This work (Research grant for New Scholar, Grant No. RGNS 65 - 212) was financially supported by the Office of the Permanent Secretary, Ministry of Higher Education, Science, Research and Innovation (OPS MHESI), Thailand Science Research and Innovation (TSRI) and King Mongkut&#39;s Institute of Technology Ladkrabang. We are grateful to the National Research Council of Thailand (NRCT) [NRCT5-RSA63001-10] for funding the research project. M.K. was funded by Thailand Science Research and Innovation Fund Chulalongkorn University. We also thank the College of Advanced Manufacturing Innovation, King Mongkut&#39;s Institute of Technology, Ladkrabang who have provided the deep learning platform and software to support the research project.

AI Program for Trypanosomiasis Detection and Classification

All authors have no financial disclosures and no conflicts of interest.

Microscopic observation for Trypanosoma protozoa infection is early and commonly used, especially during surveillance in remote areas where there is a lack of skilled technicians and labor-intensive and time-consuming processes that are all obstacles to reporting the health organization timely. Although molecular biology techniques such as immunology and polymerase chain reaction (PCR) have been approved as high-sensitivity methods to support the effectiveness of lab findings, expensive chemicals, apparatus, and professi...

Trypanosomiasis is a significant challenge to global health issues due to a variety of zoonotic species causing human disease with a wide range of geographical distribution outside the African and American continents, such as South and Southeast Asia1,2,3. Human African trypanosomiasis (HAT) or sleeping sickness, is caused by Trypanosoma brucei gambiense and T. b. rhodesiense which produce the chronic and acute forms, respectively, representing the major spread in Africa. The causative parasite belongs to the Salivaria group due to the transmission by infected saliva of Tsetse flies4. Whereas, the well-known American trypanosomiasis (Chagas&#39;s disease) caused by T. cruzi has been a public health concern for non-endemic countries; including Canada, the USA, Europe, Australia, and Japan, because of the frequent migration of individuals from endemic areas5. The trypanosome infection belongs to the Stercoraria group because it is transmitted by the infected feces of reduviid bugs. The trypanosomiases and trypanosomoses (Surra disease) caused by the T. evansi infection are endemic in Africa, South America, Western and Eastern Asia, and South and Southeast Asian countries3,6. Although human trypanosomiasis caused by the trypanosome has been reported3,4,7,8,9,10,11,12, the route of transmission of the parasite infection is debated: either the mechanical or infected blood through hematophagous insects such as tsetse flies and tabanids or horse flies6,7,8,9,10,12,13,14. No case report has been found in Thailand, however, a high prevalence of the T. evansi infection in dog15, racing horses, and water buffalo in the eastern region has been published16, suggesting an acquired transmission between domestic animals would have occurred. Several atypical human infections caused by animal trypanosomes (T. vivax, T. b. brucei, T. congolense, T. lewisi, and T. evansi) were reported, which are not the classical forms of human trypanosomes17. Awareness about atypical human infections might be underestimated, highlighting the need for improved diagnostic tests and field investigations for detection and confirmation of these atypical cases, and allowing for proper control and treatment of animal pathogenic diseases that affect global livestock, food security18, and human healthcare. This led to the development of a potential strategy integrated with an existing common method (microscopic examination) to rapidly screen blood samples in remote areas during active surveillance, enabling the identification of the hotspot zones for restricting and controlling the disease.
Having a sporadic incidence of the Surra disease in a wide range of domestic animals such as dromedaries, cattle, equines, and dogs that evoke a euryxenous T. evansi may be zoonotic to humans1,4,13,14. Human infection seems impossible because a trypanolytic factor in human serum, expressed from a sra-like gene, is capable of preventing human T. brucei and T. congolense12,19. Furthermore, as the first case report from India demonstrates, the illness has no association with immunocompromised HIV patients4. As described above, the possible human infection may be related to a high-density lipoprotein deficiency with abnormal function of the trypanosome lytic factor, which is a rare autosomal recessive genetic disorder, namely Tangier disease4. In 2016, a Vietnamese patient was discovered to possess two wild-type APOL1 alleles and a serum APOL1 concentration within the normal range. However, the theory of APOL-1 deficiency is no longer considered valid12. Therefore, one possible mechanism of trypanosome infection is direct contact of a wound with infected animal blood during occupational animal farming4,12. Microscopic examination reveals that T. evansi morphology is a monomorphic form of the trypomastigote including a predominant long slender, flagellated, and dividing trypanosome which is similar to their relative species of T. brucei1,12,13. The nucleus is in the central position with a visible small kinetoplast in the posterior position. A previous study indicated that the parasite can exist in two comparable forms, known as the classical and truncated forms. However, it remains necessary to confirm their respective pathogenic effects on hosts20. The course of symptoms varies ranging from intermittent fever associated with chills and sweating. Suramin, fortunately, is a successful first-line therapy for early-stage human African trypanosomiasis with no invasion of the central nervous system (CNS), healing patients in India and Vietnam4,12,21.
Except for clinical sign examination, several diagnostic methods for T. evansi parasites exist, including parasitological microscopic observation4,9,12, serological4,8,9,10,12, and molecular biological tests4,12. Thin-blood films stained with Giemsa are often used to visualize the parasite present under microscopic examination, which is routinely and commonly used22. However, the procedure appears to be feasible; nonetheless, it is time-consuming and labor-intensive, has inter-rater assessment variability, is sensitive to only an acute phase, and requires a personal trainee23. Both molecular biology and serological testing also needed highly skilled personnel to perform multiple processes of sample preparation, including extracting and purifying the samples before testing them with expensive apparatus, which is difficult to standardize, risk of contamination with extra-parasitic materials, and discrepancies in results24. Based on the rationale described above, rapid and early screening technology is needed to support the field surveillance study and ensure that the survey result is reported in a timely manner to identify the hotspot zone for further control of the disease transmission1,8. Computerized-based devices (CAD) have been proposed as an innovative technology for medical fields, including histopathological and cytopathological tasks25. The CAD mentioned above was performed at high speed and computed using pattern recognition, namely, artificial intelligence (AI). The AI method is accomplished using convolutional neural network algorithms that can be used to deal with a large number of dataset samples, especially, a supervised learning approach that trains a well-trained model upon data consumption.
In general, AI is the ability of computers to solve tasks that require expert intelligence, such as data labeling. Machine learning (ML), a subfield of AI, is represented as a computer system with two different processes comprised of feature extraction and pattern recognition. Deep learning (DL), or advanced ML algorithms, refers to the development of computerized programs and devices comparing human-like performance with levels of accuracy greater and equal to that accomplished by human professionals26. Currently, the role of DL in medical and veterinary fields is promisingly expanding and revolutionizing communicable disease prevention with the aim of recent prevention and guiding it to individual health staff22,27. The potential DL application is limitless with quality labels and a large number of augmented datasets, freeing specialists to manage the project task. Specifically, an advance in the digital image along with computer-assisted analysis, improved the automatic diagnostic and screening in five categories of pathology reported; including static, dynamic, robotic, whole slide imaging, and hybrid methods28. It is necessary to consider that the integration of DL algorithm approaches and digital image data could encourage local staff to utilize the technology in their daily practices.
Previously, the increase in prediction accuracy of using a hybrid model had been proven27. To identify the trypanosome parasite in microscopic images, this research presents two hybrid models, incorporating the YOLOv4-tiny (object detection) and Densenet201 (object classification) algorithms. Among several detection models, YOLOv4-tiny with a CSPDarknet53 backbone showed high performance as a prediction result in terms of localization and classification29. Since the real-time detector has modified the optimal balance among the input network resolution, the amount of the convolutional layer, the total parameter, and the number of layer outputs, it has improved prioritizing fast operating speeds and optimizing for parallel computations when compared to previous versions. Dense Convolutional Network (DenseNet) is another popular model that achieves state-of-the-art results across competitive datasets. DenseNet201 yielded a similar validation error comparable to that of ResNet101; however, DenseNet201 has fewer than 20 million parameters, which is less than ResNet101&#39;s more than 40 million parameters30. Therefore, the DenseNet model could improve prediction accuracy with an increasing number of parameters with no sign of overfitting. Here, an artificial intelligence (AI) program utilizes a hybrid deep learning algorithm with deep detection- and classification neural network backbones on the in-house CiRA CORE platform. The developed program can identify and classify the protozoan trypanosome species, namely Trypanosoma cruzi, T. brucei, and T. evansi, from oil-immersion microscopic images. This technology has the potential to revolutionize disease surveillance and control by providing a rapid, automated, and accurate screening method. It could aid local staff in making more informed decisions on transmission-blocking strategies for parasitic protozoan disease.

<table border="1" fo:keep-together.within-page="1" fo:keep-with-next.within-page="always">
	<tbody>
		<tr>
			<td>Darknet19, Darknet53 and Densenet201</td>
			<td>Gao Huang, Z. L., Laurens van der Maaten. Densely Connected Convolutional Networks. arXiv:1608.06993 [cs.CV]. (2016)</td>
			<td>https://github.com/liuzhuang13/DenseNet&#160;</td>
			<td>Deep convolutional neural network model that can function to&#160; classification 
			Generic name: YOLO model/ detection model?</td>
		</tr>
		<tr>
			<td>Olympus CX31 Model CX31RRBSFA</td>
			<td>&#160;Olympus, Tokyo, Japan</td>
			<td>SN 4G42178&#160;</td>
			<td>A light microscope&#160;</td>
		</tr>
		<tr>
			<td>Olympus DP21-SAL U-TV0.5XC-3</td>
			<td>&#160;Olympus, Tokyo, Japan</td>
			<td>SN 3D03838</td>
			<td>A digital camera 
			Generic name: Classification models/ densely CNNs</td>
		</tr>
		<tr>
			<td>Window 10</td>
			<td>Microsoft</td>
			<td>Window 10</td>
			<td>Operation system in computers</td>
		</tr>
		<tr>
			<td>YOLO v4-tiny&#160;</td>
			<td>Naing, K. M. et al. Automatic recognition of parasitic products in stool examination using object detection approach. PeerJ Comput Sci. 8 e1065, (2022).</td>
			<td>https://git.cira-lab.com/users/sign_in</td>
			<td>Deep convolutional neural network model that can function to both localization and also classification&#160;</td>
		</tr>
		<tr>
			<td></td>
			<td></td>
			<td>https://git.cira-lab.com/users/sign_in</td>
			<td></td>
		</tr>
	</tbody>
</table>

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&#39;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.

Author Spotlight: AI-Driven Trypanosome Species Detection from Microscopic Images 

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

The goal of the research is to develop an artificial intelligence AI model that makes use of a hybrid deep learning technique of object detection and classification backbones for defining the protozoa trypanosome specie, namely trypanosoma cruzi, T.brucei, and T.evansi, from oil immersion, microscopic images on the in-house, low-code AI platform CiRA CORE. The most recent development in our field of research is the development of the learning algorithms that can identify and classify trypanosome species from microscopic images. This program has the potential to revolutionize the surveillance and control by providing a rapid automated and accurate screening method that can be used by local staff in remote areas.Using the modified and hybrid algorithms between two different deep learning models within the purpose AI program can overcome many challenges. Identification of shared morphology, mixed and immature infection, and precise species characterization enables automatic standard taxonomy.

In this study, hybrid deep learning algorithms were proposed to help automatically predict the positivity of a blood sample with a trypanosome parasite infection. Archived, Giemsa-stained blood films were sorted to localize and classify the parasitized versus non-parasitic by using the object detection algorithm based on a darknet backbone neural network. Within any rectangular box prediction result obtained by the previous model, the best-selected classification model was developed to classify all three species of medic...

Watch this Scientific Journal Video about AI-Driven Trypanosome Species Detection from Microscopic Images at JoVE.com

Worldwide medical blood parasites were automatically screened using simple steps on a low-code AI platform. The prospective diagnosis of blood films was improved by using an object detection and classification method in a hybrid deep learning model. The collaboration of active monitoring and well-trained models helps to identify hotspots of trypanosome transmission.

Trypanosomiasis is a significant public health problem in several regions across the world, including South Asia and Southeast Asia. The identification of ...

superior-auto-identification-trypanosome-parasites-using-hybrid-deep

Research

JoVE Journal

Bioengineering

882 Views.  King Mongkut’s Institute of Technology Ladkrabang. The goal of the research is to develop an artificial intelligence AI model that makes use of a hybrid deep learning technique of object detection and classification backbones for defining the protozoa trypanosome specie, namely trypanosoma cruzi, T.brucei, and T.evansi, from oil immersion, microscopic images on the in-house, low-code AI platform CiRA CORE. The most recent development in our field of research is the development of the learning algorithms that can identify and classify trypanos...

Video: Author Spotlight: AI-Driven Trypanosome Species Detection from Microscopic Images 

Training Process CiRA CORE Platform and Object Detection Model Evaluation to Identify and Classify the Protozoan Trypanosome Species

To begin, move the deep train function from the general toolbar onto the screen to initiate data labeling and training. Click on the load images button within the deep train tool settings and navigate to the directory containing the images to import and label images. Then click and hold the left mouse button to label the objects and assign names to the selected objects.Click save GT to save a dot GT file in the same directory. Before starting the model training, click the gen setting button to access data augmentation features. Utilize four types of augmentations, namely rotation, contrast, noise, and blur to enrich the dataset.To begin the model training, click on the training button located within the deep train tool. In the generate training files function, choose the models, batch size, and subdivisions. Click the generate button to generate data and save it in the directory.After configuring all settings, click the train button to initiate the model training. The program will automatically execute and adjust the dataset weight while it assesses the training loss. Upon achieving optimal loss, click the export button and save the weight file in the directory.To proceed with object detection model evaluation, select evaluate in the plugin toolbar and move the eval detect function onto the screen. Afterward, click on the setting button. Wait for the three functions, detection, evaluate, and plot to appear.To start the model evaluation, click the load config button and import the trained wait file from the directory. To import test images from the image file directory, click the browse button. Then click on the load GT button to import the GT file.Next, click the evaluation button to assess the detection model in the directory. Upon completion, the results will be automatically saved as a CSV file in the same directory, sorted by class name, containing key parameters, such as true positive, false positive, false negative, recall, and precision for each class. To generate the precision recall or PR curve, navigate to the plot function.Click on browse to import the CSV files from the directory. Select the desired classes from the list and click the plot button to visualize the PR curve. Click on the save button to save an image with AUC values of the PR curve in the required image format at the selected directory.To select the image classification model training function, navigate to the image toolbar, select deep classif, and move classif train onto the screen. To import images for training, click on the open folder button within the classif train tool settings and navigate to the directory containing the images. Before training, enrich the dataset by clicking on the augmentation button and applying techniques such as rotation, contrast, flipping, noise, and blur.To initiate the model training process, click the gen train button within the classif train tool. Under gen train, select the desired models, batch size, and subdivisions. Assign a directory to save the generated data and click the generate button.Once all configurations are complete, click on the start button. The program will run continuously, assessing the training loss, and adjusting the weight of the dataset as needed. Upon achieving the desired loss level, click the export button and store the weight file in the specified directory.

Image Classification as Model Training and Model Evaluation to Identify and Classify the Protozoan Trypanosome Species

To begin, navigate to the plugin toolbar. Then drag and drop the EvaluateClassif function onto the screen. Click on the evaluate button located within the EvaluateClassif tool.Click the load folder image button to import test images, and load config to import the trained weight file from the directory. Then click the start button to evaluate the classification model. After evaluation, click the export to CSV button to save the results as a CSV file in the directory.To evaluate data at every threshold, click on the start all threshold to save the CSV file in the directory with class names, including parameters such as recall, true positive rate, false positive rate, and precision for each class. To plot the receiver operating characteristics curve, click on the plot ROC button located within the EvaluateClassif tool. Click on the browse button and import the CSV files from the directory.Inspect the imported class list and select each class label to plot the ROC curve. Next, to visualize the ROC curve, click on the plot button. Make any desired adjustments to the image properties, such as font size, font colors, rounding decimals, line styles, and line colors.Finally, click the save button to save the ROC curve image with the AUC values in the required image format in the directory.

Hybrid (Detection and Classification) as Model Testing to Identify and Classify the Protozoan Trypanosome Species

To begin, navigate to the general toolbar, click on Button Run, then click Debug to start the hybrid testing model. Next, move to the image toolbar, choose Acquisition and select Image Slide. Go to Deep Composite and click on Deep D-C.Before beginning the testing phase for the images, click on the setting option available in the image slide tool and import the test images. Then click the setting option in the deep D-C tool to import the two saved trained weight files. For the Detect function, click on the Config button, plus button, and select the backend CUDA or CPU.Provide a name and click OK.Select the weight file directory and click Choose. For the Classif function, click the Config button, followed by the plus button, and then select the backend. Provide a name and click OK.After this, select the directory for the weight file and click Choose.To view the test image results, click on the image function available in the Debug tool. Lastly, to evaluate the predicted results for each image, click on the Run button found in the Button Run tool.