Introduction
The World Health Organization (WHO) announced a “T3: Test. Treat. Track” initiative on World Malaria Day in 2012, with the goal of scaling up diagnostic, treatment, and surveillance systems in malaria-endemic countries.1 A central tenant of this initiative revolves around universal access to quality malaria diagnostics with both rapid diagnostic tests (RDTs) and microscopy.2 The WHO recommends that all suspected cases of malaria be evaluated by either an RDT or microscopy before initiating antimalarial treatment. Ideally, the two diagnostic tests are meant to be used in a complementary manner, as each has benefits and limitations.1,3,4 Malaria RDTs are designed for rapid, point-of-care malaria diagnoses; however, they are not ideal for quantifying parasitemia. These tests are designed to determine the presence or absence of infection and to dichotomize cases as either Plasmodium falciparum or a non-P. falciparum malaria species. Microscopy has the capabilities to speciate and quantify malaria diagnoses, but issues surround the general lack of availability, poor maintenance of microscopes, and training of microscopists in malaria-endemic zones.3,5–7
Microscopes are invaluable tools in clinical and public health settings. For example, in areas holoendemic for malaria, microscopy results may push clinicians to think about non-malaria diagnoses in febrile patients with a positive RDT but very low parasitemia. In addition, light microscopy is crucial in epidemiologic surveys and hospitals where accurate speciation and quantification of samples is necessary. Handheld microscopy and mobile phone microscopy have recently been evaluated to overcome some of the logistic issues with laboratory-based diagnostics in resource-constrained settings.8–13 The aim of these compact diagnostic tools is to provide portable quality laboratory testing in rural, remote, or underserviced settings rather than transferring ill individuals (or specimens) to distant reference laboratories. The goal of this study was to evaluate the “real-world” diagnostic operating characteristics of a handheld light microscope with mobile phone attachment integrated into a community-based screening program for malaria in rural Côte d'Ivoire.
Methods
The current study was embedded into an ongoing community-based cross-sectional epidemiologic study in the town of Grand Moutcho, Côte d'Ivoire (longitude 5.961, latitude −4.181), between April and June 2014. Ethical approval was granted by the Ministry of Health and Public Hygiene of Côte d'Ivoire (reference no. 32/MSLS/CNER-dkn). We obtained written informed consent from those older than 18 years and by the parents or legal guardians of children, who assented orally to participation. Those older than 6 years were invited to provide finger prick blood samples.
Before initiating the study, four Ivorian laboratory technicians received a half-day training with the Newton Nm1 microscope (Newton Microscopes; Cambridge, United Kingdom). This is a commercially available, handheld monocular light microscope with modular lenses (including a 100× lens) that weighs 480 g (Figure 1; www.newtonmicroscopes.com, accessed March 30, 2016). We attached an iPhone 5s (Apple, Cupertino, CA) to the microscope with the adaptor included with the Newton microscope. The camera function of the mobile phone is active while slides are maneuvered on the microscope stage, and images of the slide are visualized mobile phone screen in real time. The training involved a didactic session outlining how the handheld microscope functioned with and without a mobile phone attachment. This was followed by an interactive, hands-on teaching session where laboratory technicians had the opportunity to practice on at least 20 malaria test slides. The process was repeated until the disagreement between the expert microscopist and the technician was less than 10% on test slides. All training was under the direct supervision of an expert microscopist familiar with the device.
The Newton Nm1 microscope in use, mounted on a tripod, at a rural clinic located in Grand Moutcho, Côte d'Ivoire.
Citation: The American Society of Tropical Medicine and Hygiene 95, 4; 10.4269/ajtmh.16-0328
The Newton Nm1 microscope in use, mounted on a tripod, at a rural clinic located in Grand Moutcho, Côte d'Ivoire.
Citation: The American Society of Tropical Medicine and Hygiene 95, 4; 10.4269/ajtmh.16-0328
The Newton Nm1 microscope in use, mounted on a tripod, at a rural clinic located in Grand Moutcho, Côte d'Ivoire.
Citation: The American Society of Tropical Medicine and Hygiene 95, 4; 10.4269/ajtmh.16-0328
A single Giemsa-stained thick blood film was prepared from 223 individuals as per standardized protocols.14,15 Laboratory technicians initially read each slide using an Olympus CX21 microscope (Olympus, Volketswil, Switzerland) under 1,000× magnification with oil. All slides were evaluated on the same day of collection. Microscopists quantified malaria parasitemia per 200 white blood cells (WBCs), and in cases where fewer than 10 protozoa were visualized, parasitemia was quantified to 500 WBCs. Data were directly entered into an Excel database (Microsoft, Redmond, WA), and 10% of all slides were reexamined by a senior expert microscopist blinded to prior results using the Olympus CX21 microscope for quality control and validation. In cases of disagreement with the initial reading, the results were discussed with the individual technician, and the slides were read an additional time by both the laboratory technician and senior microscopist until an agreement was reached. All slides were coded and subsequently evaluated by laboratory technicians blinded to prior results using the Newton Nm1 microscope (1,000× magnification with oil) with mobile phone attachment, as per the identical procedure listed above. We calculated the sensitivity, specificity, and positive and negative predictive values of the Newton Nm1 microscope with mobile phone attachment. In addition, linear association of malaria parasitemia calculated with the Newton Nm1 and the conventional microscope was assessed by Pearson's correlation coefficient. Quantitative agreement in parasitemia estimates obtained with the two diagnostic tools was further assessed by Bland–Altman plotting. All analyses were done using R (R Foundation for Statistical Computing, Vienna, Austria).
Results
We obtained 223 thick blood films from 223 individuals. The average age was 10 years (range: 5–19), with 108 (48.4%) women. Of 223 samples, 162 (72.6%) were positive for P. falciparum infection by conventional microscopy. Of the positive samples, 65 (40.1%) were of low parasitemia requiring counting to 500 WBC, and the overall mean parasite count was 1,392/μL of blood. Sensitivity and specificity of the Newton Nm1 microscope with mobile phone attachment were 80.2% (95% confidence interval [CI]: 73.1–85.9%) and 100% (95% CI: 92.6–100.0%), respectively, with a positive and negative predictive value of 100.0% (95% CI: 96.4–100.0%) and 65.6% (95% CI: 54.9–74.9%), respectively. Pearson's correlation comparing conventional microscopy with the Newton Nm1 microscope was 0.997. Despite strong linear correlation of estimates between the two devices, Bland–Altman plotting revealed that Newton Nm1 microscopy underestimated parasitemia, particularly at low parasitemia levels (Figure 2). A scatter plot comparing conventional light microscopy with the Newton Nm1 handheld microscope for P. falciparum diagnosis is shown in Figure 3.
Bland–Altman plot comparing conventional light microscopy with the Newton Nm1 handheld microscope for the diagnosis of Plasmodium falciparum in a community-based cross-sectional screening program.
Citation: The American Society of Tropical Medicine and Hygiene 95, 4; 10.4269/ajtmh.16-0328
Bland–Altman plot comparing conventional light microscopy with the Newton Nm1 handheld microscope for the diagnosis of Plasmodium falciparum in a community-based cross-sectional screening program.
Citation: The American Society of Tropical Medicine and Hygiene 95, 4; 10.4269/ajtmh.16-0328
Bland–Altman plot comparing conventional light microscopy with the Newton Nm1 handheld microscope for the diagnosis of Plasmodium falciparum in a community-based cross-sectional screening program.
Citation: The American Society of Tropical Medicine and Hygiene 95, 4; 10.4269/ajtmh.16-0328
Scatter plot comparing conventional light microscopy with the Newton Nm1 handheld microscope for the diagnosis of Plasmodium falciparum in a community-based cross-sectional screening program.
Citation: The American Society of Tropical Medicine and Hygiene 95, 4; 10.4269/ajtmh.16-0328
Scatter plot comparing conventional light microscopy with the Newton Nm1 handheld microscope for the diagnosis of Plasmodium falciparum in a community-based cross-sectional screening program.
Citation: The American Society of Tropical Medicine and Hygiene 95, 4; 10.4269/ajtmh.16-0328
Scatter plot comparing conventional light microscopy with the Newton Nm1 handheld microscope for the diagnosis of Plasmodium falciparum in a community-based cross-sectional screening program.
Citation: The American Society of Tropical Medicine and Hygiene 95, 4; 10.4269/ajtmh.16-0328
Discussion
Malaria is a common and deadly protozoa infection in sub-Saharan Africa, and accurate diagnoses can enable swift treatment.16 Here, we demonstrate that a handheld light microscope with a mobile phone attachment can be integrated into a routine use for a community-based malaria screening program to provide accurate diagnoses.
Handheld and mobile phone microscopy may be useful diagnostic tools in resource-constrained settings where laboratory infrastructure and equipment are limited. The aim of these portable devices is to bring quality laboratory diagnostics to underserviced regions and avoid moving people long distances to the nearest laboratory. Several iterations of handheld microscopy have been validated in the past but unfortunately there has been little resulting momentum or scale-up.17,18 Newer models of handheld and mobile-phone-based microscopes have recently been designed and evaluated for several infections with global health significance, including malaria, schistosomiasis, soil-transmitted helminthiasis, giardiasis, and Loa loa.8–10,12,13,19,20 To date, most of these studies have been conducted in controlled laboratory settings, and the few field studies conducted used expert microscopists to operate novel diagnostic tools. Digitization of images (in this case, with a mobile phone camera) may be useful for several reasons, including integrating machine learning software for automated identification and quantification of pathogens, or for automated diseases mapping with geographic information systems. A major barrier to scale-up and implement these portable and handheld diagnostic devices is their validation by front-line health-care providers, who will ultimately use these tools in real-world settings.
We demonstrate that the Newton Nm1 handheld microscope with a mobile phone attachment has excellent operating characteristics as used by Ivorian laboratory technicians with a half-day training. The sensitivity of this device was consistently over 80% by all microscopists when compared with conventional light microscopy, and the infections that were missed were low parasitemia infections. This study was conducted in a community-based rather than hospital-based setting where a greater proportion of individuals with high parasitemia would be expected. One other study evaluated the utility of this device; Ugandan laboratory technicians tested the Newton Nm1 on 50 preselected thick film test slides that had 30 positive and 20 negative samples and a mean parasite count of 1,464/μL of blood.12 The laboratory technicians demonstrated an average sensitivity and specificity of 93.5% and 100%, respectively, and positive and negative predictive values of 100% and 90.5%, respectively. By comparison, our study had a slightly lower sensitivity and negative predictive value. Potential reasons include the added complexity of a mobile phone attachment to the microscope and a greater prevalence of lower intensity infections in the current study and that our study was conducted directly in the field, embedded in a cross-sectional survey, rather than under ideal laboratory conditions using slides collected during prior field studies. Challenges we encountered in the field during the course of this study included frequent power outages and a laboratory environment that was exposed to environmental elements such as wind and the occasional ant infestation. While power outages may be disruptive to personnel, the Newton Nm1 microscope utilizes three AAA batteries with over 200 hours of power, and we did not have to change batteries during the course of the study. We also used tripods to support the microscopes and prevent microscopist fatigue. A concern with the lower sensitivity and negative predictive values is that there are false-negative readings in the context of a potentially deadly disease. While the device in its current state may not be ready for use in settings where high sensitivity is essential (e.g., clinical settings), the device may be helpful in field surveys to estimate the prevalence of disease in communities, with adjustments to balance for sensitivity limitations, to enable rapid prevalence mapping.
Our study is limited in that we evaluated this handheld microscope in a community-based setting only. Future studies should assess the device in clinical settings, as used by front-line health-care personnel. In addition, future studies should focus on field applications of mobile-phone attachments harnessing computer learning technology for the automated identification and quantification of pathogens.19,20 Finally, our study was conducted in a region of very high malaria prevalence and the majority of our population had low-grade parasitemia. Future studies should evaluate the operating characteristics of the Newton Nm1 microscope in regions with lower malaria prevalence.
Handheld light microscopy with a mobile phone attachment can be integrated into routine practice of community-based screening programs to diagnose malaria infections. This valuable diagnostic tool may be helpful for malaria control efforts in resource-constrained environments.
- 1.↑
WHO, 2012. T3: Test. Treat. and Track. Scaling Up Diagnostic Testing, Treatment and Surveillance of Malaria. Geneva, Switzerland: World Health Organization.
- 2.↑
WHO, 2011. Universal Access to Malaria Diagnostic Testing—An Operational Manual. November 2011 (rev. February 2013). Geneva, Switzerland: World Health Organization.
- 3.↑
Bell D, Wongsrichanalai C, Barnwell JW, 2006. Ensuring quality and access for malaria diagnosis: how can it be achieved? Nat Rev Microbiol 4 (Suppl): S7–S20.
- 4.↑
Mouatcho JC, Goldring JPD, 2013. Malaria rapid diagnostic tests: challenges and prospects. J Med Microbiol 62: 1491–1505.
- 5.↑
Mundy C, Ngwira M, Kadewele G, Bates I, Squire SB, Gilks CF, 2000. Evaluation of microscope condition in Malawi. Trans R Soc Trop Med Hyg 94: 583–584.
- 6.
Bates I, Bekoe V, Asamoa-Adu A, 2004. Improving the accuracy of malaria-related laboratory tests in Ghana. Malar J 3: 38.
- 7.↑
Petti CA, Polage CR, Quinn TC, Ronald AR, Sande MA, 2006. Laboratory medicine in Africa: a barrier to effective health care. Clin Infect Dis 42: 377–382.
- 8.↑
Stothard JR, Kabatereine NB, Tukahebwa EM, Kazibwe F, Mathieson W, Webster JP, Fenwick A, 2005. Field evaluation of the Meade Readiview handheld microscope for diagnosis of intestinal schistosomiasis in Ugandan school children. Am J Trop Med Hyg 73: 949–955.
- 9.
Bogoch II, Andrews JR, Speich B, Utzinger J, Ame SM, Ali SM, Keiser J, 2013. Mobile phone microscopy for the diagnosis of soil-transmitted helminth infections: a proof-of-concept study. Am J Trop Med Hyg 88: 626–629.
- 10.↑
Bogoch II, Coulibaly JT, Andrews JR, Speich B, Keiser J, Stothard JR, N'Goran EK, Utzinger J, 2014. Evaluation of portable microscopic devices for the diagnosis of Schistosoma and soil-transmitted helminth infection. Parasitology 141: 1811–1818.
- 11.
Bogoch II, Andrews JR, Speich B, Ame SM, Ali SM, Stothard JR, Utzinger J, Keiser J, 2014. Quantitative evaluation of a handheld light microscope for field diagnosis of soil-transmitted helminth infection. Am J Trop Med Hyg 91: 1138–1141.
- 12.↑
Stothard JR, Nabatte B, Sousa-Figueiredo JC, Kabatereine NB, 2014. Towards malaria microscopy at the point-of-contact: an assessment of the diagnostic performance of the Newton Nm1 microscope in Uganda. Parasitology 141: 1819–1825.
- 13.↑
Ephraim RKD, Duah E, Cybulski JS, Prakash M, D'Ambrosio MV, Fletcher DA, Keiser J, Andrews JR, Bogoch II, 2015. Diagnosis of Schistosoma haematobium infection with a mobile phone-mounted foldscope and a reversed-lens CellScope in Ghana. Am J Trop Med Hyg 92: 1253–1256.
- 14.↑
WHO, 2009. Malaria Microscopy Quality Assurance Manual—Version 1. Geneva, Switzerland: World Health Organization.
- 15.↑
WHO, 2010. Basic Malaria Microscopy, Part I. Learner's Guide, 2nd edition. Geneva, Switzerland: World Health Organization.
- 16.↑
WHO Malaria Policy Advisory Committee and Secretariat, 2013. Malaria Policy Advisory Committee to the WHO: conclusions and recommendations of September 2013 meeting. Malar J 12: 456.
- 17.↑
Collier JA, Longmore JM, 1983. The reliability of the microscopic diagnosis of malaria in the field and in the laboratory. Ann Trop Med Parasitol 77: 113–117.
- 18.↑
Longmore M, 1986. Community laboratory work and hand held microscopes. Trans R Soc Trop Med Hyg 80: 849–850.
- 19.↑
Koydemir HC, Gorocs Z, Tseng D, Cortazar B, Feng S, Chan RYL, Burbano J, McLeod E, Ozcan A, 2015. Rapid imaging, detection and quantification of Giardia lamblia cysts using mobile-phone based fluorescent microscopy and machine learning. Lab Chip 15: 1284–1293.
- 20.↑
D'Ambrosio MV, Bakalar M, Bennuru S, Reber C, Skandarajah A, Nilsson L, Switz N, Kamgno J, Pion S, Boussinesq M, Nutman TB, Fletcher DA, 2015. Point-of-care quantification of blood-borne filarial parasites with a mobile phone microscope. Sci Transl Med 7: 286re4.