Field evaluation of the diagnostic performance of EasyScan GO: a digital malaria microscopy device based on machine-learning

Table 3 EasyScan Go ‘competence level’ as per the WHO-TDR Research Malaria Microscopy criteria [21]

Competence Level	Parasite detection (%)	Species identification (%)	Parasite count within 25% of true count (%)	False positive rate (%)
1	90–100	90–100	50–100	≤ 2.5
2	80–89 [EasyScan Go]	80–89 [EasyScan Go]	40–49	≤ 5
3	70–79	70–79	30–39	≤ 10
4	0–69	0–69	0–29 [EasyScan Go]	> 10 [EasyScan Go]

Parasite detection corresponds to sensitivity and false positive rate corresponds to 1–specificity. In the WHO Malaria Microscopy Quality Assurance Manual, sensitivity and specificity are combined as overall parasite detection accuracy [7]. Per this criterion, the performance of the EasyScan Go (parasite detection 82.3%) corresponds to Competence Level 2

ISSN: 1475-2875