Predicting menstrual bleeding patterns with levonorgestrel-releasing intrauterine systems

Frenz, Ann-Kathrin; Ahlers, Christiane; Beckert, Vita; Gerlinger, Christoph; Friede, Tim

doi:10.1080/13625187.2020.1843015

Cited by 4 publications

(5 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To facilitate integration into the MyIUS app, the original AI bleeding pattern prediction algorithm described by Frenz et al [ 34 ] was redeveloped in R 3.6.0 (The R Foundation). Educational content was generated in collaboration with medical affairs team members to ensure the scientific accuracy and validity of the material.…”

Section: Methodsmentioning

confidence: 99%

“…Through gamification and the provision of knowledge insights, the MyIUS app may help to improve the experience The MyIUS app could also provide reassurance to women who experience altered bleeding patterns following LNG-IUS placement and help them feel more in control of the situation. The accurate prediction (probability of correct classification >70%) of future bleeding patterns should also allow women to form realistic expectations about the amount and duration of bleeding they are likely to experience, which can allow them to have better discussions with their HCP regarding their contraceptive method moving forward [34].…”

Section: Principal Findingsmentioning

confidence: 99%

“…Available data from clinical trials provide a valuable source of information on bleeding patterns experienced by women using different LNG-IUSs that can be used to aid method selection in clinical practice and can also be used to provide guidance to women on the bleeding pattern they may expect when initiating an LNG-IUS [18][19][20][31][32][33][34]. However, the experience of bleeding following placement is unique to each woman and may be influenced by a variety of factors.…”

Section: Introductionmentioning

confidence: 99%

“…It was perceived that providing additional support and information regarding expected future bleeding patterns to women during the initial postplacement interval could provide reassurance and encourage method persistence. Therefore, an algorithm was developed that allowed the meaningful estimation of future bleeding patterns based on evidence collected on an individual basis through a digital daily bleeding diary [ 34 ]. The algorithm represents the first tool of its kind to provide an individualized prediction of future bleeding patterns and could be useful to both women and HCPs.…”

Section: Introductionmentioning

confidence: 99%

“…A logistic regression model is then used to estimate the probability of the woman having a regular cycle. The probability of correct classification using the model is high (>70%), and the generated bleeding categories are considered informative [ 34 ].…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Digital Medical Device Companion (MyIUS) for New Users of Intrauterine Systems: App Development Study

Karakoyun¹,

Podhaisky²,

Frenz³

et al. 2021

JMIR Med Inform

Self Cite

View full text Add to dashboard Cite

Background Women choosing a levonorgestrel-releasing intrauterine system may experience changes in their menstrual bleeding pattern during the first months following placement. Objective Although health care professionals (HCPs) can provide counseling, no method of providing individualized information on the expected bleeding pattern or continued support is currently available for women experiencing postplacement bleeding changes. We aim to develop a mobile phone–based medical app (MyIUS) to meet this need and provide a digital companion to women after the placement of the intrauterine system. Methods The MyIUS app is classified as a medical device and uses an artificial intelligence–based bleeding pattern prediction algorithm to estimate a woman’s future bleeding pattern in terms of intensity and regularity. We developed the app with the help of a multidisciplinary team by using a robust and high-quality design process in the context of a constantly evolving regulatory landscape. The development framework consisted of a phased approach including ideation, feasibility and concept finalization, product development, and product deployment or localization stages. Results The MyIUS app was considered useful by HCPs and easy to use by women who were consulted during the development process. Following the launch of the sustainable app in selected pilot countries, performance metrics will be gathered to facilitate further technical and feature updates and enhancements. A real-world performance study will also be conducted to allow us to upgrade the app in accordance with the new European Commission Medical Device legislation and to validate the bleeding pattern prediction algorithm in a real-world setting. Conclusions By providing a meaningful estimation of bleeding patterns and allowing an individualized approach to counseling and discussions about contraceptive method choice, the MyIUS app offers a useful tool that may benefit both women and HCPs. Further work is needed to validate the performance of the prediction algorithm and MyIUS app in a real-world setting.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Principal Findingsmentioning

confidence: 99%