Every year 500,000 people die due to Malaria and more than half the world's population is at risk of getting infected. In developing countries, diagnostic centres are often not accessible by rural populations and many of them are understaffed.
Current mistrust in diagnostic procedures (rapid diagnostic tests) leads to the preventive use of malaria medication; while manual microscopy (the gold standard) is very time consuming, expensive, requires special expertise and training.The World Health Organisation (WHO) recommends a nurse to patient ratio of 2,5:1000, while in Uganda it is 6:100000, thus being 4000% above the recommendation.
The Excelscope aims at reducing workload of medical staff. The imaging is based on smartphones, and is designed to be locally repairable. The product is fully automated, where Machine vision algorithms detect the parasites in any given blood sample.
The product was designed as a team project as part of the Advanced Embodiment Design course at TU Delft, in association with Leiden University of Medical Sciences
We are enabling the transition from manual microscopy to automated microscopy while increasing accuracy and reducing time and cost in the process. Other algorithms in existence require professional microscopes (approx. 3000€), whilst cheap microscopes (Foldscope - 1€) require enduring manual labour. We have combined the two and developed a solution that is more accessible, accurate, reliable and most importantly reduces the workload for medical professionals.
The Excelscope uses a small optical lens, mounted onto a smartphone camera to magnify and identify malaria parasites with a resolution of 1μm in blood samples. Underneath lies the blood sample on top of a 3-axes moving system, fine-tuned to achieve micro stepping of 100μm. This enables to analyze 800 field of views, complying with the WHO’s recommendation for declaring a person malaria free. For every step, the phone camera is triggered by the inbuilt electronics. An app based on machine vision algorithms then detects and counts the numbers of parasites per unit Red Blood Cells.