Contraception · 2025
Real-world validation of a bleeding prediction algorithm in levonorgestrel intrauterine device users using the MyIUS mobile app.
СтатьяНаблюдательное исследованиеВалидационное исследованиеМногоцентровое исследованиеНазвание и аннотация: Перевод в очереди
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Vieira Carolina Sales, Eggebrecht Lisa, Milhoranca Igor Andre, Neumann Robert, Acar Mariana Schiffer, von Stockum Sophia, Frenz Ann-Kathrin
Аннотация
OBJECTIVES: This study aimed to validate the real-world performance of the MyIUS mobile application algorithm in predicting bleeding intensity and regularity in levonorgestrel intrauterine device (LNG-IUD) 52 mg, 19.5 mg, and 13.5 mg users following device insertion.
STUDY DESIGN: This was an observational real-world performance study conducted in Germany, Denmark, Sweden, Spain, Mexico, and Brazil, including women aged ≥ 18 years who provided electronic written informed consent and had used the MyIUS app for 90 days following LNG-IUD 52 mg, 19.5 mg, or 13.5 mg insertion. At day 90, participants received personalized bleeding predictions for one of three bleeding intensity clusters (predominantly amenorrhea, predominantly spotting, or predominantly bleeding), with those predicted predominantly spotting or predominantly bleeding receiving regular or irregular cycle predictions. After the 270-day data collection period, bleeding data self-reported from the last 180 days were evaluated and compared with the bleeding profile prediction.
RESULTS: Based on 1734 participants with close-to-complete data sets, the overall multiclass area under the curve for the three LNG-IUDs was 0.81 (95% CI, 0.79-0.83) for bleeding intensity prediction, corresponding to sufficient discrimination in algorithm performance based on predefined thresholds. The overall area under the curve for menstrual cycle regularity prediction was 0.66 (95% CI, 0.63-0.68).
CONCLUSIONS: Results validated the bleeding intensity prediction algorithm in a real-world environment, providing evidence to support the global use of MyIUS for individualized insights into bleeding changes following LNG-IUD insertion. Further refinement of the algorithm, especially regarding its performance in predicting cycle regularity, and testing of its clinical utility are warranted.
IMPLICATIONS: This study validated the mobile application-housed bleeding prediction algorithm MyIUS in predicting bleeding intensity in LNG-IUD users following insertion in a real-world setting, supporting the global use of MyIUS for individualized insights into bleeding changes following LNG-IUD insertion.
STUDY DESIGN: This was an observational real-world performance study conducted in Germany, Denmark, Sweden, Spain, Mexico, and Brazil, including women aged ≥ 18 years who provided electronic written informed consent and had used the MyIUS app for 90 days following LNG-IUD 52 mg, 19.5 mg, or 13.5 mg insertion. At day 90, participants received personalized bleeding predictions for one of three bleeding intensity clusters (predominantly amenorrhea, predominantly spotting, or predominantly bleeding), with those predicted predominantly spotting or predominantly bleeding receiving regular or irregular cycle predictions. After the 270-day data collection period, bleeding data self-reported from the last 180 days were evaluated and compared with the bleeding profile prediction.
RESULTS: Based on 1734 participants with close-to-complete data sets, the overall multiclass area under the curve for the three LNG-IUDs was 0.81 (95% CI, 0.79-0.83) for bleeding intensity prediction, corresponding to sufficient discrimination in algorithm performance based on predefined thresholds. The overall area under the curve for menstrual cycle regularity prediction was 0.66 (95% CI, 0.63-0.68).
CONCLUSIONS: Results validated the bleeding intensity prediction algorithm in a real-world environment, providing evidence to support the global use of MyIUS for individualized insights into bleeding changes following LNG-IUD insertion. Further refinement of the algorithm, especially regarding its performance in predicting cycle regularity, and testing of its clinical utility are warranted.
IMPLICATIONS: This study validated the mobile application-housed bleeding prediction algorithm MyIUS in predicting bleeding intensity in LNG-IUD users following insertion in a real-world setting, supporting the global use of MyIUS for individualized insights into bleeding changes following LNG-IUD insertion.