Women with gestational diabetes mellitus (GDM) are at seven-fold to ten-fold increased risk of type 2 diabetes mellitus (T2DM) when compared with those who experience a normoglycaemic pregnancy, and the cumulative incidence increases with the time of follow-up post birth. This protocol outlines the development and validation of a risk prediction model assessing the 5-year and 10-year risk of T2DM in women with a prior GDM diagnosis.

Data from all birth mothers and registered births in Victoria and South Australia, retrospectively linked to national diabetes data and pathology laboratory data from 2008 to 2021, will be used for model development and validation of GDM to T2DM conversion. Candidate predictors will be selected considering existing literature, clinical significance and statistical association, including age, body mass index, parity, ethnicity, history of recurrent GDM, family history of T2DM and antenatal and postnatal glucose levels. Traditional statistical methods and machine learning algorithms will explore the best-performing and easily applicable prediction models. We will consider bootstrapping or K-fold cross-validation for internal model validation. If computationally difficult due to the expected large sample size, we will consider developing the model using 80% of available data and evaluating using a 20% random subset. We will consider external or temporal validation of the prediction model based on the availability of data. The prediction model’s performance will be assessed by using discrimination (area under the receiver operating characteristic curve, calibration (calibration slope, calibration intercept, calibration-in-the-large and observed-to-expected ratio), model overall fit (Brier score and Cox-Snell R2) and net benefit (decision curve analysis). To examine algorithm equity, the model’s predictive performance across ethnic groups and parity will be analysed. Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis-Artificial Intelligence (TRIPOD+AI) statements will be followed.

Ethics approvals have been received from Deakin University Human Research Ethics Committee (2021–179); Monash Health Human Research Ethics Committee (RES-22-0000-048A); the Australian Institute of Health and Welfare (EO2022/5/1369); the Aboriginal Health Research Ethics Committee of South Australia (SA) (04-23-1056); in addition to a Site-Specific Assessment to cover the involvement of the Preventative Health SA (formerly Wellbeing SA) (2023/SSA00065). Project findings will be disseminated in peer-reviewed journals and at scientific conferences and provided to relevant stakeholders to enable the translation of research findings into population health programmes and health policy.

Selection of antiseizure medications (ASMs) for newly diagnosed epilepsy remains largely a trial-and-error process. We have developed a machine learning (ML) model using retrospective data collected from five international cohorts that predicts response to different ASMs as the initial treatment for individual adults with new-onset epilepsy. This study aims to prospectively evaluate this model in Australia using a randomised controlled trial design.

At least 234 adult patients with newly diagnosed epilepsy will be recruited from 14 centres in Australia. Patients will be randomised 1:1 to the ML group or usual care group. The ML group will receive the ASM recommended by the model unless it is considered contraindicated by the neurologist. The usual care group will receive the ASM selected by the neurologist alone. Both the patient and neurologists conducting the follow-up will be blinded to the group assignment. Both groups will be followed up for 52 weeks to assess treatment outcomes. Additional information on adverse events, quality of life, mood and use of healthcare services and productivity will be collected using validated questionnaires. Acceptability of the model will also be assessed.

The primary outcome will be the proportion of participants who achieve seizure-freedom (defined as no seizures during the 12-month follow-up period) while taking the initially prescribed ASM. Secondary outcomes include time to treatment failure, time to first seizure after randomisation, changes in mood assessment score and quality of life score, direct healthcare costs, and loss of productivity during the treatment period.

This trial will provide class I evidence for the effectiveness of a ML model as a decision support tool for neurologists to select the first ASM for adults with newly diagnosed epilepsy.

This study is approved by the Alfred Health Human Research Ethics Committee (Project 130/23). Findings will be presented in academic conferences and submitted to peer-reviewed journals for publication.

ACTRN12623000209695.