Journal of Medical Internet Research

Preliminary Screening for Hereditary Breast and Ovarian Cancer Using an AI Chatbot as a Genetic Counselor: Clinical Study

Background Hereditary breast and ovarian cancer (HBOC) is a major type of hereditary cancer. Establishing effective screening to identify high-risk individuals for HBOC remains a challenge. We developed a prototype of a chatbot system that uses artificial intelligence (AI) for preliminary HBOC screening to determine whether individuals meet the National Comprehensive Cancer Network BRCA1/2 testing criteria. Objective This study’s objective was to validate the feasibility of this chatbot in a clinical setting by using it on a patient population that visited a hospital. Methods We validated the medical accuracy of the chatbot system by performing a test on patients who consecutively visited the Kanagawa Cancer Center. The participants completed a preoperation questionnaire to understand their background, including information technology literacy. After the operation, qualitative interviews were conducted to collect data on the usability and acceptability of the system and examine points needing improvement. Results A total of 11 participants were enrolled between October and December 2020. All of the participants were women, and among them, 10 (91%) had cancer. According to the questionnaire, 6 (54%) participants had never heard of a chatbot, while 7 (64%) had never used one. All participants were able to complete the chatbot operation, and the average time required for the operation was 18.0 (SD 5.44) minutes. The determinations by the chatbot of whether the participants met the BRCA1/2 testing criteria based on their medical and family history were consistent with those by certified genetic counselors (CGCs). We compared the medical histories obtained from the participants by the CGCs with those by the chatbot. Of the 11 participants, 3 (27%) entered information different from that obtained by the CGCs. These discrepancies were caused by the participant’s omissions or communication errors with the chatbot. Regarding the family histories, the chatbot provided new information for 3 (27%) of the 11 participants and complemented information for the family members of 5 (45%) participants not interviewed by the CGCs. The chatbot could not obtain some information on the family history of 6 (54%) participants due to several reasons, such as being outside of the scope of the chatbot’s interview questions, the participant’s omissions, and communication errors with the chatbot. Interview data were classified into the following: (1) features, (2) appearance, (3) usability and preferences, (4) concerns, (5) benefits, and (6) implementation. Favorable comments on implementation feasibility and comments on improvements were also obtained. Conclusions This study demonstrated that the preliminary screening system for HBOC using an AI chatbot was feasible for real patients.

Trust and Privacy Concerns Among Cancer Survivors Who Did Not Visit a Research Website Offering Free Genetic Counseling Services for Families: Survey Study

Background Digital health tools, such as websites, now proliferate to assist individuals in managing their health. With user input, we developed the Your Family Connects (YFC) website to promote access to genetic services for survivors of ovarian cancer and their relatives. Although we estimated that half or more would access the website, only 18% of invited survivors did so. We assessed the extent to which perceived relevance of the information provided, trust, and privacy concerns influenced decisions not to access the website. Objective We designed a theory-based cross-sectional survey to explore the following questions: (1) To what extent did nonresponders endorse privacy concerns? (2) Were privacy concerns associated with recall of receiving the website invitation, time since diagnosis, age, and race? (3) Could we identify profiles of nonresponders that would guide the development of future interventions to encourage engagement in health websites for families affected by inherited cancers? Methods A sample of survivors who were eligible to access the website yet did not respond to the study invitation was identified by linking study IDs to the Georgia Cancer Registry information. The survey was brief and contained 27 items, including recall of the invitation, interest in ovarian cancer information, benefits of using health websites, trust in health websites, and trust in university-based health research. We conducted factor analyses, regression analyses, ANOVA, correlation analyses, and logistic regression to address research questions. Results Of the 650 nonresponders to whom we sent the short survey, 368 (56.3%) responded and provided sufficient data for analysis. The mean response of 2.57 on the trust scale was significantly below the scale midpoint of 3 (t360=11.78, P<.001), suggesting that survivors who did not log on were on average distrustful of health websites. Belonging to a racial or ethnic minority group was associated with being more trusting and less skeptical about health websites. Just 196 (30.1%) nonresponders recalled the invitation to visit the website. Logistic regression analysis indicated that age was the only significant predictor of recall. Testing a model with age, racial or ethnic minority status, and the 6 privacy concerns correctly classified 58.8% of nonresponders, a rate of successful classification that was not appreciably better than a logistic regression analysis that included only age as a predictor. Conclusions The nonresponders in the present study—particularly the White nonresponders—were skeptical of website platforms regardless of whether they recalled receiving a website invitation or not. Social marketing approaches geared toward building trust in web platforms by building a relationship with an information consumer and in collaboration with trusted organizations warrant further investigation. Trial Registration ClinicalTrials.gov NCT04927013; https://clinicaltrials.gov/study/NCT04927013

Artificial Intelligence Performance in Image-Based Cancer Identification: Umbrella Review of Systematic Reviews

Background Artificial intelligence (AI) has the potential to transform cancer diagnosis, ultimately leading to better patient outcomes. Objective We performed an umbrella review to summarize and critically evaluate the evidence for the AI-based imaging diagnosis of cancers. Methods PubMed, Embase, Web of Science, Cochrane, and IEEE databases were searched for relevant systematic reviews from inception to June 19, 2024. Two independent investigators abstracted data and assessed the quality of evidence, using the Joanna Briggs Institute (JBI) Critical Appraisal Checklist for Systematic Reviews and Research Syntheses. We further assessed the quality of evidence in each meta-analysis by applying the Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) criteria. Diagnostic performance data were synthesized narratively. Results In a comprehensive analysis of 158 included studies evaluating the performance of AI algorithms in noninvasive imaging diagnosis across 8 major human system cancers, the accuracy of the classifiers for central nervous system cancers varied widely (ranging from 48% to 100%). Similarities were observed in the diagnostic performance for cancers of the head and neck, respiratory system, digestive system, urinary system, female-related systems, skin, and other sites. Most meta-analyses demonstrated positive summary performance. For instance, 9 reviews meta-analyzed sensitivity and specificity for esophageal cancer, showing ranges of 90%-95% and 80%-93.8%, respectively. In the case of breast cancer detection, 8 reviews calculated the pooled sensitivity and specificity within the ranges of 75.4%-92% and 83%-90.6%, respectively. Four meta-analyses reported the ranges of sensitivity and specificity in ovarian cancer, and both were 75%-94%. Notably, in lung cancer, the pooled specificity was relatively low, primarily distributed between 65% and 80%. Furthermore, 80.4% (127/158) of the included studies were of high quality according to the JBI Critical Appraisal Checklist, with the remaining studies classified as medium quality. The GRADE assessment indicated that the overall quality of the evidence was moderate to low. Conclusions Although AI shows great potential for achieving accelerated, accurate, and more objective diagnoses of multiple cancers, there are still hurdles to overcome before its implementation in clinical settings. The present findings highlight that a concerted effort from the research community, clinicians, and policymakers is required to overcome existing hurdles and translate this potential into improved patient outcomes and health care delivery. Trial Registration PROSPERO CRD42022364278; https://www.crd.york.ac.uk/PROSPERO/view/CRD42022364278

AI-Derived Blood Biomarkers for Ovarian Cancer Diagnosis: Systematic Review and Meta-Analysis

Background Emerging evidence underscores the potential application of artificial intelligence (AI) in discovering noninvasive blood biomarkers. However, the diagnostic value of AI-derived blood biomarkers for ovarian cancer (OC) remains inconsistent. Objective We aimed to evaluate the research quality and the validity of AI-based blood biomarkers in OC diagnosis. Methods A systematic search was performed in the MEDLINE, Embase, IEEE Xplore, PubMed, Web of Science, and the Cochrane Library databases. Studies examining the diagnostic accuracy of AI in discovering OC blood biomarkers were identified. The risk of bias was assessed using the Quality Assessment of Diagnostic Accuracy Studies–AI tool. Pooled sensitivity, specificity, and area under the curve (AUC) were estimated using a bivariate model for the diagnostic meta-analysis. Results A total of 40 studies were ultimately included. Most (n=31, 78%) included studies were evaluated as low risk of bias. Overall, the pooled sensitivity, specificity, and AUC were 85% (95% CI 83%-87%), 91% (95% CI 90%-92%), and 0.95 (95% CI 0.92-0.96), respectively. For contingency tables with the highest accuracy, the pooled sensitivity, specificity, and AUC were 95% (95% CI 90%-97%), 97% (95% CI 95%-98%), and 0.99 (95% CI 0.98-1.00), respectively. Stratification by AI algorithms revealed higher sensitivity and specificity in studies using machine learning (sensitivity=85% and specificity=92%) compared to those using deep learning (sensitivity=77% and specificity=85%). In addition, studies using serum reported substantially higher sensitivity (94%) and specificity (96%) than those using plasma (sensitivity=83% and specificity=91%). Stratification by external validation demonstrated significantly higher specificity in studies with external validation (specificity=94%) compared to those without external validation (specificity=89%), while the reverse was observed for sensitivity (74% vs 90%). No publication bias was detected in this meta-analysis. Conclusions AI algorithms demonstrate satisfactory performance in the diagnosis of OC using blood biomarkers and are anticipated to become an effective diagnostic modality in the future, potentially avoiding unnecessary surgeries. Future research is warranted to incorporate external validation into AI diagnostic models, as well as to prioritize the adoption of deep learning methodologies. Trial Registration PROSPERO CRD42023481232; https://www.crd.york.ac.uk/PROSPERO/view/CRD42023481232

Examining BRCA Previvors’ Social Media Content Creation as a Form of Self and Community Care: Qualitative Interview Study

Background Genetic testing has become a common way of identifying a woman’s risk of developing hereditary breast and ovarian cancer; however, not all medical providers have the necessary information to support patients interested in genetic testing, nor do they always have the proper information for patients once they have been diagnosed. Therefore, many “previvors”—the name given to those who have tested positive for the BRCA genetic mutation—have taken to social media to inform others about the importance of genetic testing and explain to them how to understand their test results. Historically, those desiring to speak about their medical issues online have sought out structured support groups or chat rooms; however, many previvors today are instead posting on their own personal social media accounts and creating more niche communities. Objective This study aimed to examine why BRCA previvors are sharing content on their personal social media accounts and how posting online in this way serves a purpose for their larger community. Methods A total of 16 semistructured interviews were conducted with individuals who posted about their experience being diagnosed with the BRCA genetic mutation and their subsequent treatment on their personal social media accounts, specifically for followers interested in their medical journey. The interviews were recorded, transcribed, and coded by an experienced qualitative researcher and a graduate student using inductive techniques, and a reflexive thematic analysis was applied to the transcripts. Results The results suggest BRCA previvors want to control the narrative around their personalized medical experiences rather than participating in existing groups or chat rooms. Controlling their own story, rather than adding to existing narratives, gives previvors a sense of control. It also allows them to set boundaries around the types of experiences they have online when sharing their medical journey. Finally, previvors said they feel they are serving the larger BRCA community by each sharing their individual journeys, to hopefully avoid stereotyping and homogenizing the experience of patients with BRCA genetic mutations. Conclusions Research with the objective of understanding the experiences of BRCA previvors should include exploring how and why they talk about their journeys, especially due to the lack of knowledge BRCA previvors say many of their medical providers have. We suggest further research should examine how other patients with the BRCA genetic mutation, especially racial and ethnic minority patients, are navigating their own content creation, especially considering content moderation policies that social media platforms are continuing to implement that directly impact users’ ability to share about their medical experiences.

Predictive Value of Machine Learning for Platinum Chemotherapy Responses in Ovarian Cancer: Systematic Review and Meta-Analysis

Background Machine learning is a potentially effective method for predicting the response to platinum-based treatment for ovarian cancer. However, the predictive performance of various machine learning methods and variables is still a matter of controversy and debate. Objective This study aims to systematically review relevant literature on the predictive value of machine learning for platinum-based chemotherapy responses in patients with ovarian cancer. Methods Following the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines, we systematically searched the PubMed, Embase, Web of Science, and Cochrane databases for relevant studies on predictive models for platinum-based therapies for the treatment of ovarian cancer published before April 26, 2023. The Prediction Model Risk of Bias Assessment tool was used to evaluate the risk of bias in the included articles. Concordance index (C-index), sensitivity, and specificity were used to evaluate the performance of the prediction models to investigate the predictive value of machine learning for platinum chemotherapy responses in patients with ovarian cancer. Results A total of 1749 articles were examined, and 19 of them involving 39 models were eligible for this study. The most commonly used modeling methods were logistic regression (16/39, 41%), Extreme Gradient Boosting (4/39, 10%), and support vector machine (4/39, 10%). The training cohort reported C-index in 39 predictive models, with a pooled value of 0.806; the validation cohort reported C-index in 12 predictive models, with a pooled value of 0.831. Support vector machine performed well in both the training and validation cohorts, with a C-index of 0.942 and 0.879, respectively. The pooled sensitivity was 0.890, and the pooled specificity was 0.790 in the training cohort. Conclusions Machine learning can effectively predict how patients with ovarian cancer respond to platinum-based chemotherapy and may provide a reference for the development or updating of subsequent scoring systems.

Imaging-Based AI for Predicting Lymphovascular Space Invasion in Cervical Cancer: Systematic Review and Meta-Analysis

Abstract Background The role of artificial intelligence (AI) in enhancing the accuracy of lymphovascular space invasion (LVSI) detection in cervical cancer remains debated. Objective This meta-analysis aimed to evaluate the diagnostic accuracy of imaging-based AI for predicting LVSI in cervical cancer. Methods We conducted a comprehensive literature search across multiple databases, including PubMed, Embase, and Web of Science, identifying studies published up to November 9, 2024. Studies were included if they evaluated the diagnostic performance of imaging-based AI models in detecting LVSI in cervical cancer. We used a bivariate random-effects model to calculate pooled sensitivity and specificity with corresponding 95% confidence intervals. Study heterogeneity was assessed using the I2 statistic. Results Of 403 studies identified, 16 studies (2514 patients) were included. For the interval validation set, the pooled sensitivity, specificity, and area under the curve (AUC) for detecting LVSI were 0.84 (95% CI 0.79-0.87), 0.78 (95% CI 0.75-0.81), and 0.87 (95% CI 0.84-0.90). For the external validation set, the pooled sensitivity, specificity, and AUC for detecting LVSI were 0.79 (95% CI 0.70-0.86), 0.76 (95% CI 0.67-0.83), and 0.84 (95% CI 0.81-0.87). Using the likelihood ratio test for subgroup analysis, deep learning demonstrated significantly higher sensitivity compared to machine learning (P=.01). Moreover, AI models based on positron emission tomography/computed tomography exhibited superior sensitivity relative to those based on magnetic resonance imaging (P=.01). Conclusions Imaging-based AI, particularly deep learning algorithms, demonstrates promising diagnostic performance in predicting LVSI in cervical cancer. However, the limited external validation datasets and the retrospective nature of the research may introduce potential biases. These findings underscore AI’s potential as an auxiliary diagnostic tool, necessitating further large-scale prospective validation.

Electronic Health Interventions and Cervical Cancer Screening: Systematic Review and Meta-Analysis

Background Cervical cancer is a significant cause of mortality in women. Although screening has reduced cervical cancer mortality, screening rates remain suboptimal. Electronic health interventions emerge as promising strategies to effectively tackle this issue. Objective This systematic review and meta-analysis aimed to determine the effectiveness of electronic health interventions in cervical cancer screening. Methods On December 29, 2023, we performed an extensive search for randomized controlled trials evaluating electronic health interventions to promote cervical cancer screening in adults. The search covered multiple databases, including MEDLINE, the Cochrane Central Registry of Controlled Trials, Embase, PsycINFO, PubMed, Scopus, Web of Science, and the Cumulative Index to Nursing and Allied Health Literature. These studies examined the effectiveness of electronic health interventions on cervical cancer screening. Studies published between 2013 and 2022 were included. Two independent reviewers evaluated the titles, abstracts, and full-text publications, also assessing the risk of bias using the Cochrane Risk of Bias 2 tool. Subgroup analysis was conducted based on subjects, intervention type, and economic level. The Mantel-Haenszel method was used within a random-effects model to pool the relative risk of participation in cervical cancer screening. Results A screening of 713 records identified 14 articles (15 studies) with 23,102 participants, which were included in the final analysis. The intervention strategies used in these studies included short messaging services (4/14), multimode interventions (4/14), phone calls (2/14), web videos (3/14), and internet-based booking (1/14). The results indicated that electronic health interventions were more effective than control interventions for improving cervical cancer screening rates (relative risk [RR] 1.464, 95% CI 1.285-1.667; P<.001; I2=84%), cervical cancer screening (intention-to-treat) (RR 1.382, 95% CI 1.214-1.574; P<.001; I2=82%), and cervical cancer screening (per-protocol; RR 1.565, 95% CI 1.381-1.772; P<.001; I2=74%). Subgroup analysis revealed that phone calls (RR 1.82, 95% CI 1.40-2.38), multimode (RR 1.62, 95% CI 1.26-2.08), SMS (RR 1.41, 95% CI 1.14-1.73), and video- and internet-based booking (RR 1.25, 95% CI 1.03-1.51) interventions were superior to usual care. In addition, electronic health interventions did not show a statistically significant improvement in cervical cancer screening rates among women with HPV (RR 1.17, 95% CI 0.95-1.45). Electronic health interventions had a greater impact on improving cervical cancer screening rates among women in low- and middle-income areas (RR 1.51, 95% CI 1.27-1.79). There were no indications of small study effects or publication bias. Conclusions Electronic health interventions are recommended in cervical cancer screening programs due to their potential to increase participation rates. However, significant heterogeneity remained in this meta-analysis. Researchers should conduct large-scale studies focusing on the cost-effectiveness of these interventions. Trial Registration CRD42024502884; https://www.crd.york.ac.uk/PROSPERO/display_record.php?RecordID=502884

Evaluation of the Rosa Chatbot Providing Genetic Information to Patients at Risk of Hereditary Breast and Ovarian Cancer: Qualitative Interview Study

Background Genetic testing has become an integrated part of health care for patients with breast or ovarian cancer, and the increasing demand for genetic testing is accompanied by an increasing need for easy access to reliable genetic information for patients. Therefore, we developed a chatbot app (Rosa) that is able to perform humanlike digital conversations about genetic BRCA testing. Objective Before implementing this new information service in daily clinical practice, we wanted to explore 2 aspects of chatbot use: the perceived utility and trust in chatbot technology among healthy patients at risk of hereditary cancer and how interaction with a chatbot regarding sensitive information about hereditary cancer influences patients. Methods Overall, 175 healthy individuals at risk of hereditary breast and ovarian cancer were invited to test the chatbot, Rosa, before and after genetic counseling. To secure a varied sample, participants were recruited from all cancer genetic clinics in Norway, and the selection was based on age, gender, and risk of having a BRCA pathogenic variant. Among the 34.9% (61/175) of participants who consented for individual interview, a selected subgroup (16/61, 26%) shared their experience through in-depth interviews via video. The semistructured interviews covered the following topics: usability, perceived usefulness, trust in the information received via the chatbot, how Rosa influenced the user, and thoughts about future use of digital tools in health care. The transcripts were analyzed using the stepwise-deductive inductive approach. Results The overall finding was that the chatbot was very welcomed by the participants. They appreciated the 24/7 availability wherever they were and the possibility to use it to prepare for genetic counseling and to repeat and ask questions about what had been said afterward. As Rosa was created by health care professionals, they also valued the information they received as being medically correct. Rosa was referred to as being better than Google because it provided specific and reliable answers to their questions. The findings were summed up in 3 concepts: “Anytime, anywhere”; “In addition, not instead”; and “Trustworthy and true.” All participants (16/16) denied increased worry after reading about genetic testing and hereditary breast and ovarian cancer in Rosa. Conclusions Our results indicate that a genetic information chatbot has the potential to contribute to easy access to uniform information for patients at risk of hereditary breast and ovarian cancer, regardless of geographical location. The 24/7 availability of quality-assured information, tailored to the specific situation, had a reassuring effect on our participants. It was consistent across concepts that Rosa was a tool for preparation and repetition; however, none of the participants (0/16) supported that Rosa could replace genetic counseling if hereditary cancer was confirmed. This indicates that a chatbot can be a well-suited digital companion to genetic counseling.

Effects of Message Framing on Human Papillomavirus Vaccination: Systematic Review

Background With the advancement of cervical cancer elimination strategies, promoting human papillomavirus (HPV) vaccination is essential to achieving this goal. The issue of how to structure and develop message content to promote HPV vaccination is a debatable issue. Objective The efficacy of gain-loss framing in vaccination contexts is disputed. Our study aimed to elucidate the consequences of message framing on attitudes, intentions, and behavioral tendencies toward HPV vaccination, with the objective of refining message framing strategies and their elements. Methods This systematic review adhered strictly to the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analysis) guideline reporting standards to comprehensively retrieve, extract, and integrate data. We searched databases, including PubMed, Embase, Scopus, and Web of Science, for literature published from database construction to August 15, 2023. Literature screening, data extraction, and quality evaluation were performed by 2 researchers. Intervention studies published in English, conducted with populations with children eligible for HPV vaccination, and involving message framing were included. Attitudes, intentions, and behaviors served as outcome evaluation criteria. Results A total of 19 intervention studies were included. Gain-loss framing had no clear effect on vaccination attitudes nor intentions. Loss framing showed a weak advantage at improving HPV vaccination attitudes or intentions, but the evidence was not strong enough to draw definitive conclusions. The impact of gain-loss framing on HPV vaccination behaviors could not be determined due to the limited number of studies and the qualitative nature of the analysis. Conclusions Combining gain-loss framing with other message framing approaches may be an effective way to enhance the effect of message framing. More high-quality message framing content and exploring alternative moderator or mediator variables are required to support the conclusion. Trial Registration CRD42023451612; https://www.crd.york.ac.uk/PROSPERO/display_record.php?RecordID=451612

Implementing Smartphone-Based Telemedicine for Cervical Cancer Screening in Uganda: Qualitative Study of Stakeholders’ Perceptions

Background In Uganda, cervical cancer (CaCx) is the commonest cancer, accounting for 35.7% of all cancer cases in women. The rates of human papillomavirus vaccination and CaCx screening remain low. Digital health tools and interventions have the potential to improve different aspects of CaCx screening and control in Uganda. Objective This study aimed to describe stakeholders’ perceptions of the telemedicine system we developed to improve CaCx screening in Uganda. Methods We developed and implemented a smartphone-based telemedicine system for capturing and sharing cervical images and other clinical data, as well as an artificial intelligence model for automatic analysis of images. We conducted focus group discussions with health workers at the screening clinics (n=27) and women undergoing screening (n=15) to explore their perceptions of the system. The focus group discussions were supplemented with field observations and an evaluation survey of the health workers on system usability and the overall project. Results In general, both patients and health workers had positive opinions about the system. Highlighted benefits included better cervical visualization, the ability to obtain a second opinion, improved communication between nurses and patients (to explain screening findings), improved clinical data management, performance monitoring and feedback, and modernization of screening service. However, there were also some negative perceptions. For example, some health workers felt the system is time-consuming, especially when it had just been introduced, while some patients were apprehensive about cervical image capture and sharing. Finally, commonplace challenges in digital health (eg, lack of interoperability and problems with sustainability) and challenges in cancer screening in general (eg, arduous referrals, inadequate monitoring and quality control) also resurfaced. Conclusions This study demonstrates the feasibility and value of digital health tools in CaCx screening in Uganda, particularly with regard to improving patient experience and the quality of screening services. It also provides examples of potential limitations that must be addressed for successful implementation.

Performance Evaluation of Large Language Models in Cervical Cancer Management Based on a Standardized Questionnaire: Comparative Study

Background Cervical cancer remains the fourth leading cause of death among women globally, with a particularly severe burden in low-resource settings. A comprehensive approach—from screening to diagnosis and treatment—is essential for effective prevention and management. Large language models (LLMs) have emerged as potential tools to support health care, though their specific role in cervical cancer management remains underexplored. Objective This study aims to systematically evaluate the performance and interpretability of LLMs in cervical cancer management. Methods Models were selected from the AlpacaEval leaderboard version 2.0 and based on the capabilities of our computer. The questions inputted into the models cover aspects of general knowledge, screening, diagnosis, and treatment, according to guidelines. The prompt was developed using the Context, Objective, Style, Tone, Audience, and Response (CO-STAR) framework. Responses were evaluated for accuracy, guideline compliance, clarity, and practicality, graded as A, B, C, and D with corresponding scores of 3, 2, 1, and 0. The effective rate was calculated as the ratio of A and B responses to the total number of designed questions. Local Interpretable Model-Agnostic Explanations (LIME) was used to explain and enhance physicians’ trust in model outputs within the medical context. Results Nine models were included in this study, and a set of 100 standardized questions covering general information, screening, diagnosis, and treatment was designed based on international and national guidelines. Seven models (ChatGPT-4.0 Turbo, Claude 2, Gemini Pro, Mistral-7B-v0.2, Starling-LM-7B alpha, HuatuoGPT, and BioMedLM 2.7B) provided stable responses. Among all the models included, ChatGPT-4.0 Turbo ranked first with a mean score of 2.67 (95% CI 2.54-2.80; effective rate 94.00%) with a prompt and 2.52 (95% CI 2.37-2.67; effective rate 87.00%) without a prompt, outperforming the other 8 models (P<.001). Regardless of prompts, QiZhenGPT consistently ranked among the lowest-performing models, with P<.01 in comparisons against all models except BioMedLM. Interpretability analysis showed that prompts improved alignment with human annotations for proprietary models (median intersection over union 0.43), while medical-specialized models exhibited limited improvement. Conclusions Proprietary LLMs, particularly ChatGPT-4.0 Turbo and Claude 2, show promise in clinical decision-making involving logical analysis. The use of prompts can enhance the accuracy of some models in cervical cancer management to varying degrees. Medical-specialized models, such as HuatuoGPT and BioMedLM, did not perform as well as expected in this study. By contrast, proprietary models, particularly those augmented with prompts, demonstrated notable accuracy and interpretability in medical tasks, such as cervical cancer management. However, this study underscores the need for further research to explore the practical application of LLMs in medical practice.

Effect of Immersive Virtual Reality on Chemotherapy-Related Side Effects in Patients Receiving Paclitaxel-Carboplatin With or Without Bevacizumab: 2-Arm Randomized Controlled Trial

Abstract Background Symptomatic drug treatment is generally used to treat various side effects associated with paclitaxel-carboplatin (TC) or TC plus bevacizumab (TC+Bev). However, this can lead to increased adverse effects from additional drugs. Immersive virtual reality (iVR) reduces pain and anxiety. Objective This study aimed to investigate the efficacy of iVR in managing side effects associated with TC or TC+Bev therapy. Methods This 2-arm randomized controlled trial included patients with gynecologic cancer scheduled to undergo their first course of TC/TC+Bev. Patients in the intervention group received iVR for approximately 10 minutes/day for 7 consecutive days, starting on the first day of treatment. The primary endpoint was the severity of physical and psychiatric symptoms measured using the Japanese version of the revised Edmonton Symptom Rating System (ESAS-r-J). The secondary endpoint included the proportion of patients who used additional antiemetic medications, the complete response (CR) rate to nausea, and the severity of anxiety, measured using the state-trait anxiety inventory-JYZ (STAI) Y-1. Patients in the nonintervention group received supportive and symptomatic treatments. Results The analysis included 28 and 30 patients in the intervention and nonintervention groups, respectively. The change in ESAS-r-J scores between days 1 and 7 and nausea were significantly worse in the intervention group on day 4 only (P<.001); however, the nonintervention group showed significantly worse scores on days 3, 4, and 5. Depression was not significantly worse in the intervention group on any day other than on day 1; however, the nonintervention group showed significantly worse scores on day 4. The proportion of patients who used additional antiemetic medications from days 2 to 7 was significantly lower in the intervention group than in the nonintervention group (P=.02). Regarding the change in STAI Y-1 on day 1 of TC or TC+Bev therapy, the mean score was significantly lower after the iVR experience than before the experience in the intervention group (from 43.8 to 34.8; P<.001), whereas, in the nonintervention group, no significant difference was observed before and after anticancer drug administration (from 44.9 to 43.9; P=.54). Conclusions iVR may reduce the deterioration of nausea and depression more effectively in patients with gynecologic cancer undergoing TC or TC+Bev therapy than in those undergoing nonintervention, especially in delaying the onset of nausea and accelerating recovery.

Diagnosis Test Accuracy of Artificial Intelligence for Endometrial Cancer: Systematic Review and Meta-Analysis

Background Endometrial cancer is one of the most common gynecological tumors, and early screening and diagnosis are crucial for its treatment. Research on the application of artificial intelligence (AI) in the diagnosis of endometrial cancer is increasing, but there is currently no comprehensive meta-analysis to evaluate the diagnostic accuracy of AI in screening for endometrial cancer. Objective This paper presents a systematic review of AI-based endometrial cancer screening, which is needed to clarify its diagnostic accuracy and provide evidence for the application of AI technology in screening for endometrial cancer. Methods A search was conducted across PubMed, Embase, Cochrane Library, Web of Science, and Scopus databases to include studies published in English, which evaluated the performance of AI in endometrial cancer screening. A total of 2 independent reviewers screened the titles and abstracts, and the quality of the selected studies was assessed using the Quality Assessment of Diagnostic Accuracy Studies—2 (QUADAS-2) tool. The certainty of the diagnostic test evidence was evaluated using the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) system. Results A total of 13 studies were included, and the hierarchical summary receiver operating characteristic model used for the meta-analysis showed that the overall sensitivity of AI-based endometrial cancer screening was 86% (95% CI 79%-90%) and specificity was 92% (95% CI 87%-95%). Subgroup analysis revealed similar results across AI type, study region, publication year, and study type, but the overall quality of evidence was low. Conclusions AI-based endometrial cancer screening can effectively detect patients with endometrial cancer, but large-scale population studies are needed in the future to further clarify the diagnostic accuracy of AI in screening for endometrial cancer. Trial Registration PROSPERO CRD42024519835; https://www.crd.york.ac.uk/PROSPERO/view/CRD42024519835

Performance of a Full-Coverage Cervical Cancer Screening Program Using on an Artificial Intelligence– and Cloud-Based Diagnostic System: Observational Study of an Ultralarge Population

Background The World Health Organization has set a global strategy to eliminate cervical cancer, emphasizing the need for cervical cancer screening coverage to reach 70%. In response, China has developed an action plan to accelerate the elimination of cervical cancer, with Hubei province implementing China’s first provincial full-coverage screening program using an artificial intelligence (AI) and cloud-based diagnostic system. Objective This study aimed to evaluate the performance of AI technology in this full-coverage screening program. The evaluation indicators included accessibility, screening efficiency, diagnostic quality, and program cost. Methods Characteristics of 1,704,461 individuals screened from July 2022 to January 2023 were used to analyze accessibility and AI screening efficiency. A random sample of 220 individuals was used for external diagnostic quality control. The costs of different participating screening institutions were assessed. Results Cervical cancer screening services were extended to all administrative districts, especially in rural areas. Rural women had the highest participation rate at 67.54% (1,147,839/1,699,591). Approximately 1.7 million individuals were screened, achieving a cumulative coverage of 13.45% in about 6 months. Full-coverage programs could be achieved by AI technology in approximately 1 year, which was 87.5 times more efficient than the manual reading of slides. The sample compliance rate was as high as 99.1%, and compliance rates for positive, negative, and pathology biopsy reviews exceeded 96%. The cost of this program was CN ¥49 (the average exchange rate in 2022 is as follows: US $1=CN ¥6.7261) per person, with the primary screening institution and the third-party testing institute receiving CN ¥19 and ¥27, respectively. Conclusions AI-assisted diagnosis has proven to be accessible, efficient, reliable, and low cost, which could support the implementation of full-coverage screening programs, especially in areas with insufficient health resources. AI technology served as a crucial tool for rapidly and effectively increasing screening coverage, which would accelerate the achievement of the World Health Organization’s goals of eliminating cervical cancer.

Digital Health Strategies for Cervical Cancer Control in Low- and Middle-Income Countries: Systematic Review of Current Implementations and Gaps in Research

Background Nearly 90% of deaths due to cervical cancer occur in low- and middle-income countries (LMICs). In recent years, many digital health strategies have been implemented in LMICs to ameliorate patient-, provider-, and health system–level challenges in cervical cancer control. However, there are limited efforts to systematically review the effectiveness and current landscape of digital health strategies for cervical cancer control in LMICs. Objective We aim to conduct a systematic review of digital health strategies for cervical cancer control in LMICs to assess their effectiveness, describe the range of strategies used, and summarize challenges in their implementation. Methods A systematic search was conducted to identify publications describing digital health strategies for cervical cancer control in LMICs from 5 academic databases and Google Scholar. The review excluded digital strategies associated with improving vaccination coverage against human papillomavirus. Titles and abstracts were screened, and full texts were reviewed for eligibility. A structured data extraction template was used to summarize the information from the included studies. The risk of bias and data reporting guidelines for mobile health were assessed for each study. A meta-analysis of effectiveness was planned along with a narrative review of digital health strategies, implementation challenges, and opportunities for future research. Results In the 27 included studies, interventions for cervical cancer control focused on secondary prevention (ie, screening and treatment of precancerous lesions) and digital health strategies to facilitate patient education, digital cervicography, health worker training, and data quality. Most of the included studies were conducted in sub-Saharan Africa, with fewer studies in other LMIC settings in Asia or South America. A low risk of bias was found in 2 studies, and a moderate risk of bias was found in 4 studies, while the remaining 21 studies had a high risk of bias. A meta-analysis of effectiveness was not conducted because of insufficient studies with robust study designs and matched outcomes or interventions. Conclusions Current evidence on the effectiveness of digital health strategies for cervical cancer control is limited and, in most cases, is associated with a high risk of bias. Further studies are recommended to expand the investigation of digital health strategies for cervical cancer using robust study designs, explore other LMIC settings with a high burden of cervical cancer (eg, South America), and test a greater diversity of digital strategies.

Web-Based Health Information Following the Renewal of the Cervical Screening Program in Australia: Evaluation of Readability, Understandability, and Credibility

Background Three main changes were implemented in the Australian National Cervical Screening Program (NCSP) in December 2017: an increase in the recommended age to start screening, extended screening intervals, and change from the Papanicolaou (Pap) test to primary human papillomavirus screening (cervical screening test). The internet is a readily accessible source of information to explain the reasons for these changes to the public. It is important that web-based health information about changes to national screening programs is accessible and understandable for the general population. Objective This study aimed to evaluate Australian web-based resources that provide information about the changes to the cervical screening program. Methods The term cervical screening was searched in 3 search engines. The first 10 relevant results across the first 3 pages of each search engine were selected. Overall, 2 authors independently evaluated each website for readability (Flesch Reading Ease [FRE], Flesch-Kincaid Grade Level, and Simple Measure of Gobbledygook [SMOG] index), quality of information (Patient Education Materials Assessment Tool [PEMAT] for printable materials), credibility (Journal of the American Medical Association [JAMA] benchmark criteria and presence of Health on the Net Foundation code of conduct [HONcode] certification), website design, and usability with 5 simulation questions to assess the relevance of information. A descriptive analysis was conducted for the readability measures, PEMAT, and the JAMA benchmark criteria. Results Of the 49 websites identified in the search, 15 were eligible for inclusion. The consumer-focused websites were classed as fairly difficult to read (mean FRE score 51.8, SD 13.3). The highest FRE score (easiest to read) was 70.4 (Cancer Council Australia Cervical Screening Consumer Site), and the lowest FRE score (most difficult to read) was 33.0 (NCSP Clinical Guidelines). A total of 9 consumer-focused websites and 4 health care provider–focused websites met the recommended threshold (sixth to eighth grade; SMOG index) for readability. The mean PEMAT understandability scores were 87.7% (SD 6.0%) for consumer-focused websites and 64.9% (SD 13.8%) for health care provider–focused websites. The mean actionability scores were 58.1% (SD 19.1%) for consumer-focused websites and 36.7% (SD 11.0%) for health care provider–focused websites. Moreover, 9 consumer-focused and 3 health care provider–focused websites scored above 70% for understandability, and 2 consumer-focused websites had an actionability score above 70%. A total of 3 websites met all 4 of the JAMA benchmark criteria, and 2 websites displayed the HONcode. Conclusions It is important for women to have access to information that is at an appropriate reading level to better understand the implications of the changes to the cervical screening program. These findings can help health care providers direct their patients toward websites that provide information on cervical screening that is written at accessible reading levels and has high understandability.

Effectiveness of One-Way Text Messaging on Attendance to Follow-Up Cervical Cancer Screening Among Human Papillomavirus–Positive Tanzanian Women (Connected2Care): Parallel-Group Randomized Controlled Trial

Background Rapid human papillomavirus (HPV) DNA testing is an emerging cervical cancer screening strategy in resource-limited countries, yet it requires follow-up of women who test HPV positive. Objective This study aimed to determine if one-way text messages improved attendance to a 14-month follow-up cervical cancer screening among HPV-positive women. Methods This multicenter, parallel-group randomized controlled trial was conducted at 3 hospitals in Tanzania. Eligible participants were aged between 25 and 60 years, had tested positive to a rapid HPV test during a patient-initiated screening, had been informed of their HPV result, and had a private mobile phone with a valid number. Participants were randomly assigned in a 1:1 ratio to the intervention or control group through an incorporated algorithm in the text message system. The intervention group received one-way text messages, and the control group received no text messages. The primary outcome was attendance at a 14-month health provider-initiated follow-up screening. Participants were not blinded, but outcome assessors were. The analysis was based on intention to treat. Results Between August 2015 and July 2017, 4080 women were screened for cervical cancer, of which 705 were included in this trial—358 women were allocated to the intervention group, and 347 women were allocated to the control group. Moreover, 16 women were excluded before the analysis because they developed cervical cancer or died (8 from each group). In the intervention group, 24.0% (84/350) women attended their follow-up screening, and in the control group, 23.8% (80/335) women attended their follow-up screening (risk ratio 1.02, 95% CI 0.79-1.33). Conclusions Attendance to a health provider-initiated follow-up cervical cancer screening among HPV-positive women was strikingly low, and one-way text messages did not improve the attendance rate. Implementation of rapid HPV testing as a primary screening method at the clinic level entails the challenge of ensuring a proper follow-up of women. Trial Registration ClinicalTrials.gov NCT02509702; https://clinicaltrials.gov/ct2/show/NCT02509702. International Registered Report Identifier (IRRID) RR2-10.2196/10.2196/15863

Association Between Social Media Use and Cancer Screening Awareness and Behavior for People Without a Cancer Diagnosis: Matched Cohort Study

Background The use of social media in communications regarding cancer prevention is rapidly growing. However, less is known about the general population’s social media use related to cancer screening awareness and behavior for different cancers. Objective We aimed to examine the relationship between social media use and cancer screening awareness and behavior among people without a cancer diagnosis. Methods Data were collected from the Health Information National Trends Survey 5 Cycle 1 to 3 in the United States (n=12,227). Our study included 10,124 participants without a cancer diagnosis and 3 measures of screening awareness (those who had heard of hepatitis C virus [HCV], human papillomavirus [HPV], and the HPV vaccine) and 4 measures of behavior (those who had prostate-specific antigen tests, Papanicolaou tests for cervical cancer, as well as breast cancer and colon cancer tests). Propensity-score matching was conducted to adjust for the sociodemographic variables between the social media user and nonuser participants. Multivariable logistic regression was used to assess the association of social media use by gender. Jackknife replicate weights were incorporated into the analyses. Results Of the 3794 matched participants, 1861 (57.6% weighted) were male, and the mean age was 55.5 (SD 0.42) years. Compared to social media nonusers, users were more likely to have heard of HCV (adjusted odds ratio [aOR]=2.27, 95% CI, 1.29-3.98 and aOR=2.86, 95% CI, 1.51-5.40, for male and female users, respectively) and HPV (aOR=1.82, 95% CI, 1.29-2.58 and aOR=2.35, 95% CI, 1.65-3.33, for male and female users, respectively). In addition, female users were more likely to have heard of the HPV vaccine (aOR=2.06, 95% CI, 1.41-3.00). No significant associations were found between social media use and prostate-specific antigen tests in males, Papanicolaou tests and breast cancer tests in females, or colon cancer tests in both male and female users. Conclusions While social media services can potentially promote cancer screening awareness in the general population, but they did not improve screening behavior after adjusting for socioeconomic status. These findings strengthened our understanding of social media use in targeting health communications for different cancers.

An Internet-Based Education Program for Human Papillomavirus Vaccination Among Female College Students in Mainland China: Application of the Information-Motivation-Behavioral Skills Model in a Cluster Randomized Trial

Background Patients diagnosed with cervical cancer in the last 2 decades were mainly young females. Human papillomavirus (HPV) vaccination is the most radical way to prevent HPV infection and cervical cancer. However, most female college students in mainland China have not yet been vaccinated, and their relevant knowledge is limited. Theory-based education delivered via the internet is a potentially accessible and useful way to promote HPV vaccination among this population. Objective This 3-month follow-up study intended to identify the feasibility and efficacy of an information-motivation-behavioral skills (IMB) model–based online intervention for promoting awareness and willingness regarding HPV vaccination among female college students. Methods A 7-day online HPV education program for female college students in mainland China was developed using a cluster randomized trial design. Recruitment and questionnaire surveys were performed online without face-to-face contact. SPSS 23.0 was used for statistical analysis. The chi-square test and t test were used to compare differences in qualitative and continuous variables between intervention and control groups. The generalized estimating equation was used to test the effectiveness of the intervention with a consideration of the time factor. Results Among 3867 participants, 102 had been vaccinated against HPV before the study (vaccination rate of 2.6%). A total of 3484 participants were followed up after the baseline survey, with no statistical difference in the loss rate between the intervention and control groups during the intervention and follow-up periods. At different follow-up time points, HPV-related knowledge, and the motivation, behavioral skills, and willingness regarding HPV vaccination were higher in the intervention group than in the control group. HPV-related knowledge was statistically different between the 2 groups, while the motivation, behavioral skills, and willingness regarding HPV vaccination only showed statistical differences right after the intervention, reaching a peak right after the intervention and then gradually reducing over time. Furthermore, there was no statistical difference in the HPV vaccination rate between the 2 groups. Conclusions IMB model–based online education could be a promising way to increase the HPV vaccination rate and reduce the burden of HPV infection and cervical cancer among high-risk female college students in China. Trial Registration Chinese Clinical Trial Registry ChiCTR1900025476; http://www.chictr.org.cn/showprojen.aspx? proj=42672 International Registered Report Identifier (IRRID) RR2-DOI:10.1186/s12889-019-7903-x