Project overview:
In 2014 the United Nations set out to end the HIV/AIDs endemic by 2030 with the goal of having 90% of people living with HIV know their HIV status, be on retroviral treatment, and be virally suppressed. As of 2019, the most recent data, the United Nations program on HIV/AIDs reports that only 59% of people living with HIV are virally suppressed. HIV viral load testing is the key diagnostic in this plan; to achieve these goals, 30 million diagnostic tests will be needed each year. Yet a rapid, quantitative, and accessible diagnostic test is not currently available.
The Posner Research Group has developed a novel rapid quantitative DNA test that relies on discrete puncta counting in images. However, this method loses accuracy at high viral loads within the clinically relevant range. In this work, we aim to extend the dynamic range of this diagnostic test though the use of spatial temporal data and higher order models such as ResNets. In doing so, we aim to improve the quantitative accuracy across the dynamic range of this system and bring this method closer to clinical utility.
ABC
Figure 1 (A) Image of 80 copies of DNA. Selection of discrete are puncta circled in red. (B) Image of 100,000 copies of DNA. Representative instances of undesirable merged and or non-discrete puncta are circled in red. (C) Plot of puncta counts vs DNA copy number. Power regression shown in solid red line, 95% confidence interval shown in red highlight, and clinical range show n with grey shading.