Automatic identification of human helminth eggs on microscopic fecal specimens using digital image processing and an artificial neural network

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.

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.

High-throughput detection of parasites and ova in stool using the fully automatic digital feces analyzer, orienter model fa280

BACKGROUND

Intestinal parasitic infections can harm health by causing malnutrition, anemia, impaired growth and cognitive development, and alterations in microbiota composition and immune responses. Therefore, it is crucial to examine stool samples to diagnose parasitic infections. However, the traditional microscopic detection method is time-consuming, labor-intensive, and dependent on the expertise and training of microscopists. Hence, there is a need for a low-complexity, high-throughput, and cost-effective alternative to labor-intensive microscopic examinations.

METHODS

This study aimed to compare the performance of a fully automatic digital feces analyzer, Orienter Model FA280 (People's Republic of China) with that of the formalin-ethyl acetate concentration technique (FECT). We assessed and compared the agreement between the FA280 and the FECT for parasite detection and species identification in stool samples. The first part of the study analyzed 200 fresh stool samples for parasite detection using the FECT and FA280. With the FA280, the automatic feces analyzer performed the testing, and the digital microscope images were uploaded and automatically evaluated using an artificial intelligence (AI) program. Additionally, a skilled medical technologist conducted a user audit of the FA280 findings. The second set of samples comprised 800 preserved stool samples (preserved in 10% formalin). These samples were examined for parasites using the FECT and FA280 with a user audit.

RESULTS

For the first set of stool samples, there was no statistically significant difference in the pairwise agreements between the FECT and the FA280 with a user audit (exact binomial test, P = 1). However, there were statistically significant differences between the pairwise agreements for the FECT and the FA280 with the AI report (McNemar's test, P < 0.001). The agreement for the species identification of parasites between the FA280 with AI report and FECT showed fair agreement (overall agreement = 75.5%, kappa [κ] = 0.367, 95% CI 0.248-0.486). On the other hand, the user audit for the FA280 and FECT showed perfect agreement (overall agreement = 100%, κ = 1.00, 95% CI 1.00-1.00). For the second set of samples, the FECT detected significantly more positive samples for parasites than the FA280 with a user audit (McNemar's test, P < 0.001). The disparity in results may be attributed to the FECT using significantly larger stool samples than those used by the FA280. The larger sample size used by the FECT potentially contributed to the higher parasite detection rate. Regarding species identification, there was strong agreement between the FECT and the FA280 with a user audit for helminths (κ = 0.857, 95% CI 0.82-0.894). Similarly, there was perfect agreement for the species identification of protozoa between the FECT and the FA280 with user audit (κ = 1.00, 95% CI 1.00-1.00).

CONCLUSIONS

Although the FA280 has advantages in terms of simplicity, shorter performance time, and reduced contamination in the laboratory, there are some limitations to consider. These include a higher cost per sample testing and a lower sensitivity compared to the FECT. However, the FA280 enables rapid, convenient, and safe stool examination of parasitic infections.

Parasitic egg recognition using convolution and attention network

Intestinal parasitic infections (IPIs) caused by protozoan and helminth parasites are among the most common infections in humans in low-and-middle-income countries. IPIs affect not only the health status of a country, but also the economic sector. Over the last decade, pattern recognition and image processing techniques have been developed to automatically identify parasitic eggs in microscopic images. Existing identification techniques are still suffering from diagnosis errors and low sensitivity. Therefore, more accurate and faster solution is still required to recognize parasitic eggs and classify them into several categories. A novel Chula-ParasiteEgg dataset including 11,000 microscopic images proposed in ICIP2022 was utilized to train various methods such as convolutional neural network (CNN) based models and convolution and attention (CoAtNet) based models. The experiments conducted show high recognition performance of the proposed CoAtNet that was tuned with microscopic images of parasitic eggs. The CoAtNet produced an average accuracy of 93%, and an average F1 score of 93%. The finding opens door to integrate the proposed solution in automated parasitological diagnosis.

Advances Towards Automatic Detection and Classification of Parasites Microscopic Images Using Deep Convolutional Neural Network: Methods, Models and Research Directions

In the developing world, parasites are responsible for causing several serious health problems, with relatively high infections in human beings. The traditional manual light microscopy process of parasite recognition remains the golden standard approach for the diagnosis of parasitic species, but this approach is time-consuming, highly tedious, and also difficult to maintain consistency but essential in parasitological classification for carrying out several experimental observations. Therefore, it is meaningful to apply deep learning to address these challenges. Convolution Neural Network and digital slide scanning show promising results that can revolutionize the clinical parasitology laboratory by automating the process of classification and detection of parasites. Image analysis using deep learning methods have the potential to achieve high efficiency and accuracy. For this review, we have conducted a thorough investigation in the field of image detection and classification of various parasites based on deep learning. Online databases and digital libraries such as ACM, IEEE, ScienceDirect, Springer, and Wiley Online Library were searched to identify sufficient related paper collections. After screening of 200 research papers, 70 of them met our filtering criteria, which became a part of this study. This paper presents a comprehensive review of existing parasite classification and detection methods and models in chronological order, from traditional machine learning based techniques to deep learning based techniques. In this review, we also demonstrate the summary of machine learning and deep learning methods along with dataset details, evaluation metrics, methods limitations, and future scope over the one decade. The majority of the technical publications from 2012 to the present have been examined and summarized. In addition, we have discussed the future directions and challenges of parasites classification and detection to help researchers in understanding the existing research gaps. Further, this review provides support to researchers who require an effective and comprehensive understanding of deep learning development techniques, research, and future trends in the field of parasites detection and classification.

Automatic recognition of parasitic products in stool examination using object detection approach

BACKGROUND

Object detection is a new artificial intelligence approach to morphological recognition and labeling parasitic pathogens. Due to the lack of equipment and trained personnel, artificial intelligence innovation for searching various parasitic products in stool examination will enable patients in remote areas of undeveloped countries to access diagnostic services. Because object detection is a developing approach that has been tested for its effectiveness in detecting intestinal parasitic objects such as protozoan cysts and helminthic eggs, it is suitable for use in rural areas where many factors supporting laboratory testing are still lacking. Based on the literatures, the YOLOv4-Tiny produces faster results and uses less memory with the support of low-end GPU devices. In comparison to the YOLOv3 and YOLOv3-Tiny models, this study aimed to propose an automated object detection approach, specifically the YOLOv4-Tiny model, for automatic recognition of intestinal parasitic products in stools.

METHODS

To identify protozoan cysts and helminthic eggs in human feces, the three YOLO approaches; YOLOv4-Tiny, YOLOv3, and YOLOv3-Tiny, were trained to recognize 34 intestinal parasitic classes using training of image dataset. Feces were processed using a modified direct smear method adapted from the simple direct smear and the modified Kato-Katz methods. The image dataset was collected from intestinal parasitic objects discovered during stool examination and the three YOLO models were trained to recognize the image datasets.

RESULTS

The non-maximum suppression technique and the threshold level were used to analyze the test dataset, yielding results of 96.25% precision and 95.08% sensitivity for YOLOv4-Tiny. Additionally, the YOLOv4-Tiny model had the best AUPRC performance of the three YOLO models, with a score of 0.963.

CONCLUSION

This study, to our knowledge, was the first to detect protozoan cysts and helminthic eggs in the 34 classes of intestinal parasitic objects in human stools.

Comparison of FECPAKG2, a modified Mini-FLOTAC technique and combined sedimentation and flotation for the coproscopic examination of helminth eggs in horses

Background

Due to high prevalence of anthelmintic resistance in equine helminths, selective treatment is increasingly promoted and in some countries a positive infection diagnosis is mandatory before treatment. Selective treatment is typically recommended when the number of worm eggs per gram faeces (epg) exceeds a particular threshold. In the present study we compared the semi-quantitative sedimentation/flotation method with the quantitative methods Mini-FLOTAC and FECPAK^G2 in terms of precision, sensitivity, inter-rater reliability and correlation of worm egg counts to improve the choice of optimal diagnostic tools.

Methods

Using sedimentation/flotation (counting raw egg numbers up to 200), we investigated 1067 horse faecal samples using a modified Mini-FLOTAC approach (multiplication factor of 5 to calculate epgs from raw egg counts) and FECPAK^G2 (multiplication factor of 45).

Results

Five independent analyses of the same faecal sample with all three methods revealed that variance was highest for the sedimentation/flotation method while there were no significant differences between methods regarding the coefficient of variance. Sedimentation/flotation detected the highest number of samples positive for strongyle and Parascaris spp. eggs, followed by Mini-FLOTAC and FECPAK^G2. Regarding Anoplocephalidae, no significant difference in frequency of positive samples was observed between Mini-FLOTAC and sedimentation/flotation. Cohen’s κ values comparing individual methods with the combined result of all three methods revealed almost perfect agreement (κ ≥ 0.94) for sedimentation/flotation and strong agreement for Mini-FLOTAC (κ ≥ 0.83) for strongyles and Parascaris spp. For FECPAK^G2, moderate and weak agreements were found for the detection of strongyle (κ = 0.62) and Parascaris (κ = 0.51) eggs, respectively. Despite higher sensitivity, the Mini-FLOTAC mean epg was significantly lower than that with FECPAK^G2 due to samples with > 200 raw egg counts by sedimentation/flotation, while in samples with lower egg shedding epgs were higher with Mini-FLOTAC than with FECPAK^G2.

Conclusions

For the simple detection of parasite eggs, for example, to treat foals infected with Parascaris spp., sedimentation/flotation is sufficient and more sensitive than the other two quantitative investigared in this study. Mini-FLOTAC is predicted to deliver more precise results in faecal egg count reduction tests due to higher raw egg counts. Finally, to identify animals with a strongyle epg above a certain threshold for treatment, FECPAK^G2 delivered results comparable to Mini-FLOTAC.

Grpahical Abstract

Supplementary Information

The online version contains supplementary material available at 10.1186/s13071-022-05266-y.

Helminth Eggs as a Magnetic Biomaterial: Introducing a Recognition Probe

Parasitic helminths, despite their known negative impact (biomaterial) on human health and animal production, have fascinating features. In this study, we find fantastic magnetic properties in several forms: inductor [between 20.10 and 58.85 (±2.50) H], source of detectable electrical voltage [from +0.5 to 7.3 (±0.1) V, vs. the ground, GND, measured by an AVO meter] and different inductor magnitude [between 3.33 and 41.23 (±0.76)] μH, detected by electrochemical impedance spectroscopy as well as frequency scannable electromagnetic wave horn) in several frequencies (including 100, 120, Hz, and 1, 10, 100 kHz) in “Fasciola hepatica”, “Parascaris equorum” (with and without larvae), “Dicrocoelium dendriticum,” “Taenia multiceps”, and “Moniezia expansa” eggs. This claim is attributed to some surprising characteristics, including superior inductance and intrinsic magnetic susceptibility. This feature along with a close relationship to helminth egg structure, is a novel probe with acceptable reproducibility (RSD &gt; 8.0%) and high enough trustworthiness for adequate differentiation in their magnitudes, relatively. These traits were measured by the “Single Cell Rrecording” methodology using a three-microelectrode system, implanted to each egg at the Giga ohm sealed condition (6.08 ± 0.22 GΩ cm−1, n = 5). The reliability of these results was further confirmed using multiple calibrated instruments such as a high-resolution inductance analyzer, LCR meter, impedance spectrometer, potentiometer, and an anomalous Hall effect (Magnetic field density) sensor. In addition, the critical role played (Synergistic Effect) by water-like molecules as the intermediate medium, besides the partial influence of other compounds such as dissolved oxygen, are investigated qualitatively, and specific relation between these molecules and magnetic field creation in helminth eggs was proved. These intrinsic characteristics would provide novel facilitators for efficient arriving at the researchable bio-based magnetic biomaterials, besides innovative and real-time identification probes in the “Parasitology” fields.

Primary Mobile Image Analysis of Human Intestinal Worm Detection

One among a lot of public health concerns in rural and tropical areas is the human intestinal parasite. Traditionally, diagnosis of these parasites is by visual analysis of stool specimens, which is usually tedious and time-consuming. In this study, the authors combine techniques in the Laplacian pyramid, Gabor filter, and wavelet to build a feature vector for the discrimination of intestinal worm in a low-resolution image captured with mobile devices. The dimension of the feature vector is reduced using principal component analysis, and the resultant vector is considered as input to the SVM classifier. The proposed framework was applied to the Makerere intestinal dataset. At its preliminary stage, the results demonstrate satisfactory classification with an accuracy rate of 65.22% with possible extension in future work.

Diode and Active Negative Resistance Behaviors of Helminth Eggs as a Novel Identification/Differentiation Probe

Helminths have always been studied as one of the critically annoying pathogens of parasite classes due to their adverse effects on the ecosystem of human life. They have the potency to negatively affect their hosts as points of disease, infection, cancer, and death, but in this study, we found interesting electronic properties in Fasciola hepatica, Parascaris equorum (with and without larvae), Dicrocoelium dendriticum, Taenia multiceps, and Moniezia expansa eggs. This claim is attributed to some surprising characteristics such as significant diode behavior [forward bias, 5.36-11.17 (±0.01) V, versus the ground, GND] and backward bias (-45.0 to -125.0 (±7.0) V, versus the GND) and highly active negative resistance (-2.59 to -7.11) × 1015 (±1.5) Ω in the AC mode. These traits were measured by the "blind patch-clamp, single-unit recording" methodology using a three-microelectrode system, implanted onto each tested egg under giga ohm sealed conditions (6.28 ± 0.02 GΩ cm-1 and n = 4). All the characteristic parameters were simultaneously attributed to the helminth egg structure by acceptable reproducibility (percentage of relative standard deviation: > 5%) and high enough rectitude with enough differentiation in their magnitudes, relatively. The reliability of these results was further confirmed using multiple calibrated techniques such as alternative/direct current voltammetry. Also, the significant role of water molecules as the key medium in creating these properties is evaluated qualitatively. In addition, the study aims at introducing these interesting parameters as a new approach to the fabrication of bio-based electronic elements, which are considered as a novel class of helminth egg-detection and -identification probes.

Automated Diagnostics: Advances in the Diagnosis of Intestinal Parasitic Infections in Humans and Animals

The increasingly close proximity between people and animals is of great concern for public health, given the risk of exposure to infectious diseases transmitted through animals, which are carriers of more than 60 zoonotic agents. These diseases, which are included in the list of Neglected Tropical Diseases, cause losses in countries with tropical and subtropical climates, and in regions with temperate climates. Indeed, they affect more than a billion people around the world, a large proportion of which are infected by one or more parasitic helminths, causing annual losses of billions of dollars. Several studies are being conducted in search for differentiated, more sensitive diagnostics with fewer errors. These studies, which involve the automated examination of intestinal parasites, still face challenges that must be overcome in order to ensure the proper identification of parasites. This includes a protocol that allows for elimination of most of the debris in samples, satisfactory staining of parasite structures, and a robust image database. Our objective here is therefore to offer a critical description of the techniques currently in use for the automated diagnosis of intestinal parasites in fecal samples, as well as advances in these techniques.

Helminth egg analysis platform (HEAP): An opened platform for microscopic helminth egg identification and quantification based on the integration of deep learning architectures

BACKGROUND

Millions of people throughout the world suffer from parasite infections. Traditionally, technicians use manual eye inspection of microscopic specimens to perform a parasite examination. However, manual operations have limitations that hinder the ability to obtain precise egg counts and cause inefficient identification of infected parasites on co-infections. The technician requirements for handling a large number of microscopic examinations in countries that have limited medical resources are substantial. We developed the helminth egg analysis platform (HEAP) as a user-friendly microscopic helminth eggs identification and quantification platform to assist medical technicians during parasite infection examination.

METHODS

Multiple deep learning strategies including SSD (Single Shot MultiBox Detector), U-net, and Faster R-CNN (Faster Region-based Convolutional Neural Network) are integrated to identify the same specimen allowing users to choose the best predictions. An image binning and egg-in-edge algorithm based on pixel density detection was developed to increase the performance. Computers with different operation systems can be gathered to lower the computation time using our easy-to-deploy software architecture.

RESULTS

A user-friendly interface is provided to substantially increase the efficiency of manual validation. To adapt to low-cost computers, we architected a distributed computing structure with high flexibilities.

CONCLUSIONS

HEAP serves not only as a prediction service provider but also as a parasitic egg database of microscopic helminth egg image collection, labeling data and pretrained models. All images and labeling resources are free and accessible at http://heap.cgu.edu.tw. HEAP can also be an ideal education and training resource for helminth egg examination.

Utility of an Automatic Vision-Based Examination System (AVE-562) for the Detection of Clonorchis sinensis Eggs in Stool

Stool examination is the gold standard for the detection of intestinal parasites. We assessed the performance of a newly developed AVE-562 analyzer (AVE Science & Technology Co., Hunan, China) for the vision-based detection of eggs of Clonorchis sinensis—the most common intestinal parasite in Korea—in stool samples. In total, 30 stool samples with a high or low egg count or without eggs (as negative control samples) (N=10 each) were prepared and analyzed. The performance of the AVE-562 analyzer was compared with that of the formalin-ether concentration (FEC) method. The overall correct identification rate of the AVE-562 analyzer based on FEC results was 66.6%. The sensitivity, specificity, positive predictive value, and negative predictive value of the AVE-562 analyzer for detecting C. sinensis eggs were 36.4%, 100.0%, 100.0%, and 73.1%, respectively. The average time required to run five tests simultaneously was 27 min using the AVE-562 analyzer and 58 min using the FEC method. Although the AVE-562 analyzer enables rapid and convenient stool examination, its sensitivity needs to be improved, particularly considering the prevalence of low-burden C. sinensis infection in Korea.

The effect of analyst training on fecal egg counting variability

Fecal egg counts (FECs) are essential for veterinary parasite control programs. Recent advances led to the creation of an automated FEC system that performs with increased precision and reduces the need for training of analysts. However, the variability contributed by analysts has not been quantified for FEC methods, nor has the impact of training on analyst performance been quantified. In this study, three untrained analysts performed FECs on the same slides using the modified McMaster (MM), modified Wisconsin (MW), and the automated system with two different algorithms: particle shape analysis (PSA) and machine learning (ML). Samples were screened and separated into negative (no strongylid eggs seen), 1-200 eggs per gram of feces (EPG), 201-500 EPG, 501-1000 EPG, and 1001+ EPG levels, and ten repeated counts were performed for each level and method. Analysts were then formally trained and repeated the study protocol. Between analyst variability (BV), analyst precision (AP), and the proportion of variance contributed by analysts were calculated. Total BV was significantly lower for MM post-training (p = 0.0105). Additionally, AP variability and analyst variance both tended to decrease for the manual MM and MW methods. Overall, MM had the lowest BV both pre- and post-training, although PSA and ML were minimally affected by analyst training. This research illustrates not only how the automated methods could be useful when formal training is unavailable but also how impactful formal training is for traditional manual FEC methods.

Helminth Egg Automatic Detector (HEAD): Improvements in development for digital identification and quantification of Helminth eggs and its application online

Conventional analytical techniques for evaluating Helminth eggs are based on different steps to concentrate them in a pellet for direct observation and quantification under a light microscope, which can generate under-counts or over-counts and be time consuming. To enhance this process, a new approach via automatic identification was implemented in which various image processing detectors were developed and incorporated into a Helminth Egg Automatic Detector (HEAD) system. This allowed the identification and quantification of pathogenic eggs of global medical importance. More than 2.6 billion people are currently affected and infected, and this results in approximately 80,000 child deaths each year. As a result, since 1980 the World Health Organization (WHO) has implemented guidelines, regulations and criteria for the control of the health risk. After the initial release of the analytical technique, two improvements were developed in the detector: first, a texture verification process that reduced the number of false positive results; and second, the establishment of the optimal thresholds for each species. In addition, the software was made available on a free platform. After performing an internal statistical verification of the system, testing with internationally recognized parasitology laboratories was carried out, Subsequently, the HEAD System is capable of identifying and quantifying different species of Helminth eggs in different environmental samples: wastewater, sludge, biosolids, excreta and soil, with in-service sensitivity and specificity values for the open library for machine learning TensorFlow (TF) model of 96.82% and 97.96% respectively. The current iteration uses AutoML Vision (a computer platform for the automatization of machine learning models, making it easier to train, optimize and export results to cloud applications or devices). It represents a useful and cheap tool that could be utilized by environmental monitoring facilities and laboratories around the world.•The HEAD Software will significantly reduce the costs associated with the detection and quantification of helminth eggs to a high level of accuracy.•It represents a tool, not only for microbiologists and researchers, but also for various agencies involved in sanitation, such as environmental regulation agencies, which currently require highly trained technicians.•The simplicity of the device contributes to the control the contamination of water, soil, and crops, even in poor and isolated communities.

Helminth Egg Automatic Detector (HEAD): Improvements in development for digital identification and quantification of helminth eggs and their application online

Helminths are parasitic worms that constitute a major public health problem. Conventional analytical techniques to evaluate helminth eggs in environmental samples rely on different steps, namely sedimentation, filtration, centrifugation, and flotation, to separate the eggs from a variety of particles and concentrate them in a pellet for direct observation under an optical microscope. To improve this process, a new approach was implemented in which various image processing algorithms were developed and implemented by a Helminth Egg Automatic Detector (HEAD). This allowed identification and quantification of pathogenic helminth eggs of global medical importance and it was found to be useful for relatively clean wastewater samples. After the initial version, two improvements were developed: first, a texture verification process that reduced the number of false positive results; and second, the establishment of the optimal thresholds (morphology and texture) for each helminth egg species. This second implementation, which was found to improve on the results of the former, was developed with the objective of using free software as a platform for the system. This does not require the purchase of a license, unlike the previous version that required a Mathworks® license to run. After an internal statistical verification of the system was carried out, trials in internationally recognized microbiology laboratories were performed with the aim of reinforcing software training and developing a web-based system able to receive images and perform the analysis throughout a web service. Once completed, these improvements represented a useful and cheap tool that could be used by environmental monitoring facilities and laboratories throughout the world; this tool is capable of identifying and quantifying different species of helminth eggs in otherwise difficult environmental samples: wastewater, soil, biosolids, excreta, and sludge, with a sensitivity and specificity for the TensorFlow (TF) model in the web service values of 96.82% and 97.96% respectively. Additionally, in the case of Ascaris, it may even differentiate between fertile and non-fertile eggs.

Automated diagnosis of intestinal parasites: A new hybrid approach and its benefits

Intestinal parasites are responsible for several diseases in human beings. In order to eliminate the error-prone visual analysis of optical microscopy slides, we have investigated automated, fast, and low-cost systems for the diagnosis of human intestinal parasites. In this work, we present a hybrid approach that combines the opinion of two decision-making systems with complementary properties: (DS₁) a simpler system based on very fast handcrafted image feature extraction and support vector machine classification and (DS₂) a more complex system based on a deep neural network, Vgg-16, for image feature extraction and classification. DS₁ is much faster than DS₂, but it is less accurate than DS₂. Fortunately, the errors of DS₁ are not the same of DS₂. During training, we use a validation set to learn the probabilities of misclassification by DS₁ on each class based on its confidence values. When DS₁ quickly classifies all images from a microscopy slide, the method selects a number of images with higher chances of misclassification for characterization and reclassification by DS₂. Our hybrid system can improve the overall effectiveness without compromising efficiency, being suitable for the clinical routine - a strategy that might be suitable for other real applications. As demonstrated on large datasets, the proposed system can achieve, on average, 94.9%, 87.8%, and 92.5% of Cohen's Kappa on helminth eggs, helminth larvae, and protozoa cysts, respectively.

Diagnostic performance of McMaster, Wisconsin, and automated egg counting techniques for enumeration of equine strongyle eggs in fecal samples

Fecal egg counts are the cornerstone of equine parasite control programs. Previous work led to the development of an automated, image-analysis-based parasite egg counting system. The system has been further developed to include an automated reagent dispenser unit and a custom camera (CC) unit that generates higher resolution images, as well as a particle shape analysis (PSA) algorithm and machine learning (ML) algorithm. The first aim of this study was to conduct a comprehensive comparison of method precision between the original smartphone (SP) unit with the PSA algorithm, CC/PSA, CC/ML, and the traditional McMaster (MM) and Wisconsin (MW) manual techniques. Additionally, a Bayesian analysis was performed to estimate and compare sensitivity and specificity of all five methods. Feces were collected from horses, screened with triplicate Mini-FLOTAC counts, and placed into five categories: negative (no eggs seen), > 0 - ≤ 200 eggs per gram (EPG), > 200 - ≤ 500 EPG, > 500 - ≤ 1000 EPG, and > 1000 EPG. Ten replicates per horse were analyzed for each technique. Technical variability for samples > 200 EPG was significantly higher for MM than CC/PSA and CC/ML (p <  0.0001). Biological variability for samples> 0 was numerically highest for CC/PSA, but with samples > 200 EPG, MM had a significantly lower CV than MW (p =  0.001), MW had a significantly lower CV than CC/PSA (p <  0.0001), CC/ML had a significantly lower CV than both MW and SP/PSA (p <  0.0001, p =  0.0003), and CC/PSA had a significantly lower CV than CC/SP (p =  0.0115). Sensitivity was> 98 % for all five methods with no significant differences. Specificity, however, was significantly the highest for CC/PSA, followed numerically by SP/PSA, MM, CC/ML, and finally MW. Overall, the automated counting system is a promising new development in equine parasitology. Continued refinement to the counting algorithms will help improve precision and specificity, while additional research in areas such as egg loss, analyst variability at the counting step, and accuracy will help create a complete picture of its impact as a new fecal egg count method.

Preliminary evaluation of a novel, fully automated, Telenostic device for rapid field-diagnosis of cattle parasites

New ideas for diagnostics in clinical parasitology are needed to overcome some of the difficulties experienced in the widespread adoption of detection methods for gastrointestinal parasites in livestock. Here we provide an initial evaluation of the performance of a newly developed automated device (Telenostic) to identify and quantify parasitic elements in fecal samples. This study compared the Telenostic device with the McMaster and Mini-FLOTAC for counting of strongyle eggs in a fecal sample. Three bovine fecal samples were examined, in triplicate, on each of the three fecal egg-counting devices. In addition, both manual (laboratory technician) and automated analysis (image analysis algorithm) were performed on the Telenostic device to calculate fecal egg counts (FEC). Overall, there were consistent egg counts reported across the three devices and calculation methods. The Telenostic device compared very favourably to the Mini-FLOTAC and McMaster. Only in sample C, a significant difference (P < 0.05) was observed between the egg counts obtained by Mini-FLOTAC and by the other methods. From this limited dataset it can be concluded that the Telenostic-automated test is comparable to currently used benchmark FEC methods, while improving the workflow, test turn-around time and not requiring trained laboratory personnel to operate or interpret the results.

A historical review of the techniques of recovery of parasites for their detection in human stools

Since the early 20th century, the detection of intestinal parasites has improved with the development of several techniques for parasitic structures recovery and identification, which differ in sensitivity, specificity, practicality, cost, and infrastructure demand. This study aims to review, in chronological order, the stool examination techniques and discuss their advantages, limitations, and perspectives, and to provide professionals and specialists in this field with data that lays a foundation for critical analysis on the use of such procedures. The concentration procedures that constitute the main techniques applied in routine research and in parasitological kits are a) spontaneous sedimentation; b) centrifugation-sedimentation with formalin-ethyl acetate; and c) flotation with zinc sulfate solution. While selecting a technique, one should consider the purpose of its application and the technical-operational, biological, and physicochemical factors inherent in the procedures used in stool processing, which may restrict its use. These intrinsic limitations may have undergone procedural changes driven by scientific and technological development and by development of alternative methods, which now contribute to the improvement of diagnostic accuracy.

Detection of Intestinal Protozoa in Trichrome-Stained Stool Specimens by Use of a Deep Convolutional Neural Network

Intestinal protozoa are responsible for relatively few infections in the developed world, but the testing volume is disproportionately high. Manual light microscopy of stool remains the gold standard but can be insensitive, time-consuming, and difficult to maintain competency. Artificial intelligence and digital slide scanning show promise for revolutionizing the clinical parasitology laboratory by augmenting the detection of parasites and slide interpretation using a convolutional neural network (CNN) model. The goal of this study was to develop a sensitive model that could screen out negative trichrome slides, while flagging potential parasites for manual confirmation. Conventional protozoa were trained as "classes" in a deep CNN. Between 1,394 and 23,566 exemplars per class were used for training, based on specimen availability, from a minimum of 10 unique slides per class. Scanning was performed using a 40× dry lens objective automated slide scanner. Data labeling was performed using a proprietary Web interface. Clinical validation of the model was performed using 10 unique positive slides per class and 125 negative slides. Accuracy was calculated as slide-level agreement (e.g., parasite present or absent) with microscopy. Positive agreement was 98.88% (95% confidence interval [CI], 93.76% to 99.98%), and negative agreement was 98.11% (95% CI, 93.35% to 99.77%). The model showed excellent reproducibility using slides containing multiple classes, a single class, or no parasites. The limit of detection of the model and scanner using serially diluted stool was 5-fold more sensitive than manual examinations by multiple parasitologists using 4 unique slide sets. Digital slide scanning and a CNN model are robust tools for augmenting the conventional detection of intestinal protozoa.

[Artificial intelligence empowers laboratory medicine in Industry 4.0]

Since 2017, China, the United States, and the European Union have successively issued national-level artificial intelligence (AI) strategic development plans, and the human history is about to witness the 4th industrial revolution with the theme of "intelligence". In the field of medical testing, the explosive growth of AI theories and technologies also provide a new direction for the development of medical testing theory, methods and applications. We review the evolution of AI and the recent progress in three major elements of AI, namely algorithms, data and computing power, and elaborate on the combined innovation of "AI + testing" in light of the key application dimensions of medical testing. The major applications include specimen collection robots, sample dilution robots and sample transfer robots involved in the processing of test specimens; test item mining such as tumor markers and pharmacogenomics; cytomorphology, laboratory medicine data processing, auxiliary diagnostic models, and internet-based medical tests. With the advent of the era of Industry 4.0, AI technology will promote the development of medical testing from automation to a highly intelligent stage.

Automated Diagnosis of Canine Gastrointestinal Parasites Using Image Analysis

Because canine intestinal parasites are considered cosmopolitan, they carry significant zoonotic potential to public health. These etiological agents are routinely diagnosed using microscopic examination commonly used because of its low cost, simple execution, and direct evidence. However, there are reports in the literature on the poor performance of this test due to low to moderate sensitivity resulting from frequent errors, procedures and interpretation. Therefore, to improve the diagnostic efficiency of microscopic examination in veterinary medicine, we developed and evaluated a unique new protocol. This system was tested in a study involving four genera of highly prevalent canine intestinal parasites in an endemic region in São Paulo state, Brazil. Fecal samples from 104 animals were collected for this research. The new protocol had a significantly higher (p < 0.0001) number of positive cases on image data, including parasites and impurities, and was elaborate to test them with the TF-GII/Dog technique, with a moderate agreement and Kappa index of 0.7636. We concluded that the new Prototic Coproparasitological Test for Dogs (PC-Test Dog) allowed a better visualization of the parasitic structures and showed a favorable result for the diagnosis of intestinal parasites in dogs.

Machine learning in the clinical microbiology laboratory: has the time come for routine practice?

BACKGROUND

Machine learning (ML) allows the analysis of complex and large data sets and has the potential to improve health care. The clinical microbiology laboratory, at the interface of clinical practice and diagnostics, is of special interest for the development of ML systems.

AIMS

This narrative review aims to explore the current use of ML In clinical microbiology.

SOURCES

References for this review were identified through searches of MEDLINE/PubMed, EMBASE, Google Scholar, biorXiv, arXiV, ACM Digital Library and IEEE Xplore Digital Library up to November 2019.

CONTENT

We found 97 ML systems aiming to assist clinical microbiologists. Overall, 82 ML systems (85%) targeted bacterial infections, 11 (11%) parasitic infections, nine (9%) viral infections and three (3%) fungal infections. Forty ML systems (41%) focused on microorganism detection, identification and quantification, 36 (37%) evaluated antimicrobial susceptibility, and 21 (22%) targeted the diagnosis, disease classification and prediction of clinical outcomes. The ML systems used very diverse data sources: 21 (22%) used genomic data of microorganisms, 19 (20%) microbiota data obtained by metagenomic sequencing, 19 (20%) analysed microscopic images, 17 (18%) spectroscopy data, eight (8%) targeted gene sequencing, six (6%) volatile organic compounds, four (4%) photographs of bacterial colonies, four (4%) transcriptome data, three (3%) protein structure, and three (3%) clinical data. Most systems used data from high-income countries (n = 71, 73%) but a significant number used data from low- and middle-income countries (n = 36, 37%). Performance measures were reported for the 97 ML systems, but no article described their use in clinical practice or reported impact on processes or clinical outcomes.

IMPLICATIONS

In clinical microbiology, ML has been used with various data sources and diverse practical applications. The evaluation and implementation processes represent the main gap in existing ML systems, requiring a focus on their interpretability and potential integration into real-world settings.

Segmentation-driven Hierarchical RetinaNet for Detecting Protozoa in Micrograph

Protozoa detection and identification play important roles in many practical domains such as parasitology, scientific research, biological treatment processes, and environmental quality evaluation. Traditional laboratory methods for protozoan identification are time-consuming and require expert knowledge and expensive equipment. Another approach is using micrographs to identify the species of protozoans that can save a lot of time and reduce the cost. However, the existing methods in this approach only identify the species when the protozoan are already segmented. These methods study features of shapes and sizes. In this work, we detect and identify the images of cysts and oocysts of various species such as: Giardia lamblia, Iodamoeba butschilii, Toxoplasma gondi, Cyclospora cayetanensis, Balantidium coli, Sarcocystis, Cystoisospora belli and Acanthamoeba, which have round shapes in common and affect human and animal health seriously. We propose Segmentation-driven Hierarchical RetinaNet to automatically detect, segment, and identify protozoans in their micrographs. By applying multiple techniques such as transfer learning, and data augmentation techniques, and dividing training samples into life-cycle stages of protozoans, we successfully overcome the lack of data issue in applying deep learning for this problem. Even though there are at most 5 samples per life-cycle category in the training data, our proposed method still achieves promising results and outperforms the original RetinaNet on our protozoa dataset.

Kankanet: An artificial neural network-based object detection smartphone application and mobile microscope as a point-of-care diagnostic aid for soil-transmitted helminthiases

Background

Endemic areas for soil-transmitted helminthiases often lack the tools and trained personnel necessary for point-of-care diagnosis. This study pilots the use of smartphone microscopy and an artificial neural network-based (ANN) object detection application named Kankanet to address those two needs.

Methodology/Principal findings

A smartphone was equipped with a USB Video Class (UVC) microscope attachment and Kankanet, which was trained to recognize eggs of Ascaris lumbricoides, Trichuris trichiura, and hookworm using a dataset of 2,078 images. It was evaluated for interpretive accuracy based on 185 new images. Fecal samples were processed using Kato-Katz (KK), spontaneous sedimentation technique in tube (SSTT), and Merthiolate-Iodine-Formaldehyde (MIF) techniques. UVC imaging and ANN interpretation of these slides was compared to parasitologist interpretation of standard microscopy.Relative to a gold standard defined as any positive result from parasitologist reading of KK, SSTT, and MIF preparations through standard microscopy, parasitologists reading UVC imaging of SSTT achieved a comparable sensitivity (82.9%) and specificity (97.1%) in A. lumbricoides to standard KK interpretation (97.0% sensitivity, 96.0% specificity). The UVC could not accurately image T. trichiura or hookworm. Though Kankanet interpretation was not quite as sensitive as parasitologist interpretation, it still achieved high sensitivity for A. lumbricoides and hookworm (69.6% and 71.4%, respectively). Kankanet showed high sensitivity for T. trichiura in microscope images (100.0%), but low in UVC images (50.0%).

Conclusions/Significance

The UVC achieved comparable sensitivity to standard microscopy with only A. lumbricoides. With further improvement of image resolution and magnification, UVC shows promise as a point-of-care imaging tool. In addition to smartphone microscopy, ANN-based object detection can be developed as a diagnostic aid. Though trained with a limited dataset, Kankanet accurately interprets both standard microscope and low-quality UVC images. Kankanet may achieve sensitivity comparable to parasitologists with continued expansion of the image database and improvement of machine learning technology.

For rainforest-enshrouded rural villages of Madagascar, soil-transmitted helminthiases are more the rule than the exception. However, the microscopy equipment and lab technicians needed for diagnosis are a distance of several days’ hike away. We piloted a solution for these communities by leveraging resources the villages already had: a traveling team of local health care workers, and their personal Android smartphones. We demonstrated that an inexpensive, commercially available microscope attachment for smartphones could rival the sensitivity and specificity of a regular microscope using standard field fecal sample processing techniques. We also developed an artificial neural network-based object detection Android application, called Kankanet, based on open-source programming libraries. Kankanet was used to detect eggs of the three most common soil-transmitted helminths: Ascaris lumbricoides, Trichuris trichiura, and hookworm. We found Kankanet to be moderately sensitive and highly specific for both standard microscope images and low-quality smartphone microscope images. This proof-of-concept study demonstrates the diagnostic capabilities of artificial neural network-based object detection systems. Since the programming frameworks used were all open-source and user-friendly even for computer science laymen, artificial neural network-based object detection shows strong potential for development of low-cost, high-impact diagnostic aids essential to health care and field research in resource-limited communities.

Evaluation of accuracy and precision of a smartphone based automated parasite egg counting system in comparison to the McMaster and Mini-FLOTAC methods

Fecal egg counts are emphasized for guiding equine helminth parasite control regimens due to the rise of anthelmintic resistance. This, however, poses further challenges, since egg counting results are prone to issues such as operator dependency, method variability, equipment requirements, and time commitment. The use of image analysis software for performing fecal egg counts is promoted in recent studies to reduce the operator dependency associated with manual counts. In an attempt to remove operator dependency associated with current methods, we developed a diagnostic system that utilizes a smartphone and employs image analysis to generate automated egg counts. The aims of this study were (1) to determine precision of the first smartphone prototype, the modified McMaster and ImageJ; (2) to determine precision, accuracy, sensitivity, and specificity of the second smartphone prototype, the modified McMaster, and Mini-FLOTAC techniques. Repeated counts on fecal samples naturally infected with equine strongyle eggs were performed using each technique to evaluate precision. Triplicate counts on 36 egg count negative samples and 36 samples spiked with strongyle eggs at 5, 50, 500, and 1000 eggs per gram were performed using a second smartphone system prototype, Mini-FLOTAC, and McMaster to determine technique accuracy. Precision across the techniques was evaluated using the coefficient of variation. In regards to the first aim of the study, the McMaster technique performed with significantly less variance than the first smartphone prototype and ImageJ (p<0.0001). The smartphone and ImageJ performed with equal variance. In regards to the second aim of the study, the second smartphone system prototype had significantly better precision than the McMaster (p<0.0001) and Mini-FLOTAC (p<0.0001) methods, and the Mini-FLOTAC was significantly more precise than the McMaster (p=0.0228). Mean accuracies for the Mini-FLOTAC, McMaster, and smartphone system were 64.51%, 21.67%, and 32.53%, respectively. The Mini-FLOTAC was significantly more accurate than the McMaster (p<0.0001) and the smartphone system (p<0.0001), while the smartphone and McMaster counts did not have statistically different accuracies. Overall, the smartphone system compared favorably to manual methods with regards to precision, and reasonably with regards to accuracy. With further refinement, this system could become useful in veterinary practice.

Mathematical algorithm for the automatic recognition of intestinal parasites

Parasitic infections are generally diagnosed by professionals trained to recognize the morphological characteristics of the eggs in microscopic images of fecal smears. However, this laboratory diagnosis requires medical specialists which are lacking in many of the areas where these infections are most prevalent. In response to this public health issue, we developed a software based on pattern recognition analysis from microscopi digital images of fecal smears, capable of automatically recognizing and diagnosing common human intestinal parasites. To this end, we selected 229, 124, 217, and 229 objects from microscopic images of fecal smears positive for Taenia sp., Trichuris trichiura, Diphyllobothrium latum, and Fasciola hepatica, respectively. Representative photographs were selected by a parasitologist. We then implemented our algorithm in the open source program SCILAB. The algorithm processes the image by first converting to gray-scale, then applies a fourteen step filtering process, and produces a skeletonized and tri-colored image. The features extracted fall into two general categories: geometric characteristics and brightness descriptions. Individual characteristics were quantified and evaluated with a logistic regression to model their ability to correctly identify each parasite separately. Subsequently, all algorithms were evaluated for false positive cross reactivity with the other parasites studied, excepting Taenia sp. which shares very few morphological characteristics with the others. The principal result showed that our algorithm reached sensitivities between 99.10%-100% and specificities between 98.13%- 98.38% to detect each parasite separately. We did not find any cross-positivity in the algorithms for the three parasites evaluated. In conclusion, the results demonstrated the capacity of our computer algorithm to automatically recognize and diagnose Taenia sp., Trichuris trichiura, Diphyllobothrium latum, and Fasciola hepatica with a high sensitivity and specificity.

Classification of Suncus murinus species complex (Soricidae: Crocidurinae) in Peninsular Malaysia using image analysis and machine learning approaches

Background

Taxonomists frequently identify specimen from various populations based on the morphological characteristics and molecular data. This study looks into another invasive process in identification of house shrew (Suncus murinus) using image analysis and machine learning approaches. Thus, an automated identification system is developed to assist and simplify this task. In this study, seven descriptors namely area, convex area, major axis length, minor axis length, perimeter, equivalent diameter and extent which are based on the shape are used as features to represent digital image of skull that consists of dorsal, lateral and jaw views for each specimen. An Artificial Neural Network (ANN) is used as classifier to classify the skulls of S. murinus based on region (northern and southern populations of Peninsular Malaysia) and sex (adult male and female). Thus, specimen classification using Training data set and identification using Testing data set were performed through two stages of ANNs.

Results

At present, the classifier used has achieved an accuracy of 100% based on skulls’ views. Classification and identification to regions and sexes have also attained 72.5%, 87.5% and 80.0% of accuracy for dorsal, lateral, and jaw views, respectively. This results show that the shape characteristic features used are substantial because they can differentiate the specimens based on regions and sexes up to the accuracy of 80% and above. Finally, an application was developed and can be used for the scientific community.

Conclusions

This automated system demonstrates the practicability of using computer-assisted systems in providing interesting alternative approach for quick and easy identification of unknown species.

Identification and quantification of pathogenic helminth eggs using a digital image system

A system was developed to identify and quantify up to seven species of helminth eggs (Ascaris lumbricoides -fertile and unfertile eggs-, Trichuris trichiura, Toxocara canis, Taenia saginata, Hymenolepis nana, Hymenolepis diminuta, and Schistosoma mansoni) in wastewater using different image processing tools and pattern recognition algorithms. The system was developed in three stages. Version one was used to explore the viability of the concept of identifying helminth eggs through an image processing system, while versions 2 and 3 were used to improve its efficiency. The system development was based on the analysis of different properties of helminth eggs in order to discriminate them from other objects in samples processed using the conventional United States Environmental Protection Agency (US EPA) technique to quantify helminth eggs. The system was tested, in its three stages, considering two parameters: specificity (capacity to discriminate between species of helminth eggs and other objects) and sensitivity (capacity to correctly classify and identify the different species of helminth eggs). The final version showed a specificity of 99% while the sensitivity varied between 80 and 90%, depending on the total suspended solids content of the wastewater samples. To achieve such values in samples with total suspended solids (TSS) above 150 mg/L, it is recommended to dilute the concentrated sediment just before taking the images under the microscope. The system allows the helminth eggs most commonly found in wastewater to be reliably and uniformly detected and quantified. In addition, it provides the total number of eggs as well as the individual number by species, and for Ascaris lumbricoides it differentiates whether or not the egg is fertile. The system only requires basically trained technicians to prepare the samples, as for visual identification there is no need for highly trained personnel. The time required to analyze each image is less than a minute. This system could be used in central analytical laboratories providing a remote analysis service.

•

The system identifies and quantifies seven species of helminth eggs.

•

The system shows a specificity of 99% and a sensitivity between 80 and 90%.

•

The time required to analyze each image is less than a minute.

•

The system reduces the need for highly trained personnel for the identification of helminth eggs.

Automated parasite faecal egg counting using fluorescence labelling, smartphone image capture and computational image analysis

Intestinal parasites are a concern in veterinary medicine worldwide and for human health in the developing world. Infections are identified by microscopic visualisation of parasite eggs in faeces, which is time-consuming, requires technical expertise and is impractical for use on-site. For these reasons, recommendations for parasite surveillance are not widely adopted and parasite control is based on administration of rote prophylactic treatments with anthelmintic drugs. This approach is known to promote anthelmintic resistance, so there is a pronounced need for a convenient egg counting assay to promote good clinical practice. Using a fluorescent chitin-binding protein, we show that this structural carbohydrate is present and accessible in shells of ova of strongyle, ascarid, trichurid and coccidian parasites. Furthermore, we show that a cellular smartphone can be used as an inexpensive device to image fluorescent eggs and, by harnessing the computational power of the phone, to perform image analysis to count the eggs. Strongyle egg counts generated by the smartphone system had a significant linear correlation with manual McMaster counts (R(2)=0.98), but with a significantly lower coefficient of variation (P=0.0177). Furthermore, the system was capable of differentiating equine strongyle and ascarid eggs similar to the McMaster method, but with significantly lower coefficients of variation (P<0.0001). This demonstrates the feasibility of a simple, automated on-site test to detect and/or enumerate parasite eggs in mammalian faeces without the need for a laboratory microscope, and highlights the potential of smartphones as relatively sophisticated, inexpensive and portable medical diagnostic devices.

Automated identification of copepods using digital image processing and artificial neural network

Background

Copepods are planktonic organisms that play a major role in the marine food chain. Studying the community structure and abundance of copepods in relation to the environment is essential to evaluate their contribution to mangrove trophodynamics and coastal fisheries. The routine identification of copepods can be very technical, requiring taxonomic expertise, experience and much effort which can be very time-consuming. Hence, there is an urgent need to introduce novel methods and approaches to automate identification and classification of copepod specimens. This study aims to apply digital image processing and machine learning methods to build an automated identification and classification technique.

Results

We developed an automated technique to extract morphological features of copepods' specimen from captured images using digital image processing techniques. An Artificial Neural Network (ANN) was used to classify the copepod specimens from species Acartia spinicauda, Bestiolina similis, Oithona aruensis, Oithona dissimilis, Oithona simplex, Parvocalanus crassirostris, Tortanus barbatus and Tortanus forcipatus based on the extracted features. 60% of the dataset was used for a two-layer feed-forward network training and the remaining 40% was used as testing dataset for system evaluation. Our approach demonstrated an overall classification accuracy of 93.13% (100% for A. spinicauda, B. similis and O. aruensis, 95% for T. barbatus, 90% for O. dissimilis and P. crassirostris, 85% for O. similis and T. forcipatus).

Conclusions

The methods presented in this study enable fast classification of copepods to the species level. Future studies should include more classes in the model, improving the selection of features, and reducing the time to capture the copepod images.

Methods for Quantification of Soil-Transmitted Helminths in Environmental Media: Current Techniques and Recent Advances

Limiting the environmental transmission of soil-transmitted helminths (STHs), which infect 1.5 billion people worldwide, will require sensitive, reliable, and cost-effective methods to detect and quantify STHs in the environment. We review the state-of-the-art of STH quantification in soil, biosolids, water, produce, and vegetation with regard to four major methodological issues: environmental sampling; recovery of STHs from environmental matrices; quantification of recovered STHs; and viability assessment of STH ova. We conclude that methods for sampling and recovering STHs require substantial advances to provide reliable measurements for STH control. Recent innovations in the use of automated image identification and developments in molecular genetic assays offer considerable promise for improving quantification and viability assessment.

Automatic Identification of Human Erythrocytes in Microscopic Fecal Specimens

Traditional fecal erythrocyte detection is performed via a manual operation that is unsuitable because it depends significantly on the expertise of individual inspectors. To recognize human erythrocytes automatically and precisely, automatic segmentation is very important for extraction of characteristics. In addition, multiple recognition algorithms are also essential. This paper proposes an algorithm based on morphological segmentation and a fuzzy neural network. The morphological segmentation process comprises three operational steps: top-hat transformation, Otsu's method, and image binarization. Following initial screening by area and circularity, fuzzy c-means clustering and the neural network algorithms are used for secondary screening. Subsequently, the erythrocytes are screened by combining the results of five images obtained at different focal lengths. Experimental results show that even when the illumination, noise pollution, and position of the erythrocytes are different, they are all segmented and labeled accurately by the proposed method. Thus, the proposed method is robust even in images with significant amounts of noise.

A robust and automatic method for human parasite egg recognition in microscopic images

With the accelerated movement of population, human parasitoses become an increasingly serious public health's problem. Currently, detections of parasite eggs through microscopic images are still the golden standard for diagnoses. However, this conventional method relies heavily on the experiences of inspectors, thus giving rise to misdiagnoses and missed diagnoses occasionally. And, as the number of clinical specimens increases rapidly, manual identification seems impractical. Hence, a fully automatic method is in desperate need. In this paper, we propose a robust method to segment and recognize the parasite eggs. Their contours are extracted using phase coherence technology, and the support vector machine (SVM) method based on shape and texture features is employed to classification of parasite eggs. Our novel method was comparable to the traditional method. The overall recognition rate was up to 95%, and the overall robustness indexes, including si, fnvf, fvpf, tpvf, were 95.7, 4.9, 3.7, 95.1, respectively, suggesting that our method is effective and the robustness is good, which has great potential to become a diagnostic method in the parasitological clinic.

OvaSpec - A vision-based instrument for assessing concentration and developmental stage of Trichuris suis parasite egg suspensions

BACKGROUND

OvaSpec is a new, fully automated, vision-based instrument for assessing the quantity (concentration) and quality (embryonation percentage) of Trichuris suis parasite eggs in liquid suspension. The eggs constitute the active pharmaceutical ingredient in a medicinal drug for the treatment of immune-mediated diseases such as Crohn׳s disease, ulcerative colitis, and multiple sclerosis.

METHODS

This paper describes the development of an automated microscopy technology, including methodological challenges and design decisions of relevance for the future development of comparable vision-based instruments. Morphological properties are used to distinguish eggs from impurities and two features of the egg contents under brightfield and darkfield illumination are used in a statistical classification to distinguish eggs with undifferentiated contents (non-embryonated eggs) from eggs with fully developed larvae inside (embryonated eggs).

RESULTS

For assessment of the instrument׳s performance, six egg suspensions of varying quality were used to generate a dataset of unseen images. Subsequently, annotation of the detected eggs and impurities revealed a high agreement with the manual, image-based assessments for both concentration and embryonation percentage (both error rates <1.0%). Similarly, a strong correlation was demonstrated in a final, blinded comparison with traditional microscopic assessments performed by an experienced laboratory technician.

CONCLUSIONS

The present study demonstrates the applicability of computer vision in the production, analysis, and quality control of T. suis eggs used as an active pharmaceutical ingredient for the treatment of autoimmune diseases.

Cascaded-Automatic Segmentation for Schistosoma japonicum eggs in images of fecal samples

BACKGROUND

To recognize parasite eggs automatically, the automatic segmentation of parasite egg images is very important for the extraction of characteristics and genera classification.

METHODS

A Cascaded-Automatic Segmentation approach was proposed. Firstly, image contrast between the border of an egg and its background for all samples was strengthened by the Radon-Like Features algorithm and the enhanced image was processed into a binary image to get an initial set. Then, the elliptical targets are located with Randomized Hough Transform (RHT). The fitted data of an elliptical border are considered the initial border data and the accurate border of a Schistosoma japonicum egg can be finally segmented using an Active Contour Model (Snake).

RESULTS

Seventy-three cases of S. japonicum eggs in fecal samples were found; 61 images contained a parasite egg and 12 did not. Although the illumination, noise pollution, boundary definitions of eggs, and egg position are different, they are all segmented and labeled accurately.

DISCUSSION

The results proved that accurate borders of S. japonicum eggs could be recognized precisely using the proposed method, and the robustness of the method is good even in images with heavy noise. This indicates that the proposed method can overcome the disadvantages of the traditional threshold segmentation method, which has limited adaptability to images with heavy background noise.