Date of Completion
Martin G. Cherniack; Robert A. Levine
Field of Study
Master of Science
Malaria has profoundly influenced human history for over four thousand years and despite numerous attempts at eradication, the prevention, diagnosis, and treatment of malaria have been largely ineffective. More than five hundred million people are affected by malaria every year resulting in over one million deaths. Drug resistance development by the parasite has diminished the effectiveness of numerous treatment options due, in part, to overtreatment of negative patients based on insufficient clinical algorithms and diagnostic methods. The goal of this research was to develop an image analysis algorithm to diagnose malaria with a high degree of sensitivity and specificity in addition to performing auxiliary features that offer substantial clinical utility.
The image analysis algorithm was initially implemented using a novel point-of-care hematology analyzer, known as the Abbott Laboratories Imaging Platform (ALIP), for the computation of complete blood counts. This facilitated a simultaneous malaria diagnosis and complete blood count that can be used to con_rm or reject malaria infection based on hemoglobin and platelet results. The image analysis algorithm exhibited a sensitivity of 100% and a specificity of 97.7%, outperforming the most commonly used malaria diagnostic instruments. The parasitemia of malaria specimen was automatically calculated to within 0.13% of the actual parasitemia determined by microscopy review. Infected red blood cell indices were calculated and showed a negligible variance in hemoglobin content, but a slight increase in cell volume due to the inclusion of the intraerythrocytic parasite. All four erythrocytic stages were identifiable using the ALIP by the detection of intraerythrocytic hemozoin. All information may be presented to the clinician in less time than any current malaria diagnostic method.
The supplementary information made available by a device like the ALIP allows clinicians to make a more informed diagnosis and determine appropriate treatment methods on a patient-by-patient basis. The application of the image processing methodology with ALIP-like devices will contribute to the early detection of malaria, reduce disease transmission, limit unnecessary exposure to antimalarial drugs, improve resource waste in endemic regions, and has the capacity to save over 100,000 lives annually.
Jorgensen, Michael B., "Automated Point-of-Care Image Processing Methodology for the Diagnosis of Malaria" (2013). Master's Theses. 387.
Donald R. Peterson