Internal Validation of Automated Visual Evaluation (AVE) on Smartphone Images for Cervical Cancer Screening in a Prospective Study in Zambia.

Objectives: Visual inspection with acetic acid (VIA) is a low-cost approach for cervical cancer screening used in most low- and middle-income countries (LMICs) but, similar to other visual tests like histopathology, is subjective and requires sustained training and quality assurance. We developed, trained, and validated an artificial-intelligence-based "Automated Visual Evaluation" (AVE) tool that can be adapted to run on smartphones to assess smartphone-captured images of the cervix and identify precancerous lesions, helping augment performance of VIA. Design: Prospective study. Setting: Eight public health facilities in Zambia. Participants: 8,204 women aged 25-55. Interventions: Cervical images captured on commonly used low-cost smartphone models were matched with key clinical information including human immunodeficiency virus (HIV) and human papillomavirus (HPV) status, plus histopathology analysis (where applicable), to develop and train an AVE algorithm and evaluate its performance for use as a primary screen and triage test for women who are HPV positive. Main outcome measures: Area under the receiver operating curve (AUC); sensitivity; specificity. Results: As a general population screening for cervical precancerous lesions, AVE identified cases of cervical precancerous and cancerous (CIN2+) lesions with high performance (AUC = 0.91, 95% confidence interval [CI] = 0.89 to 0.93), which translates to a sensitivity of 85% (95% CI = 81% to 90%) and specificity of 86% (95% CI = 84% to 88%) based on maximizing the Youden's index. This represents a considerable improvement over VIA, which a meta-analysis by the World Health Organization (WHO) estimates to have sensitivity of 66% and specificity of 87%. For women living with HIV, the AUC of AVE was 0.91 (95% CI = 0.88 to 0.93), and among those testing positive for high-risk HPV types, the AUC was 0.87 (95% CI = 0.83 to 0.91). Conclusions: These results demonstrate the feasibility of utilizing AVE on images captured using a commonly available smartphone by screening nurses and support our transition to clinical evaluation of AVE's sensitivity, specificity, feasibility, and acceptability across a broader range of settings. The performance of the algorithm as reported may be inflated, as biopsies were obtained only from study participants with visible aceto-white cervical lesions, which can lead to verification bias; and the images and data sets used for testing of the model, although "unseen" by the algorithm during training, were acquired from the same set of patients and devices, limiting the study to that of an internal validation of the AVE algorithm.

Validation in Zambia of a cervical screening strategy including HPV genotyping and artificial intelligence (AI)-based automated visual evaluation.

BACKGROUND: WHO has recommended HPV testing for cervical screening where it is practical and affordable. If used, it is important to both clarify and implement the clinical management of positive results. We estimated the performance in Lusaka, Zambia of a novel screening/triage approach combining HPV typing with visual assessment assisted by a deep-learning approach called automated visual evaluation (AVE). METHODS: In this well-established cervical cancer screening program nested inside public sector primary care health facilities, experienced nurses examined women with high-quality digital cameras; the magnified illuminated images permit inspection of the surface morphology of the cervix and expert telemedicine quality assurance. Emphasizing sensitive criteria to avoid missing precancer/cancer, ~ 25% of women screen positive, reflecting partly the high HIV prevalence. Visual screen-positive women are treated in the same visit by trained nurses using either ablation (~ 60%) or LLETZ excision, or referred for LLETZ or more extensive surgery as needed. We added research elements (which did not influence clinical care) including collection of HPV specimens for testing and typing with BD Onclarity™ with a five channel output (HPV16, HPV18/45, HPV31/33/52/58, HPV35/39/51/56/59/66/68, human DNA control), and collection of triplicate cervical images with a Samsung Galaxy J8 smartphone camera™ that were analyzed using AVE, an AI-based algorithm pre-trained on a large NCI cervical image archive. The four HPV groups and three AVE classes were crossed to create a 12-level risk scale, ranking participants in order of predicted risk of precancer. We evaluated the risk scale and assessed how well it predicted the observed diagnosis of precancer/cancer. RESULTS: HPV type, AVE classification, and the 12-level risk scale all were strongly associated with degree of histologic outcome. The AVE classification showed good reproducibility between replicates, and added finer predictive accuracy to each HPV type group. Women living with HIV had higher prevalence of precancer/cancer; the HPV-AVE risk categories strongly predicted diagnostic findings in these women as well. CONCLUSIONS: These results support the theoretical efficacy of HPV-AVE-based risk estimation for cervical screening. If HPV testing can be made affordable, cost-effective and point of care, this risk-based approach could be one management option for HPV-positive women.

Gynaecologic oncology surgical cancellations in Zambia.

Introduction: Cancellations of elective surgery in low-and middle-income countries (LMIC) are common and a major hindrance for patients who are in need of surgical therapeutic modalities. This is especially important in the context of scaling up needed surgical interventions for gynaecological cancer care. There is a knowledge gap in the literature related to cancellation of gynaecologic oncology surgeries in LMIC, where there is enormous need for this specific cancer surgical capacity. We report in an observational descriptive fashion, our experience at the UTH/CDH in Lusaka, Zambia, on the causes of surgical cancellations in gynaecologic oncology. Methods: From January 1, 2021 through June 31, 2023, we retrospectively evaluated the surgical registry for gynaecologic oncology at the UTH/CDH in Lusaka, Zambia to assess the number and causes of surgical cancellations. Results: There were a total of 66 (16.96%) surgical cancellations out of 389 scheduled gynaecologic oncology cases. Lack of available blood and/or low haemoglobin was the most frequent cause of surgical cancellations, 27 cases (40.90%). Conclusion: We highlight in our series that the lack of blood, leading to surgical cancellations was the most frequent impediment related to performing scheduled gynaecologic oncology surgical procedures. As gynaecologic oncology services scale up in LMIC, given the radical nature of surgery and its association with blood loss, it is incumbent on the entire clinical ecosystem to address this issue and to develop mitigating strategies, specific to their respective resource setting.

The use of thermal ablation in diverse cervical cancer "screen-and-treat" service platforms in Zambia.

OBJECTIVE: Thermal ablation (TA) was implemented in public sector cervical cancer prevention services in Zambia in 2012. Initially introduced as a treatment modality in primary healthcare clinics, it was later included in mobile outreach campaigns and clinical research trials. We report the feasibility, acceptability, safety, and provider uptake of TA in diverse clinical contexts. METHODS: Screening services based on visual inspection with acetic acid were offered by trained nurses to non-pregnant women aged 25-59 years. Women with a type 1 transformation zone (TZ) were treated with same-visit TA. Those with a type 2 or 3 TZ, or suspicious for cancer, were managed with same-visit electrosurgical excision or punch biopsy, respectively. A provider survey was conducted. RESULTS: Between 2012 and 2020, 2123 women were treated with TA: primary healthcare clinics, n = 746; mobile outreach clinics, n = 1127; research clinics, n = 250. Of the 996 women treated in primary healthcare and research clinics, 359 (48%) were HIV positive. Mild cramping during treatment was the most common adverse effect. No treatment interruptions occurred. No major complications were reported in the early (6 weeks) follow-up period. Providers expressed an overwhelming preference for TA over cryotherapy. CONCLUSION: TA was feasible, safe, and acceptable in diverse clinical contexts. It was the preferred ablation method of providers when compared with cryotherapy.

Partnering with traditional Chiefs to expand access to cervical cancer prevention services in rural Zambia.

OBJECTIVE: To evaluate how the influence of traditional Chiefs can be leveraged to promote access to cervical cancer prevention services in rural Zambia. METHODS: A retrospective review of outcome data was conducted for all screening outreach events that occurred in Zambian Chiefdoms between October 4, 2015, and October 3, 2016. Members of the health promotion team of the Cervical Cancer Prevention Program in Zambia visited local Chiefs to inform them of the importance of cervical cancer prevention. The local Chiefs then summoned adults living within their Chiefdoms to assemble for cervical cancer prevention health talks. Screen-and-treat services were implemented within each of the Chiefdoms over a 1-week period. RESULTS: VIA-enhanced digital imaging of the cervix (digital cervicography) was offered to 8399 women in ten Chiefdoms as part of a village-based screening (VBS) program. In all, 419 (4.9%) women had positive screening test results. Of these women, 276 (65.8%) were treated immediately with thermocoagulation and 143 (34.1%) were referred to provincial government hospitals to undergo either the loop electrosurgical excision procedure/large loop excision of the transformation zone (n=109, 26.0%) or punch biopsy (n=34, 8.1%). CONCLUSION: The influence of traditional Chiefs was leveraged to facilitate access to cervical cancer prevention services in rural Zambia.

Population-level scale-up of cervical cancer prevention services in a low-resource setting: development, implementation, and evaluation of the cervical cancer prevention program in Zambia.

BACKGROUND: Very few efforts have been undertaken to scale-up low-cost approaches to cervical cancer prevention in low-resource countries. METHODS: In a public sector cervical cancer prevention program in Zambia, nurses provided visual-inspection with acetic acid (VIA) and cryotherapy in clinics co-housed with HIV/AIDS programs, and referred women with complex lesions for histopathologic evaluation. Low-cost technological adaptations were deployed for improving VIA detection, facilitating expert physician opinion, and ensuring quality assurance. Key process and outcome indicators were derived by analyzing electronic medical records to evaluate program expansion efforts. FINDINGS: Between 2006-2013, screening services were expanded from 2 to 12 clinics in Lusaka, the most-populous province in Zambia, through which 102,942 women were screened. The majority (71.7%) were in the target age-range of 25-49 years; 28% were HIV-positive. Out of 101,867 with evaluable data, 20,419 (20%) were VIA positive, of whom 11,508 (56.4%) were treated with cryotherapy, and 8,911 (43.6%) were referred for histopathologic evaluation. Most women (87%, 86,301 of 98,961 evaluable) received same-day services (including 5% undergoing same-visit cryotherapy and 82% screening VIA-negative). The proportion of women with cervical intraepithelial neoplasia grade 2 and worse (CIN2+) among those referred for histopathologic evaluation was 44.1% (1,735/3,938 with histopathology results). Detection rates for CIN2+ and invasive cervical cancer were 17 and 7 per 1,000 women screened, respectively. Women with HIV were more likely to screen positive, to be referred for histopathologic evaluation, and to have cervical precancer and cancer than HIV-negative women. INTERPRETATION: We creatively disrupted the 'no screening' status quo prevailing in Zambia and addressed the heavy burden of cervical disease among previously unscreened women by establishing and scaling-up public-sector screening and treatment services at a population level. Key determinants for successful expansion included leveraging HIV/AIDS program investments, and context-specific information technology applications for quality assurance and filling human resource gaps.