There is limited access to effective cervical cancer screening programs in many resource‐limited settings, resulting in continued high cervical cancer burden. Human papillomavirus (HPV) testing is increasingly recognized to be the preferable primary screening approach if affordable due to superior long‐term reassurance when negative and adaptability to self‐sampling. Visual inspection with acetic acid (VIA) is an inexpensive but subjective and inaccurate method widely used in resource‐limited settings, either for primary screening or for triage of HPV‐positive individuals. A deep learning (DL)‐based automated visual evaluation (AVE) of cervical images has been developed to help improve the accuracy and reproducibility of VIA as assistive technology. However, like any new clinical technology, rigorous evaluation and proof of clinical effectiveness are required before AVE is implemented widely. In the current article, we outline essential clinical and technical considerations involved in building a validated DL‐based AVE tool for broad use as a clinical test.

Background: Accelerated global control of cervical cancer would require primary prevention with human papillomavirus (HPV) vaccination in addition to novel screening program strategies that are simple, inexpensive, and effective. We present the feasibility and outcome of a community-based HPV self-sampled screening program. Methods: In Ile Ife, Nigeria, 9406 women aged 30–49 years collected vaginal self-samples, which were tested for HPV in the local study laboratory using Hybrid Capture-2 (HC2) (Qiagen). HPV-positive women were referred to the colposcopy clinic. Gynecologist colposcopic impression dictated immediate management; biopsies were taken when definite acetowhitening was present to produce a histopathologic reference standard of precancer (and to determine final clinical management). Retrospective linkage to the medical records identified 442 of 9406 women living with HIV (WLWH). Results: With self-sampling, it was possible to screen more than 100 women per day per clinic. Following an audio-visual presentation and in-person instructions, overall acceptability of self-sampling was very high (81.2% women preferring self-sampling over clinician collection). HPV positivity was found in 17.3% of women. Intensive follow-up contributed to 85.9% attendance at the colposcopy clinic. Of those referred, 8.2% were initially treated with thermal ablation and 5.6% with large loop excision of transformation zone (LLETZ). Full visibility of the squamocolumnar junction, necessary for optimal visual triage and ablation, declined from 68.5% at age 30 to 35.4% at age 49. CIN2+ and CIN3+ (CIN- Cervical intraepithelial neoplasia), including five cancers, were identified by histology in 5.9 and 3.2% of the HPV-positive women, respectively (0.9 and 0.5% of the total screening population), leading to additional treatment as indicated. The prevalences of HPV infection and CIN2+ were substantially higher (40.5 and 2.5%, respectively) among WLWH. Colposcopic impression led to over- and under-treatment compared to the histopathology reference standard. Conclusion: A cervical cancer screening program using self-sampled HPV testing, with colposcopic immediate management of women positive for HPV, proved feasible in Nigeria. Based on the collected specimens and images, we are now evaluating the use of a combination of partial HPV typing and automated visual evaluation (AVE) of cervical images to improve the accuracy of the screening program.

Background: We propose an economical cervical screening research and implementation strategy designed to take into account the typically slow natural history of cervical cancer and the severe but hopefully temporary impact of COVID-19. The commentary introduces the practical validation of some critical components of the strategy, described in three manuscripts detailing recent project results in Asia and Africa. The main phases of a cervical screening program are 1) primary screening of women in the general population, 2) triage testing of the small minority of women that screen positive to determine need for treatment, and 3) treatment of triage-positive women thought to be at highest risk of precancer or even cancer. In each phase, attention must now be paid to safety in relation to SARS-CoV-2 transmission. The new imperatives of the COVID-19 pandemic support self-sampled HPV testing as the primary cervical screening method. Most women can be reassured for several years by a negative test performed on a self-sample collected at home, without need of clinic visit and speculum examination. The advent of relatively inexpensive, rapid and accurate HPV DNA testing makes it possible to return screening results from self-sampling very soon after specimen collection, minimizing loss to follow-up. Partial HPV typing provides important risk stratification useful for triage of HPV-positive women. A second “triage” test is often useful to guide management. In lower-resource settings, visual inspection with acetic acid (VIA) is still proposed but it is inaccurate and poorly reproducible, misclassifying the risk stratification gained by primary HPV testing. A deep-learning based approach to recognizing cervical precancer, adaptable to a smartphone camera, is being validated to improve VIA performance. The advent and approval of thermal ablation permits quick, affordable and safe, immediate treatment at the triage clinic of the majority of HPV-positive, triage-positive women. Conclusions: Overall, only a small percentage of women in cervical screening programs need to attend the hospital clinic for a surgical procedure, particularly when screening is targeted to the optimal age range for detection of precancer rather than older ages with decreased visual screening performance and higher risks of hard-to-treat outcomes including invasive cancer.

Accelerated cervical cancer control will require widespread human papillomavirus (HPV) vaccination and screening. For screening, sensitive HPV testing with an option of self‐collection is increasingly desirable. HPV typing predicts risk of precancer/cancer, which could be useful in management, but most current typing assays are expensive and/or complicated. An existing 15‐type isothermal amplification assay (AmpFire, Atila Biosystems, USA) was redesigned as a 13‐type assay (ScreenFire) for public health use. The redesigned assay groups HPV types into four channels with differential cervical cancer risk: (a) HPV16, (b) HPV18/45, (c) HPV31/33/35/52/58 and (d) HPV39/51/56/59/68. Since the assay will be most useful in resource‐limited settings, we chose a stratified random sample of 453 provider‐collected samples from a population‐based screening study in rural Nigeria that had been initially tested with MY09‐MY11‐based PCR with oligonucleotide hybridization genotyping. Frozen residual specimens were masked and retested at Atila Biosystems. Agreement on positivity between ScreenFire and prior PCR testing was very high for each of the channels. When we simulated intended use, that is, a hierarchical result in order of clinical importance of the type groups (HPV16 > 18/45 > 31/33/35/52/58 > 39/51/56/59/68), the weighted kappa for ScreenFire vs PCR was 0.90 (95% CI: 0.86‐0.93). The ScreenFire assay is mobile, relatively simple, rapid (results within 20‐60 minutes) and agrees well with reference testing particularly for the HPV types of greatest carcinogenic risk. If confirmed, ScreenFire or similar isothermal amplification assays could be useful as part of risk‐based screening and management.

Objective: Rapid, high-volume screening programs are needed as part of cervical cancer prevention in China. Methods: In a 5-day screening project in Inner Mongolia, 3345 women volunteered following a community awareness campaign, and self-swabbed to permit rapid HPV testing. Two AmpFire™ HPV detection systems (Atila Biosystems) were sufficient to provide pooled 15-HPV type data within an hour. HPV+ patients had same-day digital colposcopy (DC) performed by 1 of 6 physicians, using the EVA™ system (MobileODT). Digital images were obtained and, after biopsy of suspected lesions for later confirmatory diagnosis, women were treated immediately based on colposcopic impression. Suspected low- grade lesions were offered treatment with thermal ablation (Wisap), and suspected high-grade lesions were treated with LLETZ. Results: Of 3345 women screened, 624 (18.7%) were HPV+. Of these, 88.5% HPV+ women underwent same-day colposcopy and 78 were treated. Later consensus histology results obtained on 197 women indicated 20 CIN2+, of whom 15 were detected and treated/referred at screening (10 by thermal ablation, 4 by LLETZ, 1 by referral). Conclusions: Global control of cervical cancer will require both vaccination and screening of a huge number of women. This study illustrates a cervical screening strategy that can be used to screen-and-treat large numbers of women. HPV self-sampling facilitates high-volume screening. Specimens can be tested rapidly, promoting minimal loss-to-follow-up. Specifically, the AmpFire™ system used in this study is highly portable, simple, rapid (92 specimens per 65 min per unit), and economical. Visual triage can be performed on HPV+ women with a portable digital colposcope that provides magnification, lighting, and a recorded image. Diagnosis and appropriate treatment remain the most subjective elements. The digital image is under study for deep-learning based automated evaluation that could assist the management decision, either by itself or combined with HPV typing.

A number of challenges hinder artificial intelligence (AI) models from effective clinical translation. Foremost among these challenges is the lack of generalizability, which is defined as the ability of a model to perform well on datasets that have different characteristics from the training data. We recently investigated the development of an AI pipeline on digital images of the cervix, utilizing a multi-heterogeneous dataset of 9,462 women (17,013 images) and a multi-stage model selection and optimization approach, to generate a diagnostic classifier able to classify images of the cervix into “normal”, “indeterminate” and “precancer/cancer” (denoted as “precancer+”) categories. In this work, we investigate the performance of this multiclass classifier on external data not utilized in training and internal validation, to assess the generalizability of the classifier when moving to new settings. We assessed both the classification performance and repeatability of our classifier model across the two axes of heterogeneity present in our dataset: image capture device and geography, utilizing both out-of-the-box inference and retraining with external data. Our results demonstrate that device-level heterogeneity affects our model performance more than geography-level heterogeneity. Classification performance of our model is strong on images from a new geography without retraining, while incremental retraining with inclusion of images from a new device progressively improves classification performance on that device up to a point of saturation. Repeatability of our model is relatively unaffected by data heterogeneity and remains strong throughout. Our work supports the need for optimized retraining approaches that address data heterogeneity (e.g., when moving to a new device) to facilitate effective use of AI models in new settings.