Smartphone Use for Cervical Cancer Screening in Low-Resource Countries: A Pilot Study Conducted in Madagascar

Abstract

BackgroundVisual inspection of the cervix after application of 5% acetic acid (VIA) is a screening technique for cervical cancer used widely in low and middle-income countries (LMIC). To improve VIA screening performance, digital images after acid acetic application (D-VIA) are taken. The aim of this study was to evaluate the use of a smartphone for on- and off-site D-VIA diagnosis.Materials and MethodsWomen aged 30–65 years, living in the city of Ambanja, Madagascar, were recruited through a cervical cancer screening campaign. Each performed a human papillomavirus (HPV) self-sample as a primary screen. Women testing positive for HPV were referred for VIA followed by D-VIA, cervical biopsy and endocervical curettage according to routine protocol. In addition, the same day, the D-VIA was emailed to a tertiary care center for immediate assessment. Results were scored as either D-VIA normal or D-VIA abnormal, requiring immediate therapy or referral to a tertiary center. Each of the three off-site physicians were blinded to the result reported by the one on-site physician and each gave their individual assessment followed by a consensus diagnosis. Statistical analyses were conducted using STATA software.ResultsOf the 332 women recruited, 137 (41.2%) were HPV-positive and recalled for VIA triage; compliance with this invitation was 69.3% (n = 95). Cervical intraepithelial neoplasia was detected in 17.7% and 21.7% of digital images by on-site and off-site physicians, respectively. The on-site physician had a sensitivity of 66.7% (95%CI: 30.0–90.3) and a specificity of 85.7% (95%CI: 76.7–91.6); the off-site physician consensus sensitivity was 66.7% (95%CI: 30.0–90.3) with a specificity of 82.3% (95%CI: 72.4–89.1).ConclusionThis pilot study supports the use of telemedicine for off-site diagnosis of cervical intraepithelial neoplasia, with diagnostic performance similar to those achieved on-site. Further studies need to determine if smartphones can improve cervical cancer screening efficiency in LMIC.

Computer Vision Malaria Diagnostic Systems—Progress and Prospects

“…The TBDx system by applied visual sciences uses proprietary algorithms to detect and count acid fast bacilli at 40× magnification. Digital cytology systems have been extensively used in screening for cervical cancer ( 51 , 52 ). Combining these particular tests into a unified platform would provide a particularly attractive device in developing world settings where these diseases are highly prevalent and poorly diagnosed.…”

The Cost-Effectiveness of Visual Triage of Human Papillomavirus–Positive Women in Three Low- and Middle-Income Countries

“…Our review of the literature found that VIA sensitivity to detect CIN2þ in HPV-positive women was as low as 25% in Cameroon (37,38). The highest VIA sensitivity in HPVpositive women was 82% in a study from India (39), but this was considerably higher than the next highest sensitivity rate of 67% in Madagascar (36). We set VIA sensitivity in HPV-positive women at 70% to observe outcomes under a scenario of high performance.…”

Cervical cancer data and data systems in limited‐resource settings: Challenges and opportunities

“…The embedded mobile portal allows upload and real‐time data aggregation and monitoring and evaluation . In other applications, data exchange over mobile networks has been facilitated by cloud‐based data storage, multimedia messaging service or MMS, and electronic medical pathways . While mHealth technologies offer promising solutions for low‐resource settings, additional evidence on scalability and impact on health outcomes is needed …”

“…19,40,41 In Zambia, a system for telemedicine and quality assurance that combinesmobiletelecommunicationsanddigitalcervicographyplayed avitalroleinimprovingandscalingupvisualinspectionwithaceticacid (VIA)-based cervical cancer screening. 42,43 A similar approach using smartphonestocaptureandtransmitdigitalcervicalimageshasbeen successfullytestedinMadagascar, 44 andscaled-upinBotswana. 45,46 Takingmobiledigitalimagecaptureastepfurther,anenhancedvisual assessment (EVA) system tested in Kenya combines a smartphoneintegrated colposcope with a decision-support job aid that records dataonVIAtestresults,treatmentdecisions,andfollow-up/rescreen dates.Theembeddedmobileportalallowsuploadandreal-timedata aggregation and monitoring and evaluation.…”

“…45,46 Takingmobiledigitalimagecaptureastepfurther,anenhancedvisual assessment (EVA) system tested in Kenya combines a smartphoneintegrated colposcope with a decision-support job aid that records dataonVIAtestresults,treatmentdecisions,andfollow-up/rescreen dates.Theembeddedmobileportalallowsuploadandreal-timedata aggregation and monitoring and evaluation. 47 In other applications, data exchange over mobile networks has been facilitated by cloudbased data storage, 48 multimedia messaging service or MMS, 44 and electronic medical pathways. 11,49 While mHealth technologies offer promisingsolutionsforlow-resourcesettings,additionalevidenceon scalabilityandimpactonhealthoutcomesisneeded.…”