Actionable companion · Learning design & outcomes

Learning with agency: Application of Artificial Intelligence in Early Diagnosis of Spontaneous Preterm Labor and Birth

AI-assisted editorial implementation guide, separate from the original research. Proposed actions are not claims or recommendations attributed to the source authors. No independent peer review of this guide is asserted.

Read original paper Search paper agent CC BY 4.0

Evidence and context

Title and abstract screening; stage labels describe reported settings, not verified numerical ages. Design labels require full-text appraisal. Reported stage(s): Secondary school. Verify the participant population and task before transferring the findings.

Original authors’ abstract, reproduced under the linked source license:

Abstract This study reviews the current status and future prospective of knowledge on the use of artificial intelligence for the prediction of spontaneous preterm labor and birth (“preterm birth” hereafter). The summary of review suggests that different machine learning approaches would be optimal for different types of data regarding the prediction of preterm birth: the artificial neural network, logistic regression and/or the random forest for numeric data; the support vector machine for electrohysterogram data; the recurrent neural network for text data; and the convolutional neural network for image data. The ranges of performance measures were 0.79–0.94 for accuracy, 0.22–0.97 for sensitivity, 0.86–1.00 for specificity, and 0.54–0.83 for the area under the receiver operating characteristic curve. The following maternal variables were reported to be major determinants of preterm birth: delivery and pregestational body mass index, age, parity, predelivery systolic and diastolic blood pressure, twins, below high school graduation, infant sex, prior preterm birth, progesterone medication history, upper gastrointestinal tract symptom, gastroesophageal reflux disease, Helicobacter pylori, urban region, calcium channel blocker medication history, gestational diabetes mellitus, prior cone biopsy, cervical length, myomas and adenomyosis, insurance, marriage, religion, systemic lupus erythematosus, hydroxychloroquine sulfate, and increased cerebrospinal fluid and reduced cortical folding due to impaired brain growth. Keywords: preterm birth, early diagnosis, artificial intelligence

What kind of evidence is this?

Supporting AI learning science · Design requires full-text appraisal

Stage: Secondary school. Screening labels describe reported context and study-design language. They are not a risk-of-bias rating or proof of effectiveness for an age band.

Study methods · source passage

The artificial neural network and the random forest have been known for their performance that is comparable or superior to those of traditional methods such as logistic regression regarding the prediction of preterm birth [17,18,19,20,21]. For example, a recent study used data with 19,970 participants at the Duke University Medical Center from 1 January 1988 to 1 June 1997 [17,18,19]. The dependent variable (or the class) was preterm birth and the independent variables (or the attributes) were 1622 demographic, obstetric variables including age, race (Asian, black, white, Hispanic, American Indian, unknown), region (Durham, Orange, unknown), religion (Catholic, Jewish, Protestant, other), education (graduate school, college, technical school, high school, grade school, unknown), insurance (private, public, other), and marital status (married, divorced, separated, single, widowed, unknown). The artificial neural network showed a better performance than logistic regression and the decision tree: the areas under the receiver operating characteristic curves were 0.66, 0.65, and 0.68 for logistic regression, the decision tree, and the artificial neural network, respectively.

Limitations and uncertainty · source passage

Existing literature on the early diagnosis of preterm birth based on artificial intelligence suffers from several limitations and more improvement is needed in these directions. Firstly, these studies used a cross-sectional design because of limited data availability. Expanding data with a longitudinal design is expected to improve the performance of artificial intelligence much more. Secondly, these studies did not consider possible mediating effects among variables. Thirdly, these studies used data with small sample sizes in single centers. Expanding these studies to big data (such as national health insurance claims data) will be a good topic for future research. Fourthly, the areas under the receiver operating characteristic curves of machine learning methods in these studies (0.54–0.83) might not be optimal as diagnostic tests yet. Fifthly, binary categories (no, yes) are common for preterm birth but their categories can be refined further, for example, previable (less than 24 weeks of gestation), very early (24–27 weeks), early (28–31 weeks), and late (32–37 weeks). Analyzing various factors of preterm birth based on more refined categories will be an interesting issue for future research. Sixthly, little investigation has been carried out and more study is needed on possible pathways between gastroesophageal reflux disease and preterm birth [21]. Seventhly, combining different types of deep learning models for different types of preterm birth data would break new ground and bring more profound clinical insights. Finally, there has been no basic or translational research based on artificial intelligence regarding any type of preterm birth, spontaneous or medically indicated.

Design the system before the pilot

Use this study to inform a learning mechanism or implementation question. It is not direct evidence that autonomous agents improve learning. Keep the proposed agent behavior separate from what the source actually tested.

Specify a measurable learning objective, a comparator and a task without tool assistance before adopting a system.

A 90-day district pilot

Days 1–15 · Appraise. Have the curriculum lead and a teacher read the methods and limitations. Record the study design, sample, comparator, outcomes and evidence gaps. Define a local question that the study can help inform.
Days 16–30 · Design. Ask learners to attempt a problem unaided, compare an agent hint, explain a revision and verify the final reasoning against a trusted source. Make the permitted use explicit for each assignment. Name an accountable owner and set the baseline, success threshold and budget before implementation.
Days 31–60 · Test. Start with a small voluntary cohort, preserve an accessible non-AI route and require teacher review. Use synthetic or de-identified examples; approve data handling and tool terms through district processes.
Days 61–90 · Decide. Measure unaided transfer and delayed retention, calibration of confidence and the ability to identify an incorrect AI explanation. Record whether help-seeking becomes answer copying. Compare with baseline, document uncertainty and stop or revise if learning worsens, access is unequal or critical errors persist.

The board decision brief

Report the learning problem, evidence quality, participating population, total cost including review time, measured benefit, errors, access differences and remaining uncertainties. Make a dated recommendation to stop, revise or expand. A positive perception survey alone does not establish learning improvement.

Attribution and reuse

Kwang-Sig Lee; Ki Hoon Ahn. Application of Artificial Intelligence in Early Diagnosis of Spontaneous Preterm Labor and Birth. Diagnostics (2020-09-22). https://doi.org/10.3390/diagnostics10090733

Companion prepared 2026-10-30. Original authorship and DOI are preserved. The abstract is an attributed source extract; the implementation plan is new AI-assisted editorial material, offered under the same CC BY 4.0 license. This guide has not undergone independent peer review.