A novel approach for identification of zoonotic trypanosome utilizing deep metric learning and vector database-based image retrieval system.

Trypanosomiasis, a significant health concern in South America, South Asia, and Southeast Asia, requires active surveys to effectively control the disease. To address this, we have developed a hybrid model that combines deep metric learning (DML) and image retrieval. This model is proficient at identifying Trypanosoma species in microscopic images of thin-blood film examinations. Utilizing the ResNet50 backbone neural network, a trained-model has demonstrated outstanding performance, achieving an accuracy exceeding 99.71 % and up to 96 % in recall. Acknowledging the necessity for automated tools in field scenarios, we demonstrated the potential of our model as an autonomous screening approach. This was achieved by using prevailing convolutional neural network (CNN) applications, and vector database based-images returned by the KNN algorithm. This achievement is primarily attributed to the implementation of the Triplet Margin Loss function as 98 % of precision. The robustness of the model demonstrated in five-fold cross-validation highlights the ResNet50 neural network, based on DML, as a state-of-the-art CNN model as AUC >98 %. The adoption of DML significantly improves the performance of the model, remaining unaffected by variations in the dataset and rendering it a useful tool for fieldwork studies. DML offers several advantages over conventional classification model to manage large-scale datasets with a high volume of classes, enhancing scalability. The model has the capacity to generalize to novel classes that were not encountered during training, proving particularly advantageous in scenarios where new classes may consistently emerge. It is also well suited for applications requiring precise recognition, especially in discriminating between closely related classes. Furthermore, the DML exhibits greater resilience to issues related to class imbalance, as it concentrates on learning distances or similarities, which are more tolerant to such imbalances. These contributions significantly make the effectiveness and practicality of DML model, particularly in in fieldwork research.

Superior Auto-Identification of Trypanosome Parasites by Using a Hybrid Deep-Learning Model.

Trypanosomiasis is a significant public health problem in several regions across the world, including South Asia and Southeast Asia. The identification of hotspot areas under active surveillance is a fundamental procedure for controlling disease transmission. Microscopic examination is a commonly used diagnostic method. It is, nevertheless, primarily reliant on skilled and experienced personnel. To address this issue, an artificial intelligence (AI) program was introduced that makes use of a hybrid deep learning technique of object identification and object classification neural network backbones on the in-house low-code AI platform (CiRA CORE). The program can identify and classify the protozoan trypanosome species, namely Trypanosoma cruzi, T. brucei, and T. evansi, from oil-immersion microscopic images. The AI program utilizes pattern recognition to observe and analyze multiple protozoa within a single blood sample and highlights the nucleus and kinetoplast of each parasite as specific characteristic features using an attention map. To assess the AI program's performance, two unique modules are created that provide a variety of statistical measures such as accuracy, recall, specificity, precision, F1 score, misclassification rate, receiver operating characteristics (ROC) curves, and precision versus recall (PR) curves. The assessment findings show that the AI algorithm is effective at identifying and categorizing parasites. By delivering a speedy, automated, and accurate screening tool, this technology has the potential to transform disease surveillance and control. It could also assist local officials in making more informed decisions on disease transmission-blocking strategies.

Complete mitochondrial genome analyses confirm that bat Polychromophilus and ungulate Plasmodium constitute a distinct clade independent of other Plasmodium species.

In recent phylogenetic studies, bat Polychromophilus and ungulate Plasmodium, two relatively understudied haemosporidian parasites within the Apicomplexa phylum, have often been overlooked. Instead, the focus has been primarily on haemosporidian parasites in primates, rodents, and birds. Several phylogenetic analyses of bat Polychromophilus have relied on limited datasets and short informative DNA sequences. As a result of these inherent limitations, the substantiation of their evolutionary stance has encountered a diminished degree of robust validation. This study successfully obtained complete mitochondrial genome sequences from 11 Polychromophilus parasites originating from Hipposideros gentilis and Myotis siligoensis bats for the first time. Additionally, the authors have sequenced the apicoplast caseinolytic protease C genes from Polychromophilus murinus and a potentially new Polychromophilus species. These mitochondrial genomes range in length from 5994 to 6001 bp and consist of three protein-coding genes (PCGs), seven small subunit ribosomal RNA genes (SSU rRNA), 12 large subunit ribosomal RNA genes (LSU rRNA), and seven miscellaneous RNA genes. Phylogenetic analyses using Bayesian Inference and Maximum Likelihood methods indicated robust support for the grouping of ungulate Plasmodium and bat Polychromophilus in a single clade separate from other Plasmodium spp., confirming previous reports, albeit with stronger evidence in this study. The divergence between Polychromophilus in bats and Plasmodium in ungulates occurred approximately 29.61 to 55.77 million years ago (Mya), with a node age estimated at 40.63 Mya. These findings highlight that the genus Plasmodium, which includes species found in ungulates, birds, reptiles, and other mammals, does not form a monophyletic group. By incorporating Polychromophilus in bats and Plasmodium in ungulates, this study contributes significantly to understanding the phylogenetic relationships within the Haemosporida order. It provides valuable insights into the evolutionary history and interconnections among these diverse parasites, thereby expanding knowledge in this field.

Zoonotic simian malaria parasites in free-ranging Macaca fascicularis macaques and human malaria patients in Thailand, with a note on genetic characterization of recent isolates.

Despite the natural occurrences of human infections by Plasmodium knowlesi, P. cynomolgi, P. inui, and P. fieldi in Thailand, investigating the prevalence and genetic diversity of the zoonotic simian malaria parasites in macaque populations has been limited to certain areas. To address this gap, a total of 560 long-tailed macaques (Macaca fascicularis) and 20 southern pig-tailed macaques (M. nemestrina) were captured from 15 locations across 10 provinces throughout Thailand between 2018 and 2021 for investigation of malaria, as were 15 human samples residing in two simian-malaria endemic provinces, namely Songkhla and Satun, who exhibited malaria-like symptoms. Using PCR techniques targeting the mitochondrial cytb and cox1 genes coupled with DNA sequencing, 40 long-tailed macaques inhabiting five locations had mono-infections with one of the three simian malaria species. Most of the positive cases of macaque were infected with P. inui (32/40), while infections with P. cynomolgi (6/40) and P. knowlesi (2/40) were less common and confined to specific macaque populations. Interestingly, all 15 human cases were mono-infected with P. knowlesi, with one of them residing in an area with two P. knowlesi-infected macaques. Nucleotide sequence analysis showed a high level of genetic diversity in P. inui, while P. cynomolgi and P. knowlesi displayed limited genetic diversity. Phylogenetic and haplotype network analyses revealed that P. inui in this study was closely related to simian and Anopheles isolates from Peninsular Malaysia, while P. cynomolgi clustered with simian and human isolates from Asian countries. P. knowlesi, which was found in both macaques and humans in this study, was closely related to isolates from macaques, humans, and An. hackeri in Peninsular Malaysia, suggesting a sylvatic transmission cycle extending across these endemic regions. This study highlights the current hotspots for zoonotic simian malaria and sheds light on the genetic characteristics of recent isolates in both macaques and human residents in Thailand.

Establishment of molecular diagnostics targeting the 23S ribosomal RNA gene for the detection of Mycoplasma suis infection in Thai domestic pigs.

Mycoplasma (M.) suis is a pathogenic hemotropic Mycoplasma sp. that causes acute hemolytic anemia or chronic infection in pigs. M. suis infection can be diagnosed using several methods, including molecular diagnosis such as conventional PCR (cPCR) and quantitative PCR (qPCR). In these cases, the common target is the 16S rRNA gene; however, this genetic marker cannot distinguish hemoplasma at the species level owing to high sequence identity. Therefore, the 23S rRNA gene has emerged as another target gene. Other than PCR, the loop-mediated isothermal amplification (LAMP) method can be applied for M. suis. The objective of the present study was to establish cPCR, TaqMan qPCR, and LAMP assays in which the 23S rRNA gene is used to detect M. suis infection in Thai domestic pigs. The analytical sensitivity of cPCR was determined as 7.46 × 104 copies/µl of plasmid DNA, whereas those of qPCR and LAMP were 7.46 × 102 copies/µl. There was no cross reaction with other pathogens in any of the assays. To evaluate the diagnostic performance of the assays, they were tested using 173 samples of genomic DNA. The detection percentage of M. suis infection was 24.86% (43/173; 95% CI: 18.61%-31.89%), 28.32% (49/173; 95% CI: 21.75%-35.66%), and 29.48% (51/173; 95% CI: 22.80%-36.88%) using cPCR, qPCR, and LAMP, respectively. Using qPCR as a reference assay, cPCR showed 81.63% sensitivity, 97.58% specificity, and an almost perfect level of agreement (kappa = 0.823). In comparison, LAMP showed 77.55% sensitivity, 89.52% specificity, and a substantial level of agreement (kappa = 0.662). All assays tested here could be applied in veterinary diagnostic laboratories for monitoring porcine health in the herds. Furthermore, the LAMP assay could be used as a screening test in farm practice without the need for any special equipment.

Molecular detection and genetic analysis of porcine haemoplasmas in commercial pig farms from Thailand reveal a putative novel species.

Haemoplasma is a trivial name for haemotropic Mycoplasma spp., which can attach to the surface of red blood cells leading to deformity and anaemia in a wide range of mammalian animals, including pigs. In Thailand, there is only one study that reported the occurrence of Mycoplasma suis without other haemoplasma species. In this study, we examined the molecular occurrence and genetic diversity of porcine haemoplasmas in Thai domestic pigs (Sus scrofa domesticus) from commercial farms using a PCR assay targeting the 16S rRNA gene, DNA sequencing, nucleotide sequence type (ntST) analysis and phylogenetic analysis. A total of 665 blood samples were collected from pigs at thirteen farms located in eight provinces of Thailand during 2019-2020. Genomic DNA was extracted from blood samples and tested by PCR. The frequency of haemoplasma infection was 37.1% (247/665, 95% CI: 33.5%-40.9%) in all pigs. Among 247 PCR positive samples, 194 were sequenced and analysed by nucleotide BLAST, ntST diversity, phylogenetic trees and ntST networks. The results of this genetic analysis indicated that at least four species with 27 nucleotide sequence types (Mycoplasma suis, Mycoplasma parvum, Candidatus Mycoplasma haemosuis and a putative novel species) of porcine haemoplasmas were identified. Thus, it appears that haemoplasmas show a high genetic diversity in the Thai pig population. In addition, a putative novel species was genetically characterized by other markers, namely, the 23S rRNA and RNase P RNA (rnpB) genes. For phylogenetic analysis, Candidatus Mycoplasma haemosuis was placed into the Mycoplasma haemofelis group, and the three remaining species were placed into the Mycoplasma suis group in all trees containing the 16S rRNA, 23S rRNA and rnpB genes. Further studies, such as pathobiology and epidemiology, should be conducted to better characterize this putative novel species.

Classification for avian malaria parasite Plasmodium gallinaceum blood stages by using deep convolutional neural networks.

The infection of an avian malaria parasite (Plasmodium gallinaceum) in domestic chickens presents a major threat to the poultry industry because it causes economic loss in both the quality and quantity of meat and egg production. Computer-aided diagnosis has been developed to automatically identify avian malaria infections and classify the blood infection stage development. In this study, four types of deep convolutional neural networks, namely Darknet, Darknet19, Darknet19-448 and Densenet201 are used to classify P. gallinaceum blood stages. We randomly collected a dataset of 12,761 single-cell images consisting of three parasite stages from ten-infected blood films stained by Giemsa. All images were confirmed by three well-trained examiners. The study mainly compared several image classification models and used both qualitative and quantitative data for the evaluation of the proposed models. In the model-wise comparison, the four neural network models gave us high values with a mean average accuracy of at least 97%. The Darknet can reproduce a superior performance in the classification of the P. gallinaceum development stages across any other model architectures. Furthermore, the Darknet has the best performance in multiple class-wise classification, with average values of greater than 99% in accuracy, specificity, and sensitivity. It also has a low misclassification rate (< 1%) than the other three models. Therefore, the model is more suitable in the classification of P. gallinaceum blood stages. The findings could help us create a fast-screening method to help non-experts in field studies where there is a lack of specialized instruments for avian malaria diagnostics.

Development and application of a novel multiplex PCR assay for the differentiation of four haemosporidian parasites in the chicken Gallus gallus domesticus.

Haemosporidian infections in domestic chickens (Gallus gallus domesticus) are not only widely prevalent but also cause economic loss. Diagnosis is usually made by microscopic examination; however, the method has several drawbacks such as requiring an experienced microscopist, being unreliable when parasitemia is low and being unable to accurately differentiate between co-infections from multiple parasite species. Therefore, the current extent of haemosporidian infections might be underestimated and neglected. We have developed a novel multiplex PCR assay to simultaneously detect and differentiate between four haemosporidian parasites: Leucocytozoon caulleryi, Leucocytozoon sabrazesi, Plasmodium juxtanucleare and Plasmodium gallinaceum. Primers in the present study specifically amplified the corresponding targets with no cross-species amplification detected. The multiplex PCR exhibited a significantly greater detection rate when compared with microscopic examination (p = 0.0001). The results demonstrate that the detection rate of multiplex PCR for L. sabrazesi, P. juxtanucleare, and P. gallinaceum are all greater than that of microscopic examination with p = 0.002, 0.0001 and 0.004, respectively. Co-infections were also detected more effectively by multiplex PCR. We applied the current method to field samples originating from Nan, Prachinburi, and Chachoengsao Provinces. The current study revealed that positive rates of haemosporidian parasites in chickens in the three study sites ranging from 39.5%-93.8%. The present assay offers a timesaving option for molecular diagnosis instead of using singleplex PCRs for detecting the parasites individually. Within a single reaction, this assay would be a useful tool for the detection of avian haemosporidian parasites either single or under co-infection conditions and for large-scale epidemiology studies.

First molecular investigation of haemosporidian parasites in Thai bat species.

Malaria parasites in the phylum Apicomplexa (Order: Haemosporida) infect diverse vertebrates and invertebrate hosts. At least seven genera of haemosporidian parasites have been described to exclusively infect bats. Most of these parasites remain enigmatic with a poorly known host range. Here, we investigated 271 bats belonging to 21 species and seven families from six provinces of Thailand. Overall, 124 out of 271 bats (45.8%) were positive for haemosporidian parasites, while none had Plasmodium, based on microscopic examination of blood smears and PCR amplification. We obtained 19 distinct cytochrome b (cytb) nucleotide haplotypes of Hepatocystis from seven bat species (families: Craseonycteridae, Hipposideridae, Pteropodidae, and Rhinolophidae). Nycteria was found in four bat species (Craseonycteridae, Emballonuridae, Megadermatidae, and Pteropodidae) and Polychromophilus in two species (Emballonuridae, Vespertilionidae). Phylogenetic analysis inferred from cytb sequences placed Hepatocystis into 2 different clades. Most Hepatocystis infections were found in insectivorous bats and clustered together with a sequence from Hipposideros larvatus in Cambodia (in subclade 1a). A single sequence of Hepatocystis obtained from a frugivorous bat, Cynopterus brachyotis, was placed in the same clade with Hepatocystis from the same bat species previously reported in Malaysia (clade 2). Nycteria in these Thai bats were clearly separated from the African isolates previously reported in bats in the family Rhinolophidae. Polychromophilus murinus from Myotis siligorensis was placed in a distinct clade (clade 2) from Polychromophilus melanipherus isolated from Taphozous melanopogon (clade 1). These results confirmed that at least two distinct species of Polychromophilus are found in Thailand. Collectively, Hepatocystis presented no host specificity. Although Megaderma spasma seemed to be infected by only Nycteria, its respective parasite does not show specificity to only a single bat host. Polychromophilus murinus and P. melanipherus seem to infect a narrower host range or are somehow restricted to bats in the families Vespertilionidae and Emballonuridae, respectively.