Marcela Arrivillaga

Innovative prototypes for cervical cancer prevention in low-income primary care settings: A human-centered design approach

This article presents the design process of innovative prototypes for cervical cancer prevention in primary care centers located in low-income settings in Cali, Colombia, using the Human-Centered Design (HCD). The project was developed in collaboration with a public healthcare network comprised of 38 urban and rural centers with women between the ages of 25 and 65 years, healthcare providers of the cancer program, healthcare administrators and the general manager of said network. Our HCD process involved five stages: research, need synthesis, ideation and co-design process, prototyping and in-context usability testing. In practice, some of the stages are overlapped and iterated throughout the design process. We conducted observations, open-ended interviews and conversations, multi-stakeholder workshops, focus groups, systematic text condensation analyses and tests in real contexts. As a result, we designed four prototypes: (1) 'Encanto': An educational manicure service, (2) 'No le des la espalda a la citología': A media-based strategy, (3) An educational wireless queuing device in the waiting room, and (4) Citobot: A cervical cancer early detection device, system, and method. The tests carried out with each prototype showed their value, limitations and possibilities in terms of subsequent development and validation through public health research or clinical research. We recognize that a longer-term evaluation is required in order to determine whether the prototypes will be used regularly, integrated into cervical cancer screening services and effectively improve access to cytology as a screening test. We conclude that HCD is a useful for design-based prevention in the field of cervical cancer. The integration of this approach with public health research would allow the generation of evidence during to the formulation of policies and programs as well as optimize existing interventions and, ultimately, facilitate the scalability and financing of what actually works.

Patient acceptability of CITOBOT for cervical cancer screening: A mixed-method study

This study assessed the acceptability of CITOBOT, a device for early cervical cancer screening in a real-world pilot setting as part of a translational research project aimed at designing and clinically validating a portable, cost-effective device supported by artificial intelligence. The authors adopted the Theoretical Framework of Acceptability for its utility in evaluating patient acceptability within complex interventions’ development, piloting, and feasibility phases. We employed a mixed-method study, with 20 consecutive participants recruited from a specialized cancer healthcare center in Cali, Colombia. Data collection included a sociodemographic, gynecological-obstetric, behavioral survey, a validated patient acceptability scale, alongside open-ended interview questions. No adverse effects were reported seven days post-testing. The findings were promising, with all participants expressing high overall acceptability. Retrospective acceptability, focusing on the evaluation after device pilot testing, revealed that participants felt comfortable with the device, found it coherent with the purpose of early cervical cancer detection, and did not perceive the test as an additional burden compared to conventional cytology screening. Regarding prospective acceptability, which assesses anticipated acceptability before full implementation, three results stand out: i) All participants stated that they would intend to attend their health service if called for testing with CITOBOT; ii) they perceived opportunity costs, such as timely delivery of results, expedited diagnosis and treatment, and improved accessibility for women with limited resources or geographical barriers to healthcare access; and iii) participants viewed CITOBOT as highly effective in preventing cervical cancer deaths, indicating a strong belief in its potential to impact public health outcomes positively. Addressing concerns related to discomfort, inconvenience, and timely delivery of results, CITOBOT shows promise in enhancing cervical cancer screening participation and adherence, especially among underserved populations.

Artificial intelligence for cervical cancer screening: Scoping review, 2009–2022

AbstractBackgroundThe intersection of artificial intelligence (AI) with cancer research is increasing, and many of the advances have focused on the analysis of cancer images.ObjectivesTo describe and synthesize the literature on the diagnostic accuracy of AI in early imaging diagnosis of cervical cancer following Preferred Reporting Items for Systematic Reviews and Meta‐Analyses Extension for Scoping Reviews (PRISMA‐ScR).Search StrategyArksey and O'Malley methodology was used and PubMed, Scopus, and Google Scholar databases were searched using a combination of English and Spanish keywords.Selection CriteriaIdentified titles and abstracts were screened to select original reports and cross‐checked for overlap of cases.Data Collection and AnalysisA descriptive summary was organized by the AI algorithm used, total of images analyzed, data source, clinical comparison criteria, and diagnosis performance.Main ResultsWe identified 32 studies published between 2009 and 2022. The primary sources of images were digital colposcopy, cervicography, and mobile devices. The machine learning/deep learning (DL) algorithms applied in the articles included support vector machine (SVM), random forest classifier, k‐nearest neighbors, multilayer perceptron, C4.5, Naïve Bayes, AdaBoost, XGboots, conditional random fields, Bayes classifier, convolutional neural network (CNN; and variations), ResNet (several versions), YOLO+EfficientNetB0, and visual geometry group (VGG; several versions). SVM and DL methods (CNN, ResNet, VGG) showed the best diagnostic performances, with an accuracy of over 97%.ConclusionWe concluded that the use of AI for cervical cancer screening has increased over the years, and some results (mainly from DL) are very promising. However, further research is necessary to validate these findings.