ML regression-based predictive modeling for disease outbreak threshold estimation
Metadatos
Mostrar el registro completo del ítemEditorial
Universidad de Granada
Materia
Blood Smears Diagnostic accuracy Parasite detection Automated diagnosis
Fecha
2024-12-31Referencia bibliográfica
G. Prabhakar, Spurthi Vadlakonda, Vanam snehitha, Sravani Siddam (2024). ML regression-based predictive modeling for disease outbreak threshold estimation. Journal for Educators, Teachers and Trainers,Vol.15(5).435-443. ISSN 1989-9572
Resumen
Malaria diagnosis relied heavily on manual microscopy, where a skilled technician examines blood
smears under a microscope to identify and count malaria parasites. This method, established in the early
1900s, has been the gold standard for malaria diagnosis but is labor-intensive, time-consuming, and
requires significant expertise. In regions with limited healthcare resources, this has often led to
misdiagnosis or delayed treatment. The objective of this study is to leverage ML learning techniques to
develop an automated, accurate, and efficient diagnostic tool for detecting malaria infections from
medical images, thereby improving diagnostic accuracy and reducing the time required for analysis.
The title "ML Learning-Based Analysis for Malaria Infection Diagnosis" refers to the application of ML
learning algorithms, a subset of machine learning, to analyze medical images and diagnose malaria
infections. The approach aims to automate the detection process, making it faster and more reliable
compared to traditional methods. Before the advent of machine learning or AI, the primary method for
diagnosing malaria was manual microscopy, as mentioned earlier. This involved staining blood smears
with special dyes, followed by meticulous examination under a microscope. The accuracy of this
method largely depended on the technician’s experience and the quality of the equipment, which could
vary significantly, especially in low-resource settings. Traditional microscopy for malaria diagnosis,
while effective, has several limitations, including the need for skilled personnel, the potential for human
error, and the slow turnaround time for results. The motivation for this research stems from the need to
address the shortcomings of traditional malaria diagnostic methods. With the global burden of malaria
remaining high, especially in low-income countries, there is a pressing need for diagnostic tools that
are not only accurate but also accessible and scalable. The proposed system involves the development
of a ML learning model trained on a large dataset of labeled blood smear images. This model will
automatically detect and classify malaria parasites in the images, offering a quick and accurate diagnosis. By reducing reliance on human expertise, this system can provide consistent results across
different settings, enhance early detection, and enable prompt treatment, ultimately contributing to
better malaria control and eradication efforts.