Pocket Colposcope Research Articles

Automated image clarity detection for the improvement of colposcopy imaging with multiple devices

The proportion of women dying from cervical cancer in middle- and low-income countries is over 60%, twice that of their high-income counterparts. A primary screening strategy to eliminate this burden is cervix visualization and application of 3–5% acetic acid, inducing contrast in potential lesions. Recently, machine learning tools have emerged to aid visual diagnosis. As low-cost visualization tools expand, it is important to maximize image quality at the time of the exam or of images used in algorithms. Objective: We present the use of an object detection algorithm, the YOLOv5 model, to localize the cervix and describe blur within a multi-device image database. Methods: We took advantage of the Fourier domain to provide pseudo-labeling of training and testing images. A YOLOv5 model was trained using Pocket Colposcope, Mobile ODT EVA, and standard of care digital colposcope images. Results: When tested on all devices, this model achieved a mean average precision score, sensitivity, and specificity of 0.9, 0.89, and 0.89, respectively. Mobile ODT EVA and Pocket Colposcope hold out sets yielded mAP scores of 0.81 and 0.83, respectively, reflecting the generalizability of the algorithm. Compared to physician annotation, it yielded an accuracy of 0.72. Conclusion: This method provides an informed quantitative, generalizable analysis of captured images that is highly concordant with expert annotation. Significance: This quality control framework can assist in the standardization of colposcopy workflow, data acquisition, and image analysis and in doing so increase the availability of usable positive images for the development of deep learning algorithms.

Automated analysis of digital medical images in cervical cancer screening: A systematic review.

Cervical cancer screening programs are poorly implemented in LMICs due to a shortage of specialists and expensive diagnostic infrastructure. To address the barriers of implementation researchers have been developing low-cost portable devices and automating image analysis for decision support.However, as the knowledge base is growing rapidly, progress on the implementation status of novel imaging devices and algorithms in cervical cancer screening has become unclear. The aim of this project was to provide a systematic review summarizing the full range of automated technology systems used in cervical cancer screening. A search on academic databases was conducted and the search results were screened by two independent reviewers. Study selection was based on eligibility in meeting the terms of inclusion and exclusion criteria which were outlined using a Population, Intervention, Comparator and Outcome framework. 17 studies reported algorithms developed with source images from mobile device, viz. Pocket Colposcope, MobileODT EVA Colpo, Smartphone Camera, Smartphone-based Endoscope System, Smartscope, mHRME, and PiHRME. While 56 studies reported algorithms with source images from conventional/commercial acquisition devices. Most interventions were in the feasibility stage of development, undergoing initial clinical validations. Researchers have proven superior prediction performance of computer aided diagnostics (CAD) in colposcopy (>80% accuracies) versus manual analysis (<70.0% accuracies). Furthermore, this review summarized evidence of the algorithms which are being created utilizing portable devices, to circumvent constraints prohibiting wider implementation in LMICs (such as expensive diagnostic infrastructure). However clinical validation of novel devices with CAD is not yet implemented adequately in LMICs.

Can portable Colposcopes Replace Standard-of-care Colposcopes? A Crossover Trial of Two Portable Colposcopes with a Standard-of-Care Video Colposcope

Screen positive women need to be triaged by colposcopy which is a major challenge in low-middle income countries. Portable colposcopes may overcome many challenges, reduce referrals and enable a single visit approach. This study assessed the performance of portable colposcopes and potential to reduce referral. This crossover randomised study enrolled women aged 25 to 65 years with abnormal screening result or cervical symptoms. All women underwent visual inspection with acetic acid (VIA), HPV test, colposcopy with two portable colposcopes (Gynocular®, Gynius, Sweden, and Pocket® transvaginal colposcope, Duke University, NC, USA) and a standard video colposcope, and biopsy. Colposcopic Swede score agreement between portable and video colposcopes, as well as agreement of Swede score with histology were calculated for each device. The potential impact of portable colposcopes in a single visit approach was assessed based on the final diagnosis. Among 250 subjects, 27(10.80%) had high-grade cervical intraepithelial neoplasia (CIN2+) lesions. Swede scores for Pocket and Gynocular colposcopes were similar to video colposcope in 248 (99.20%) and 247 (98.80%) subjects, respectively (agreement scores 0.9969 and 0.9954, respectively). At a Swede score cut-off of ≥5, all three devices had identical sensitivity, specificity, positive and negative predictive value of 96.30%, 92.30%, 60.50% and 99.50,. Ablative treatment offered at field setting would result in optimal treatment in 52.0% and 85.1% cases when screened with VIA and HPV test respectively; using Pocket colposcope could improve this to 94.0% and 95.9%, respectively. Overtreatment and referral rates reduced from 46.8% and 12.4% to 4.8% and 6.0%, respectively, when VIA test is followed by triage with pocket colposcope. These outcomes were comparable to screening with HPV followed by colposcopy triage. Pocket colposcope performed comparably to the video colposcope. Used by healthcare providers in the field setting, they can augment the results of VIA significantly.

Implementing Technology That Disrupts Health Systems but not Quality of Care: The Pocket Colposcope in Lima, Peru

PURPOSE Innovative devices are often targeted at increasing access, improving quality, or reducing costs—the three axes of the infamous Iron Triangle of Health Care, which are notoriously difficult to simultaneously optimize. The main aim of this study was to demonstrate that disruptive technologies, if high quality and appropriately implemented, can result in improved access, cost, and quality of care, overcoming the conventional constraints of the Iron Triangle framework. METHODS Our team conducted a global value chain analysis of the Pocket Colposcope in Lima, Peru, and developed surveys and conducted in-depth interviews to evaluate Pocket Colposcope stakeholders. All surveys were developed with consultations from the Duke Evidence Lab and had institutional review board approval. RESULTS The global value chain identified 5 leverage points: regulatory approval, task shifting, collaboration, telemedicine, and patient acceptance. We also identified stakeholders and processes that affect the degree to which the Pocket Colposcope is successfully implemented. Of women surveyed, 39.4% answered that they had previously wanted a cervical cancer screening test, but had been unable to receive one as a result of some barrier. The most common responses were distance to clinics (31.0%), participants could not leave work (27.6%), and patients were afraid of receiving a cancer diagnosis (20.7%). All 4 midwives who participated in the focus group identified the portability of the Pocket Colposcope as the device’s most appealing feature. Providers identified the quality of the image, cost to patient, and ease of use as the 3 most important aspects of the Pocket Colposcope. CONCLUSION The Pocket Colposcope provides an opportunity to make high-quality diagnostic technology more accessible at a cheaper price for more people. Often, disruptive technology in low-income settings is expected to increase access at the cost of reducing quality. In the case of the Pocket Colposcope, the disruptive technology is significantly cheaper than existing technology, but quality is still high enough to succeed.

Combining multiple contrasts for improving machine learning-based classification of cervical cancers with a low-cost point-of-care Pocket colposcope.

We apply feature-extraction and machine learning methods to multiple sources of contrast (acetic acid, Lugol's iodine and green light) from the white Pocket Colposcope, a low-cost point of care colposcope for cervical cancer screening. We combine features from the sources of contrast and analyze diagnostic improvements with addition of each contrast. We find that overall AUC increases with additional contrast agents compared to using only one source.

Development of Algorithms for Automated Detection of Cervical Pre-Cancers With a Low-Cost, Point-of-Care, Pocket Colposcope.

In this paper, we propose methods for (1) automatic feature extraction and classification for acetic acid and Lugol's iodine cervigrams and (2) methods for combining features/diagnosis of different contrasts in cervigrams for improved performance. We developed algorithms to pre-process pathology-labeled cervigrams and extract simple but powerful color and textural-based features. The features were used to train a support vector machine model to classify cervigrams based on corresponding pathology for visual inspection with acetic acid, visual inspection with Lugol's iodine, and a combination of the two contrasts. The proposed framework achieved a sensitivity, specificity, and accuracy of 81.3%, 78.6%, and 80.0%, respectively, when used to distinguish cervical intraepithelial neoplasia (CIN+) relative to normal and benign tissues. This is superior to the average values achieved by three expert physicians on the same data set for discriminating normal/benign cases from CIN+ (77% sensitivity, 51% specificity, and 63% accuracy). The results suggest that utilizing simple color- and textural-based features from visual inspection with acetic acid and visual inspection with Lugol's iodine images may provide unbiased automation of cervigrams. This would enable automated, expert-level diagnosis of cervical pre-cancer at the point of care.

Portable Pocket colposcopy performs comparably to standard-of-care clinical colposcopy using acetic acid and Lugol's iodine as contrast mediators: an investigational study in Peru.

Our goal was to develop a tele-colposcopy platform for primary-care clinics to improve screening sensitivity and access. Specifically, we developed a low-cost, portable Pocket colposcope and evaluated its performance in a tertiary healthcare centre in Peru. Images of the cervix were captured with a standard-of-care and Pocket colposcope at la Liga Contra el Cáncer in Lima, Peru. Two hundred Peruvian women with abnormal cytology and/or human papillomavirus positivity were enrolled. Images were collected using acetic acid and Lugol's iodine as contrast agents. Biopsies were taken as per standard-of-care procedures. After passing quality review, images from 129 women were sent to four physicians who provided a diagnosis for each image. Physician interpretation of images from the two colposcopes agreed 83.1% of the time. The average sensitivity and specificity of physician interpretation compared with pathology was similar for the Pocket (sensitivity=71.2%, specificity=57.5%) and standard-of-care (sensitivity=79.8%, specificity=56.6%) colposcopes. When compared with a previous study where only acetic acid was applied to the cervix, results indicated that adding Lugol's iodine as a secondary contrast agent improved the percent agreement between colposcopes for all pathological categories by up to 8.9% and the sensitivity and specificity of physician interpretation compared with pathology by over 6.0 and 9.0%, respectively. The Pocket colposcope performance was similar to that of a standard-of-care colposcope when used to identify precancerous and cancerous lesions using acetic acid and Lugol's iodine during colposcopy examinations in Peru. The Pocket colposcope performance was similar to that of a standard-of-care colposcope when identifying cervical lesions.

An integrated strategy for improving contrast, durability, and portability of a Pocket Colposcope for cervical cancer screening and diagnosis.

IntroductionWe have previously developed a portable Pocket Colposcope for cervical cancer screening in resource-limited settings. In this manuscript we report two different strategies (cross-polarization and an integrated reflector) to improve image contrast levels achieved with the Pocket Colposcope and evaluate the merits of each strategy compared to a standard-of-care digital colposcope. The desired outcomes included reduced specular reflection (glare), increased illumination beam pattern uniformity, and reduced electrical power budget. In addition, anti-fogging and waterproofing features were incorporated to prevent the Pocket Colposcope from fogging in the vaginal canal and to enable rapid disinfection by submersion in chemical agents.MethodsCross-polarization (Generation 3 Pocket Colposcope) and a new reflector design (Generation 4 Pocket Colposcope) were used to reduce glare and improve contrast. The reflector design (including the angle and height of the reflector sidewalls) was optimized through ray-tracing simulations. Both systems were characterized with a series of bench tests to assess specular reflection, beam pattern uniformity, and image contrast. A pilot clinical study was conducted to compare the Generation 3 and 4 Pocket Colposcopes to a standard-of-care colposcope (Leisegang Optik 2). Specifically, paired images of cervices were collected from the standard-of-care colposcope and either the Generation 3 (n = 24 patients) or the Generation 4 (n = 32 patients) Pocket Colposcopes. The paired images were blinded by device, randomized, and sent to an expert physician who provided a diagnosis for each image. Corresponding pathology was obtained for all image pairs. The primary outcome measures were the level of agreement (%) and κ (kappa) statistic between the standard-of-care colposcope and each Pocket Colposcope (Generation 3 and Generation 4).ResultsBoth generations of Pocket Colposcope had significantly higher image contrast when compared to the standard-of-care colposcope. The addition of anti-fog and waterproofing features to the Generation 3 and 4 Pocket Colposcope did not impact image quality based on qualitative and quantitative metrics. The level of agreement between the Generation 3 Pocket Colposcope and the standard-of-care colposcope was 75.0% (kappa = 0.4000, p = 0.0028, n = 24). This closely matched the level of agreement between the Generation 4 Pocket Colposcope and the standard-of-care colposcope which was also 75.0% (kappa = 0.4941, p = 0.0024, n = 32).ConclusionOur results indicate that the Generation 3 and 4 Pocket Colposcopes perform comparably to the standard-of-care colposcope, with the added benefit of being low-cost and waterproof, which is ideal for use in resource-limited settings. Additionally, the reflector significantly reduces the electrical requirements of the Generation 4 Pocket Colposcope enhancing portability without altering performance compared to the Generation 3 system.

Design and preliminary analysis of a vaginal inserter for speculum-free cervical cancer screening.

ObjectiveCervical cancer screening usually requires use of a speculum to provide a clear view of the cervix. The speculum is one potential barrier to screening due to fear of pain, discomfort and embarrassment. The aim of this paper is to present and demonstrate the feasibility of a tampon-sized inserter and the POCkeT Colposcope, a miniature pen sized-colposcope, for comfortable, speculum-free and potentially self-colposcopy.Study designWe explored different designs using 3D computer-aided design (CAD) software and performed mechanical testing simulations on each. Designs were rapid prototyped and tested using a custom vaginal phantom across a range of vaginal pressures and uterine tilts to select an optimal design. Two final designs were tested with fifteen volunteers to assess cervix visualization, comfort and usability compared to the speculum and the optimal design, the curved-tip inserter, was selected for testing in volunteers.ResultsWe present a vaginal inserter as an alternative to the standard speculum for use with the POCkeT Colposcope. The device has a slim tubular body with a funnel-like curved tip measuring approximately 2.5 cm in diameter. The inserter has a channel through which a 2 megapixel (MP) mini camera with LED illumination fits to enable image capture. Mechanical finite element testing simulations with an applied pressure of 15 cm H2O indicated a high factor of safety (90.9) for the inserter. Testing of the device with a custom vaginal phantom, across a range of supine vaginal pressures and uterine tilts (retroverted, anteverted and sideverted), demonstrated image capture with a visual area comparable to the speculum for a normal/axial positioned uteri and significantly better than the speculum for anteverted and sideverted uteri (p<0.00001). Volunteer studies with self-insertion and physician-assisted cervix image capture showed adequate cervix visualization for 83% of patients. In addition, questionnaire responses from volunteers indicated a 92.3% overall preference for the inserter over the speculum and all indicated that the inserter was more comfortable than the speculum. The inserter provides a platform for self-cervical cancer screening and also enables acetic acid/Lugol’s iodine application and insertion of swabs for Pap smear sample collection.ConclusionThis study demonstrates the feasibility of an inserter and miniature-imaging device for comfortable cervical image capture of women with potential for synergistic HPV and Pap smear sample collection.

Design of a Novel Low Cost Point of Care Tampon (POCkeT) Colposcope for Use in Resource Limited Settings

IntroductionCurrent guidelines by WHO for cervical cancer screening in low- and middle-income countries involves visual inspection with acetic acid (VIA) of the cervix, followed by treatment during the same visit or a subsequent visit with cryotherapy if a suspicious lesion is found. Implementation of these guidelines is hampered by a lack of: trained health workers, reliable technology, and access to screening facilities. A low cost ultra-portable Point of Care Tampon based digital colposcope (POCkeT Colposcope) for use at the community level setting, which has the unique form factor of a tampon, can be inserted into the vagina to capture images of the cervix, which are on par with that of a state of the art colposcope, at a fraction of the cost. A repository of images to be compiled that can be used to empower front line workers to become more effective through virtual dynamic training. By task shifting to the community setting, this technology could potentially provide significantly greater cervical screening access to where the most vulnerable women live. The POCkeT Colposcope’s concentric LED ring provides comparable white and green field illumination at a fraction of the electrical power required in commercial colposcopes. Evaluation with standard optical imaging targets to assess the POCkeT Colposcope against the state of the art digital colposcope and other VIAM technologies.ResultsOur POCkeT Colposcope has comparable resolving power, color reproduction accuracy, minimal lens distortion, and illumination when compared to commercially available colposcopes. In vitro and pilot in vivo imaging results are promising with our POCkeT Colposcope capturing comparable quality images to commercial systems.ConclusionThe POCkeT Colposcope is capable of capturing images suitable for cervical lesion analysis. Our portable low cost system could potentially increase access to cervical cancer screening in limited resource settings through task shifting to community health workers.