: Minimally Invasive Tissue Sampling as a technique to accurately ascertain the cause of death; Genetic Surveillance to trace the pathogen genomic diversity and emergence of resistance; Artificial Intelligence as a multidisciplinary tool; Portable noninvasive imaging methods; and Prognostic Biomarkers to triage and risk stratify pediatric patients.

EXPERT OPINION

To overcome the specific hurdles in child health for LMICs, some innovative strategies appear at the forefront of research. If the development of these next-generation tools remains focused on accessibility, sustainability and capacity building, reshaping epidemiological surveillance, diagnosis, and treatment in LMICs, can become a reality and result in a significant public health impact. Their integration with existing healthcare infrastructures may revolutionize disease detection and surveillance, and improve child health and survival.

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.

Diagnosis of soil-transmitted helminth infections with digital mobile microscopy and artificial intelligence in a resource-limited setting

BACKGROUND

Infections caused by soil-transmitted helminths (STHs) are the most prevalent neglected tropical diseases and result in a major disease burden in low- and middle-income countries, especially in school-aged children. Improved diagnostic methods, especially for light intensity infections, are needed for efficient, control and elimination of STHs as a public health problem, as well as STH management. Image-based artificial intelligence (AI) has shown promise for STH detection in digitized stool samples. However, the diagnostic accuracy of AI-based analysis of entire microscope slides, so called whole-slide images (WSI), has previously not been evaluated on a sample-level in primary healthcare settings in STH endemic countries.

METHODOLOGY/PRINCIPAL FINDINGS

Stool samples (n = 1,335) were collected during 2020 from children attending primary schools in Kwale County, Kenya, prepared according to the Kato-Katz method at a local primary healthcare laboratory and digitized with a portable whole-slide microscopy scanner and uploaded via mobile networks to a cloud environment. The digital samples of adequate quality (n = 1,180) were split into a training (n = 388) and test set (n = 792) and a deep-learning system (DLS) developed for detection of STHs. The DLS findings were compared with expert manual microscopy and additional visual assessment of the digital samples in slides with discordant results between the methods. Manual microscopy detected 15 (1.9%) Ascaris lumbricoides, 172 (21.7%) Tricuris trichiura and 140 (17.7%) hookworm (Ancylostoma duodenale or Necator americanus) infections in the test set. Importantly, more than 90% of all STH positive cases represented light intensity infections. With manual microscopy as the reference standard, the sensitivity of the DLS as the index test for detection of A. lumbricoides, T. trichiura and hookworm was 80%, 92% and 76%, respectively. The corresponding specificity was 98%, 90% and 95%. Notably, in 79 samples (10%) classified as negative by manual microscopy for a specific species, STH eggs were detected by the DLS and confirmed correct by visual inspection of the digital samples.

CONCLUSIONS/SIGNIFICANCE

Analysis of digitally scanned stool samples with the DLS provided high diagnostic accuracy for detection of STHs. Importantly, a substantial number of light intensity infections were missed by manual microscopy but detected by the DLS. Thus, analysis of WSIs with image-based AI may provide a future tool for improved detection of STHs in a primary healthcare setting, which in turn could facilitate monitoring and evaluation of control programs.

A high-speed microscopy system based on deep learning to detect yeast-like fungi cells in blood

Background: Blood-invasive fungal infections can cause the death of patients, while diagnosis of fungal infections is challenging. Methods: A high-speed microscopy detection system was constructed that included a microfluidic system, a microscope connected to a high-speed camera and a deep learning analysis section. Results: For training data, the sensitivity and specificity of the convolutional neural network model were 93.5% (92.7-94.2%) and 99.5% (99.1-99.5%), respectively. For validating data, the sensitivity and specificity were 81.3% (80.0-82.5%) and 99.4% (99.2-99.6%), respectively. Cryptococcal cells were found in 22.07% of blood samples. Conclusion: This high-speed microscopy system can analyze fungal pathogens in blood samples rapidly with high sensitivity and specificity and can help dramatically accelerate the diagnosis of fungal infectious diseases.

Neglected tropical disease (NTD) diagnostics: current development and operations to advance control

Neglected tropical diseases (NTDs) have become important public health threats that require multi-faceted control interventions. As late treatment and management of NTDs contribute significantly to the associated burdens, early diagnosis becomes an important component for surveillance and planning effective interventions. This review identifies common NTDs and highlights the progress in the development of diagnostics for these NTDs. Leveraging existing technologies to improve NTD diagnosis and improving current operational approaches for deployment of developed diagnostics are crucial to achieving the 2030 NTD elimination target. Point-of-care NTD (POC-NTD) diagnostic tools are recommended preferred diagnostic options in resource-constrained areas for mapping risk zones and monitoring treatment efficacy. However, few are currently available commercially. Technical training of remote health care workers on the use of POC-NTD diagnostics, and training of health workers on the psychosocial consequences of these diagnostics are critical in harnessing POC-NTD diagnostic potential. While the COVID-19 pandemic has challenged the possibility of achieving NTD elimination in 2030 due to the disruption of healthcare services and dwindling financial support for NTDs, the possible contribution of NTDs in exacerbating COVID-19 pandemic should motivate NTD health system strengthening.

Validation of artificial intelligence-based digital microscopy for automated detection of Schistosoma haematobium eggs in urine in Gabon

INTRODUCTION

Schistosomiasis is a significant public health concern, especially in Sub-Saharan Africa. Conventional microscopy is the standard diagnostic method in resource-limited settings, but with limitations, such as the need for expert microscopists. An automated digital microscope with artificial intelligence (Schistoscope), offers a potential solution. This field study aimed to validate the diagnostic performance of the Schistoscope for detecting and quantifying Schistosoma haematobium eggs in urine compared to conventional microscopy and to a composite reference standard (CRS) consisting of real-time PCR and the up-converting particle (UCP) lateral flow (LF) test for the detection of schistosome circulating anodic antigen (CAA).

METHODS

Based on a non-inferiority concept, the Schistoscope was evaluated in two parts: study A, consisting of 339 freshly collected urine samples and study B, consisting of 798 fresh urine samples that were also banked as slides for analysis with the Schistoscope. In both studies, the Schistoscope, conventional microscopy, real-time PCR and UCP-LF CAA were performed and samples with all the diagnostic test results were included in the analysis. All diagnostic procedures were performed in a laboratory located in a rural area of Gabon, endemic for S. haematobium.

RESULTS

In study A and B, the Schistoscope demonstrated a sensitivity of 83.1% and 96.3% compared to conventional microscopy, and 62.9% and 78.0% compared to the CRS. The sensitivity of conventional microscopy in study A and B compared to the CRS was 61.9% and 75.2%, respectively, comparable to the Schistoscope. The specificity of the Schistoscope in study A (78.8%) was significantly lower than that of conventional microscopy (96.4%) based on the CRS but comparable in study B (90.9% and 98.0%, respectively).

CONCLUSION

Overall, the performance of the Schistoscope was non-inferior to conventional microscopy with a comparable sensitivity, although the specificity varied. The Schistoscope shows promising diagnostic accuracy, particularly for samples with moderate to higher infection intensities as well as for banked sample slides, highlighting the potential for retrospective analysis in resource-limited settings.

TRIAL REGISTRATION

NCT04505046 ClinicalTrials.gov.

Automatic detection of Opisthorchis viverrini egg in stool examination using convolutional-based neural networks

Background

Human opisthorchiasis is a dangerous infectious chronic disease distributed in many Asian areas in the water-basins of large rivers, Siberia, and Europe. The gold standard for human opisthorchiasis laboratory diagnosis is the routine examination of Opisthorchis spp. eggs under a microscope. Manual detection is laborious, time-consuming, and dependent on the microscopist's abilities and expertise. Automatic screening of Opisthorchis spp. eggs with deep learning techniques is a useful diagnostic aid.

Methods

Herein, we propose a convolutional neural network (CNN) for classifying and automatically detecting O. viverrini eggs from digitized images. The image data acquisition was acquired from infected human feces and was processed using the gold standard formalin ethyl acetate concentration technique, and then captured under the microscope digital camera at 400x. Microscopic images containing artifacts and O.viverrini egg were augmented using image rotation, filtering, noising, and sharpening techniques. This augmentation increased the image dataset from 1 time to 36 times in preparation for the training and validation step. Furthermore, the overall dataset was subdivided into a training-validation and test set at an 80:20 ratio, trained with a five-fold cross-validation to test model stability. For model training, we customized a CNN for image classification. An object detection method was proposed using a patch search algorithm to detect eggs and their locations. A performance matrix was used to evaluate model efficiency after training and IoU analysis for object detection.

Results

The proposed model, initially trained on non-augmented data of artifacts (class 0) and O. viverrini eggs (class 1), showed limited performance with 50.0% accuracy, 25.0% precision, 50.0% recall, and a 33.0% F1-score. After implementing data augmentation, the model significantly improved, reaching 100% accuracy, precision, recall, and F1-score. Stability assessments using 5-fold cross-validation indicated better stability with augmented data, evidenced by an ROC-AUC metric improvement from 0.5 to 1.00. Compared to other models such as ResNet50, InceptionV3, VGG16, DenseNet121, and Xception, the proposed model, with a smaller file size of 2.7 MB, showed comparable perfect performance. In object detection, the augmented data-trained model achieved an IoU score over 0.5 in 139 out of 148 images, with an average IoU of 0.6947.

Conclusion

This study demonstrated the successful application of CNN in classifying and automating the detection of O. viverrini eggs in human stool samples. Our CNN model's performance metrics and true positive detection rates were outstanding. This innovative application of deep learning can automate and improve diagnostic precision, speed, and efficiency, particularly in regions where O. viverrini infections are prevalent, thereby possibly improving infection sustainable control and treatment program.

Artificial intelligence, ChatGPT, and other large language models for social determinants of health: Current state and future directions

This perspective highlights the importance of addressing social determinants of health (SDOH) in patient health outcomes and health inequity, a global problem exacerbated by the COVID-19 pandemic. We provide a broad discussion on current developments in digital health and artificial intelligence (AI), including large language models (LLMs), as transformative tools in addressing SDOH factors, offering new capabilities for disease surveillance and patient care. Simultaneously, we bring attention to challenges, such as data standardization, infrastructure limitations, digital literacy, and algorithmic bias, that could hinder equitable access to AI benefits. For LLMs, we highlight potential unique challenges and risks including environmental impact, unfair labor practices, inadvertent disinformation or "hallucinations," proliferation of bias, and infringement of copyrights. We propose the need for a multitiered approach to digital inclusion as an SDOH and the development of ethical and responsible AI practice frameworks globally and provide suggestions on bridging the gap from development to implementation of equitable AI technologies.

Toward automating the diagnosis of gastrointestinal parasites in cats and dogs

Diagnosing gastrointestinal parasites by microscopy slide examination often leads to human interpretation errors, which may occur due to fatigue, lack of training and infrastructure, presence of artifacts (e.g., various types of cells, algae, yeasts), and other reasons. We have investigated the stages in automating the process to cope with the interpretation errors. This work presents advances in two stages focused on gastrointestinal parasites of cats and dogs: a new parasitological processing technique, named TF-Test VetPet, and a microscopy image analysis pipeline based on deep learning methods. TF-Test VetPet improves image quality by reducing cluttering (i.e., eliminating artifacts), which favors automated image analysis. The proposed pipeline can identify three species of parasites in cats and five in dogs, distinguishing them from fecal impurities with an average accuracy of 98,6%. We also make available the two datasets with images of parasites of dogs and cats, which were obtained by processing fecal smears with temporary staining using TF-Test VetPet.

A general framework to support cost-efficient fecal egg count methods and study design choices for large-scale STH deworming programs-monitoring of therapeutic drug efficacy as a case study

BACKGROUND

Soil-transmitted helminth (STH) control programs currently lack evidence-based recommendations for cost-efficient survey designs for monitoring and evaluation. Here, we present a framework to provide evidence-based recommendations, using a case study of therapeutic drug efficacy monitoring based on the examination of helminth eggs in stool.

METHODS

We performed an in-depth analysis of the operational costs to process one stool sample for three diagnostic methods (Kato-Katz, Mini-FLOTAC and FECPAKG2). Next, we performed simulations to determine the probability of detecting a truly reduced therapeutic efficacy for different scenarios of STH species (Ascaris lumbricoides, Trichuris trichiura and hookworms), pre-treatment infection levels, survey design (screen and select (SS); screen, select and retest (SSR) and no selection (NS)) and number of subjects enrolled (100-5,000). Finally, we integrated the outcome of the cost assessment into the simulation study to estimate the total survey costs and determined the most cost-efficient survey design.

PRINCIPAL FINDINGS

Kato-Katz allowed for both the highest sample throughput and the lowest cost per test, while FECPAKG2 required both the most laboratory time and was the most expensive. Counting of eggs accounted for 23% (FECPAKG2) or ≥80% (Kato-Katz and Mini-FLOTAC) of the total time-to-result. NS survey designs in combination with Kato-Katz were the most cost-efficient to assess therapeutic drug efficacy in all scenarios of STH species and endemicity.

CONCLUSIONS/SIGNIFICANCE

We confirm that Kato-Katz is the fecal egg counting method of choice for monitoring therapeutic drug efficacy, but that the survey design currently recommended by WHO (SS) should be updated. Our generic framework, which captures laboratory time and material costs, can be used to further support cost-efficient choices for other important surveys informing STH control programs. In addition, it can be used to explore the value of alternative diagnostic techniques, like automated egg counting, which may further reduce operational costs.

TRIAL REGISTRATION

ClinicalTrials.gov NCT03465488.

A review on innovative optical devices for the diagnosis of human soil-transmitted helminthiasis and schistosomiasis: from research and development to commercialization

Diagnosis of soil-transmitted helminth (STH) and schistosome infections relies largely on conventional microscopy which has limited sensitivity, requires highly trained personnel and is error-prone. Rapid advances in miniaturization of optical systems, sensors and processors have enhanced research and development of digital and automated microscopes suitable for the detection of these diseases in resource-limited settings. While some studies have reported proof-of-principle results, others have evaluated the performance of working prototypes in field settings. The extensive commercialization of these innovative devices has, however, not yet been achieved. This review provides an overview of recent publications (2010–2022) on innovative field applicable optical devices which can be used for the diagnosis of STH and schistosome infections. Using an adapted technology readiness level (TRL) scale taking into account the WHO target product profile (TPP) for these diseases, the developmental stages of the devices were ranked to determine the readiness for practical applications in field settings. From the reviewed 18 articles, 19 innovative optical devices were identified and ranked. Almost all of the devices (85%) were ranked with a TRL score below 8 indicating that, most of the devices are not ready for commercialization and field use. The potential limitations of these innovative devices were discussed. We believe that the outcome of this review can guide the end-to-end development of automated digital microscopes aligned with the WHO TPP for the diagnosis of STH and schistosome infections in resource-limited settings.

Evaluation of genome skimming to detect and characterise human and livestock helminths

The identification of gastrointestinal helminth infections of humans and livestock almost exclusively relies on the detection of eggs or larvae in faeces, followed by manual counting and morphological characterisation to differentiate species using microscopy-based techniques. However, molecular approaches based on the detection and quantification of parasite DNA are becoming more prevalent, increasing the sensitivity, specificity and throughput of diagnostic assays. High-throughput sequencing, from single PCR targets through to the analysis of whole genomes, offers significant promise towards providing information-rich data that may add value beyond traditional and conventional molecular approaches; however, thus far, its utility has not been fully explored to detect helminths in faecal samples. In this study, low-depth whole genome sequencing, i.e. genome skimming, has been applied to detect and characterise helminth diversity in a set of helminth-infected human and livestock faecal material. The strengths and limitations of this approach are evaluated using three methods to characterise and differentiate metagenomic sequencing data based on (i) mapping to whole mitochondrial genomes, (ii) whole genome assemblies, and (iii) a comprehensive internal transcribed spacer 2 (ITS2) database, together with validation using quantitative PCR (qPCR). Our analyses suggest that genome skimming can successfully identify most single and multi-species infections reported by qPCR and can provide sufficient coverage within some samples to resolve consensus mitochondrial genomes, thus facilitating phylogenetic analyses of selected genera, e.g. Ascaris spp. Key to this approach is both the availability and integrity of helminth reference genomes, some of which are currently contaminated with bacterial and host sequences. The success of genome skimming of faecal DNA is dependent on the availability of vouchered sequences of helminths spanning both taxonomic and geographic diversity, together with methods to detect or amplify minute quantities of parasite nucleic acids in mixed samples.

A review on low-cost microscopes for Open Science

This article presents a review after an exhaustive search that yielded 23 works carried out in the last decade for the availability of optical microscopes with open hardware as a low-cost alternative to commercial systems. These works were developed with the aim of covering needs within several areas such as: Bio Sciences research in institutions with limited resources, diagnosis of diseases and health screenings in large populations in developing countries, and training in educational contexts with a need for high availability of equipment and low replacement cost. The analysis of the selected works allows us to classify the analyzed solutions into two main categories, for which their essential characteristics are enumerated: portable field microscopes and multipurpose automated microscopes. Moreover, this work includes a discussion on the degree of maturity of the solutions in terms of the adoption of practices aligned with the development of Open Science. RESEARCH HIGHLIGHTS: Concise review on low-cost microscopes for developing Open Science, exposing the role of smartphone-based microscopy. The work classifies microscopes in two main categories: (1) portable field microscopes, and (2) multipurpose automated microscopes.

Worms and bugs of the gut: the search for diagnostic signatures using barcoding, and metagenomics-metabolomics

Gastrointestinal (GI) helminth infections cause significant morbidity in both humans and animals worldwide. Specific and sensitive diagnosis is central to the surveillance of such infections and to determine the effectiveness of treatment strategies used to control them. In this article, we: (i) assess the strengths and limitations of existing methods applied to the diagnosis of GI helminth infections of humans and livestock; (ii) examine high-throughput sequencing approaches, such as targeted molecular barcoding and shotgun sequencing, as tools to define the taxonomic composition of helminth infections; and (iii) discuss the current understanding of the interactions between helminths and microbiota in the host gut. Stool-based diagnostics are likely to serve as an important tool well into the future; improved diagnostics of helminths and their environment in the gut may assist the identification of biomarkers with the potential to define the health/disease status of individuals and populations, and to identify existing or emerging anthelmintic resistance.