AI-In-Medicine: Scientist Develop AI-Powered Schistoscope To Help Diagnosis Of Urogenital Schistosomiasis

AI-In-Medicine: Schistosomiasis, a parasitic disease caused by blood flukes of the genus Schistosoma, affects millions of people worldwide and presents a significant public health challenge, particularly in impoverished regions. Endemic in 76 countries, with an estimated 252 million people infected and 779 million at risk of infection, schistosomiasis is a disease of poverty that disproportionately impacts vulnerable populations in Sub-Saharan Africa. This disease is characterized by chronic and debilitating symptoms, including anemia, organ damage, and delayed physical and cognitive development in children.


Schematics of the proposed two-stage diagnosis framework urogenital schistosomiasis with DeepLabV3-MobileNetV3 deep learning architecture for semantic segmentation of eggs and refined segmentation for overlapping eggs separation and count

Furthermore, schistosomiasis increases the risk of HIV transmission, highlighting the urgent need for accurate and accessible diagnostic tools to support control and elimination efforts.
 
Addressing the global burden of schistosomiasis and achieving the World Health Organization (WHO) control and elimination targets require innovative approaches to disease diagnosis. Currently, microscopy is the WHO-recommended method for diagnosing schistosomiasis, involving the microscopic examination of urine or stool samples for the presence of parasite eggs. However, this approach has limitations, including its labor-intensive nature, dependence on trained personnel, and potential for human error, making it challenging to implement in resource-limited settings.
 
To overcome these challenges, a team of researchers has developed a groundbreaking solution: the AI-powered Schistoscope. This optical device integrates cutting-edge technology with deep learning algorithms to revolutionize the diagnosis of urogenital schistosomiasis. The Schistoscope is equipped with an autofocusing and automated slide scanning system, enabling rapid and accurate capture of microscopy images from urine samples. This innovation streamlines the diagnostic process, reduces the need for specialized expertise, and offers a cost-effective solution for disease detection in low-resource settings.
 
Central to the success of the Schistoscope is its ability to accurately identify and quantify Schistosoma haematobium (SH) eggs, a common cause of urogenital schistosomiasis. In a study published in the Journal of Medical Imaging, the researchers developed a robust dataset consisting of over 12,000 images of urine samples collected from a rural area in central Nigeria. These images were meticulously annotated to differentiate between SH eggs and potential artifacts that could hinder accurate diagnosis, such as crystals, glass debris, air bubbles, and fibers.
 
The diagnostic process is orchestrated by a sophisticated two-stage framework. In the first stage, a deep convolutional neural network (CNN) architecture, DeepLabv3 with a MobilenetV3 backbone, performs semantic segmentation to identify candidate SH eggs within the captured images. This stage effectively discerns the eggs from the background and artifacts, laying the foundation for accurate detection. The second stage involves a refined segmentation process, wherein overlapping eggs are separated through the fitting of overlapping ellipses. This approach enha nces the precision of egg counting by distinguishing boundaries of clustered eggs, addressing a critical challenge in accurate diagnosis.
 
The effectiveness of the proposed diagnostic framework was rigorously tested in real-world field settings. The researchers implemented the framework on an edge AI system, combining a Raspberry Pi and a Coral USB accelerator, and applied it to 65 clinical urine samples obtained from Nigeria.
 
The results demonstrated exceptional sensitivity, specificity, and precision percentages of 93.75, 93.94, and 93.75, respectively. Notably, the automated egg count closely mirrored manual counts conducted by an expert microscopist, affirming the reliability and accuracy of the AI-powered diagnostic approach.
 
The implications of this research extend beyond technological innovation, impacting global health efforts and underscoring the potential for AI to transform disease diagnosis and management. The AI-powered Schistoscope and the diagnostic framework offer a promising solution to the challenges posed by urogenital schistosomiasis, particularly in regions with limited resources. The ability to rapidly and accurately detect SH eggs is a critical step towards timely intervention, enabling healthcare professionals to map disease prevalence, monitor the effectiveness of control programs, and make informed decisions regarding mass drug administration.
 
Dr Jan Carel Diehl, the study's corresponding author and a professor at Delft University of Technology-Netherlands, highlights the significance of this work in addressing a pressing global health issue. By providing an automated and accurate diagnostic tool, the AI-powered Schistoscope contributes to the fight against schistosomiasis, aiming to alleviate the suffering of millions and promote sustainable development in endemic regions.
 
Dr Diehl told AI In Medicine reporters at TMN, “By automating the egg detection process, the Schistoscope and the proposed diagnostic framework offer a promising solution for the rapid and accurate diagnosis of urogenital schistosomiasis, particularly in low-resource settings. Future studies will further validate the framework's performance and compare it with other diagnostic methods, such as schistosome circulating antigen detection and DNA-based assays, to establish its role in schistosomiasis monitoring and control."
 
Looking ahead, the potential of the AI-powered Schistoscope extends beyond schistosomiasis to other infectious diseases and healthcare challenges. The successful integration of AI, deep learning, and optical technology exemplifies the transformative power of interdisciplinary research and collaboration. As technology continues to evolve, the AI-powered Schistoscope paves the way for a future where advanced diagnostics and healthcare interventions are accessible to all, regardless of their socio-economic status or geographical location.
 
In conclusion, the AI-powered Schistoscope represents a significant advancement in the field of medical diagnostics, particularly in the context of urogenital schistosomiasis. By combining cutting-edge technology with deep learning algorithms, this innovative device offers a rapid, accurate, and cost-effective solution for disease detection in resource-limited settings. As global health efforts continue to prioritize the control and elimination of schistosomiasis, the AI-powered Schistoscope emerges as a beacon of hope, promising a brighter and healthier future for millions of individuals affected by this debilitating disease.
 
The study findings were published in the peer reviewed Journal of Medical Imaging.
https://www.spiedigitallibrary.org/journals/journal-of-medical-imaging/volume-10/issue-04/044005/Two-stage-automated-diagnosis-framework-for-urogenital-schistosomiasis-in-microscopy/10.1117/1.JMI.10.4.044005.full
 
For the latest on AI In Medicine, keep on logging to Thailand Medical News.