Dungate, B., Tucker, D. R., Goodwin, E., & Yong, P. J. (2024). Assessing the Utility of artificial intelligence in endometriosis: Promises and pitfalls. Women's health (London, England). https://doi.org/10.1177/17455057241248121
Dungate B, Tucker DR, Goodwin E, Yong PJ. Assessing the Utility of artificial intelligence in endometriosis: Promises and pitfalls. Women's health (London, England). 2024. doi:10.1177/17455057241248121
Dungate, Brie, et al. "Assessing the Utility of artificial intelligence in endometriosis: Promises and pitfalls." Women's health (London, England), 2024.
For EndNote, Zotero or Mendeley:
Open access
No open license is recorded for this paper. Reuse terms are set by the publisher.
Open Access
Free full text on PubMed Central (PMC11062212)
RRM Academy Synopsis
AI models detect signs of endometriosis on imaging in early studies
A 2024 narrative review reports that AI models detected signs of endometriosis on ultrasound and MRI in published studies. The authors list common pitfalls. They include small samples, overfitting, and incomplete reporting.
Key Findings
A neural network model for transvaginal ultrasound videos classified pouch of Douglas obliteration, verified by two sonologists, with an area under the curve (AUC) of 0.965 and 88.7% accuracy.
Pre-training an MRI model with an ultrasound model raised its AUC for the same finding from 65.0% to 90.6%.
A salivary microRNA signature predicted endometriosis with an AUC of 0.98 in a study of 200 patients using a Random Forest algorithm.
Symptom-based screening models reached AUCs of 0.88 to 0.94 without validation on outside data. In a study with similar predictors, the best models reached 0.889 on a validation cohort.
The review cites a rule of thumb of 10 samples per feature and notes scoping-review examples of machine learning with 100 features and fewer than 100 samples.
Interpretation
The paper is a narrative review. The authors state that they did not conduct a complete analysis of the field, so the studies described are a selection. The accuracy figures come from the original studies. A higher AUC means a model separates people with and without the finding better. Some symptom-based models lacked validation on outside data. The review states that models built on small or non-representative samples are expected to perform worse in real-world use. The studies covered detection and outcome prediction. The authors write that it is yet unknown how artificial intelligence (AI) in endometriosis research will change clinical care.
RRM Context
Restorative reproductive medicine starts by finding the cause of symptoms. The review reports an average delay of 8–12 years before endometriosis is diagnosed. Tools that flag likely disease sooner support that aim. Earlier answers shorten the years spent without a diagnosis.
Our editorial summary of this paper, not the article's abstract.
Abstract
Endometriosis, a chronic condition characterized by the growth of endometrial-like tissue outside of the uterus, poses substantial challenges in terms of diagnosis and treatment. Artificial intelligence (AI) has emerged as a promising tool in the field of medicine, offering opportunities to address the complexities of endometriosis. This review explores the current landscape of endometriosis diagnosis and treatment, highlighting the potential of AI to alleviate some of the associated burdens and underscoring common pitfalls and challenges when employing AI algorithms in this context. Women's health research in endometriosis has suffered from underfunding, leading to limitations in diagnosis, classification, and treatment approaches. The heterogeneity of symptoms in patients with endometriosis has further complicated efforts to address this condition. New, powerful methods of analysis have the potential to uncover previously unidentified patterns in data relating to endometriosis. AI, a collection of algorithms replicating human decision-making in data analysis, has been increasingly adopted in medical research, including endometriosis studies. While AI offers the ability to identify novel patterns in data and analyze large datasets, its effectiveness hinges on data quality and quantity and the expertise of those implementing the algorithms. Current applications of AI in endometriosis range from diagnostic tools for ultrasound imaging to predicting treatment success. These applications show promise in reducing diagnostic delays, healthcare costs, and providing patients with more treatment options, improving their quality of life. AI holds significant potential in advancing the diagnosis and treatment of endometriosis, but it must be applied carefully and transparently to avoid pitfalls and ensure reproducibility. This review calls for increased scrutiny and accountability in AI research. Addressing these challenges can lead to more effective AI-driven solutions for endometriosis and other complex medical conditions.