The clinical backdrop is stark. Aboriginal mothers experience significant perinatal mental health disparities shaped by social, historical and cultural determinants, from intergenerational trauma to systemic barriers in mainstream health services. Routine risk screening is supposed to guide follow-up care, but conventional approaches often miss cultural considerations entirely. Standard screening questionnaires and risk scores tend to catalogue deficits and symptoms while overlooking the protective factors that matter in Aboriginal communities: connection to family, cultural identity, pride, and community belonging. The Baby Coming You Ready program was designed to flip that framing, using a strengths-based, culturally adapted digital assessment that lets mothers tell their own story rather than simply answer deficit-oriented checklists.

Yet even a well-designed culturally safe tool can be undermined by how its results are used. According to the research team, some clinicians continue to rely primarily on conventional risk scores, which can neglect the broader contextual circumstances influencing Aboriginal mothers, including cultural strengths and protective factors. That reliance limits the program’s full impact. The question the team set out to explore was whether machine learning could help: could an algorithm, trained on BCYR data, predict perinatal psychological distress in a way that is both accurate and transparent enough to support culturally safe clinical conversations rather than replace them?

The study analyzed deidentified pilot data from 293 Aboriginal mothers recruited across six sites in Western Australia. After excluding five records with incomplete outcome data, the analytic dataset included 288 participants. The researchers faced a familiar obstacle in clinical machine learning: the dataset was modest in size, which constrains how well complex models can learn. Their solution was generative AI data augmentation. Using a technique known as a Tabular Variational Autoencoder, or TVAE, the team generated synthetic training examples that preserved the statistical structure of the real data while expanding the effective training set. Crucially, synthetic data was used only to augment training; model performance was always assessed on held-out real test data, ensuring that the reported results reflect genuine generalization rather than artifacts of artificial samples.

For the prediction engine itself, the researchers chose an Explainable Boosting Machine, a member of a newer family of models called generalized additive models with structured interactions. Unlike deep neural networks, whose internal reasoning is notoriously opaque, an Explainable Boosting Machine is intrinsically interpretable: each feature’s contribution to a prediction is expressed as a transparent, inspectable function, and pairwise interactions between features remain visible to the user. This matters enormously in a clinical and cultural context where a black-box score that simply labels a mother as high risk could reinforce exactly the deficit-focused, paternalistic practices the BCYR program was built to counter. An interpretable model, by contrast, can show which factors drove an estimate, inviting scrutiny and dialogue.

The results were striking. The best internal performance came from the TVAE-augmented Explainable Boosting Machine at an augmentation ratio of 1.0, meaning the synthetic data roughly doubled the training set. That configuration achieved an accuracy of 0.867, an F1 score of 0.817, and an area under the receiver operating characteristic curve, or AUC, of 0.934. In practical terms, an AUC above 0.9 indicates the model ranked mothers by their estimated risk of perinatal psychological distress with high discrimination across repeated internal train-test splits. The F1 score, which balances precision and recall, suggests the model maintained a reasonable trade-off between flagging mothers who genuinely need support and avoiding false alarms that could burden families and services.

Just as important as the numbers is what the model revealed when asked to explain itself. The explanations identified self-blame, loneliness, family pride, day-to-day management, and sleep as influential contributors to model-estimated risk, alongside related contextual factors. The pattern is telling: alongside recognized distress markers such as loneliness, disturbed sleep and self-blame, the model gave weight to protective and contextual factors like family pride and the capacity to manage daily life. That mixture reflects the strengths-based philosophy of the underlying assessment and stands in sharp contrast to conventional risk scores, which typically reduce a mother’s situation to a narrow symptom count. The researchers are explicit that these explanations are intended as prompts for culturally safe discussion and clinical reflection, not as verdicts delivered by a machine.

The vision emerging from this work is a collaborative one. In the imagined clinical workflow, an interpretable model surfaces the factors shaping a mother’s estimated risk, and a clinician, ideally working within a culturally safe framework and in partnership with Aboriginal health workers, uses those factors to open a conversation: about loneliness, about sleep, about family strengths and daily pressures. The AI becomes a mirror and a discussion aid rather than an authority. This is what the authors mean by interpretable generative AI supporting culturally contextualised prediction: the technology’s transparency is not a technical nicety but the very feature that makes it compatible with culturally responsive care, where mothers’ own narratives and community context must remain central.

The team is careful about the limits of the study, and appropriately so. This was an exploratory model-development exercise using pilot data from a single program, evaluated with repeated internal train-test splits rather than on an entirely new population. The authors state that broader community-governed external validation and prospective clinical workflow evaluation are required before any clinical implementation. Community governance is a key phrase: for Aboriginal health research, legitimacy depends on Aboriginal communities themselves controlling how data is used and how tools are deployed, a principle consistent with Indigenous data sovereignty movements worldwide. External validation would test whether the model’s performance and its explanations hold in different settings, while prospective evaluation would examine how the tool behaves in real clinics, with real time pressures and real therapeutic relationships.

Even with those caveats, the study offers a template that reaches far beyond perinatal mental health. It demonstrates that generative AI can be used to stretch small, culturally specific datasets without compromising honest evaluation, and that intrinsically interpretable models can deliver competitive accuracy while remaining auditable by clinicians and communities. As health systems worldwide grapple with deploying AI in sensitive domains, the example set by Aboriginal mothers, researchers and clinicians in Western Australia suggests a different path from the black-box norm: one where the world’s oldest living cultures help define what trustworthy, culturally safe artificial intelligence should look like, and where the measure of success is not just a high AUC but a better, more respectful conversation between a mother and those caring for her.

Subject of Research: Interpretable generative AI for culturally safe perinatal mental health prediction among Aboriginal mothers

Article Title: The world’s oldest living culture leads the way with interpretable generative AI: Aboriginal women shaping the future of perinatal mental health services

Article References: Wang, G., Bennamoun, H., Kwok, W. H., Raza, R., Sohel, F., Kelly, B., Ratajczak, T., Reid, C., Marriott, R., Walker, R., Kotz, J., & Reibel, T. (2026). The world’s oldest living culture leads the way with interpretable generative AI: Aboriginal women shaping the future of perinatal mental health services. PLOS Digital Health, 5(10), e0001709. https://doi.org/10.1371/journal.pdig.0001709

Image Credits: AI Generated

DOI: 10.1371/journal.pdig.0001709

Keywords: Aboriginal health, perinatal mental health, interpretable AI, generative AI, Explainable Boosting Machine, data augmentation, TVAE, Baby Coming You Ready, cultural safety, machine learning, PLOS Digital Health, Western Australia

The core of the argument is methodological. Conventional assessments of water security rely on hydrological modelling, infrastructure inventories and supply-side statistics: how many pipes exist, how much water is available, how many connections are installed. These measures say nothing about whether the water that reaches a household is safe, reliable, acceptable or affordable, or about the trade-offs families make when it is not. Experiential scales change that. The Household Water Insecurity Experiences Scale, known as HWISE, was developed and validated for low- and middle-income settings and asks households directly about problems such as worrying about having enough water, going without washing, or drinking water they believed was unsafe. Similarly, the Food Insecurity Experiences Scale and the Household Food Insecurity Access Scale capture lived experiences of hunger and dietary compromise. When deployed together, these instruments reveal coupled deprivation that neither measure can expose alone.

The clearest illustration comes from Walgett, a remote town of roughly 2,000 people in northern New South Wales, where the Barwon and Namoi rivers meet. Aboriginal Community Controlled Organisations there had raised concerns about drinking water quality for years. In 2018, national media reported that sodium concentrations in the town supply had reached approximately 300 milligrams per litre, nearly double the Australian palatability guideline and far above levels recommended for people managing chronic conditions such as hypertension and kidney disease, which occur at high prevalence in Aboriginal populations. From a purely infrastructural standpoint, Walgett had a functioning water system. From the perspective of the people drinking the water, it was a system that many could not safely or acceptably use.

In 2022, supported by a National Health and Medical Research Council Ideas Grant, the community ran a survey using the internationally validated HWISE and food insecurity instruments. The results were stark. Forty-six per cent of participants experienced food insecurity and 44 per cent experienced water insecurity, while almost one third experienced both at the same time. Participants who were water insecure were far more likely to be food insecure, a statistical association that turns the abstract policy language of integration into a measurable household reality. Four out of five respondents relied on bottled water, typically spending between 30 and 50 Australian dollars a week on it. In a low-income community, that recurring expenditure directly competes with the household food budget, effectively converting a water quality failure into a nutrition problem.

The surveys also changed the politics. Wendy Spencer, General Manager of the Dharriwaa Elders Group, described the shift in terms that capture why experiential data matters beyond its technical content. Prior to the surveys, she said, the community knew there was a problem but no one would listen; the data gave them a voice and helped inform their solutions. That observation goes to the heart of the authors’ case. Infrastructure descriptors and aggregate supply statistics cannot register the daily decisions of a parent choosing between bottled water and fresh vegetables, but a validated experiential scale can, and in doing so it creates evidence that institutions are obliged to engage with.

Walgett is not an outlier. Modelling by the Australian National University, published in npj Clean Water, estimated that more than 600,000 people across more than 400 locations are likely consuming water that does not meet health guidelines for quality, and that 44 per cent of those locations are Aboriginal communities. Read together with the Walgett survey findings, this points to a systemic national problem rather than a collection of local anomalies. Yet the authors note that Australia’s current consultation on a National Food Security Strategy does not explicitly position water insecurity as a priority area. Without experiential indicators embedded in national monitoring, they argue, the strategy risks addressing food supply and affordability while overlooking the fundamental constraint that unsafe or unacceptable water places on what households can prepare, drink and afford.

The mechanism linking water to food is worth spelling out, because it operates through several distinct pathways. Poor water quality forces substitution: households avoid the tap and buy bottled water or switch to sugar-sweetened beverages, a pattern documented in international research on tap water avoidance and diet-related disease. Unreliable supply disrupts food preparation and hygiene, raising the risk of gastrointestinal illness that itself worsens nutritional status. And the financial burden of alternative water sources crowds out spending on healthy food. Each pathway is invisible in infrastructure statistics, and each becomes quantifiable when water insecurity is measured experientially alongside food insecurity. This is why the authors describe integration as a methodological necessity rather than an advocacy aspiration.

The same logic is now spreading through the Asia Pacific. In Timor-Leste, an Australian Centre for International Agricultural Research-funded project led by Permatil, a community-based organisation promoting permaculture and sustainable livelihoods, has begun training local practitioners and piloting the Food Insecurity Experiences Scale and the Water Insecurity Experiences Scales. Outcome data are not yet available, but the initiative demonstrates a capacity-building model in which local organisations lead survey implementation, ethical data collection and community-centred evidence generation, embedding experiential measurement within broader water management and food system work. For a young nation where households face interacting pressures from climate variability, the approach offers a way to detect vulnerability that conventional indicators alone cannot capture.

For the water research and engineering community, the implications are practical rather than rhetorical. The authors point to case studies from Nepal and Sierra Leone showing that experiential water insecurity measures are useful for monitoring and evaluating WASH programs, revealing behavioural and economic barriers that determine whether technically sound interventions actually improve lives. Paired with biophysical indicators, experiential data provide a fuller picture of system performance: they can sharpen targeting, strengthen accountability, and test whether policies are moving the outcomes that matter to households. The metrics the authors propose are concrete and demanding: fewer families forced to spend scarce income on bottled water, fewer households facing trade-offs between water and food, and more equitable access to safe, reliable drinking water.

What makes the argument compelling is that it reframes who defines the problem. A water system can look adequate on paper while failing the people it serves, and a food strategy can project national abundance while households in remote communities ration both what they drink and what they eat. Measuring experiences closes that gap between system-level adequacy and household-level reality, and the Walgett case shows the data can shift power as well as policy, giving Aboriginal community organisations the evidence to demand and design their own solutions. As climate change accelerates and water variability deepens across Australia and the Asia Pacific, the authors’ conclusion is blunt: integrating experiential measures of water insecurity into food security measurement and policy is no longer optional. It is the difference between a strategy that counts pipes and one that counts people.

Subject of Research: Experiential measurement of household water and food insecurity to inform integrated food and water security policy in Australia and the Asia Pacific

Article Title: Measuring people’s experiences to inform integrated solutions to food and water security in Australia and the Asia Pacific region

Article References: Webster, J., Spencer, W., Corby, C., Lemos, E., & Leslie, G. (2026). Measuring people’s experiences to inform integrated solutions to food and water security in Australia and the Asia Pacific region. PLOS Water, 5(10), e0000647. https://doi.org/10.1371/journal.pwat.0000647

Image Credits: AI Generated

DOI: 10.1371/journal.pwat.0000647

Keywords: water insecurity, food security, HWISE scale, Walgett, Timor-Leste, Aboriginal health, drinking water quality, experiential measures, PLOS Water, climate variability, national food strategy, Asia Pacific

The Fitzroy Valley is home to roughly 3,500 people spread across 32 communities, about 400 kilometres from the nearest regional centre of Broome. Around 80 per cent of residents identify as Aboriginal and belong to five predominant language groups: Bunuba, Gooniyandi, Walmajarri, Wangkatjungka and Nyikina. The region has earned international recognition for its community-led response to foetal alcohol spectrum disorder, spearheaded by senior women whose authority flows from kinship and Country. Yet families there have consistently reported limited access to health and disability services, and children with developmental delays or social-emotional difficulties face elevated risks of mental illness and other comorbidities later in life without early support.

That gap prompted the Marninwarntikura Women’s Resource Centre, an Aboriginal community-controlled organisation based in Fitzroy Crossing, to form the Marurra-U partnership with Royal Far West, a New South Wales-based specialist developmental health service. The partnership operates a two-way governance model in which Aboriginal knowledge and clinical expertise meet on equal ground, delivering a wraparound model of care that includes telehealth therapy, capacity building for local teachers, and the family camp that formed the focus of the new study. The camp itself is held at the Broome Camp School, the closest facility to the Fitzroy Valley offering accommodation, meals and activities in a single location, and was run twice, in April 2024 and April 2025.

Each camp brought together occupational therapists, speech pathologists and a social worker alongside Aboriginal family support workers from the Fitzroy Valley. Children aged between one and eighteen years attended with their parents and siblings, participating in music sessions, sports, arts and crafts, low-ropes courses, swimming and raft building. Behind the playful surface lay deliberate therapeutic aims: building vocabulary, speech fluency, oral motor skills, listening and memory, social and expressive language, and emotion regulation through play-based activity. The camp also deliberately incorporated risky play, activities that require children to assess their own capabilities, test limits and manage risk, which developmental research links to confidence and personal growth.

Crucially, parents were not bystanders. While children played, parents attended workshops grounded in the Aboriginal and Torres Strait Islander social and emotional wellbeing model, working on personal artwork while learning about child brain development, grief and loss, intergenerational trauma and practical strategies for supporting children with complex needs. Workshop topics were chosen by the Aboriginal and non-Aboriginal Marurra-U team based on priorities identified through the centre’s research, advocacy and community work. Local language and knowledge were woven throughout: parents and children taught the group songs in Gooniyandi and Walmajarri, traditional instruments featured in music sessions, and parents showed the team how to make damper. One Aboriginal family support worker observed that the clinicians’ approach made families feel good because there was no judgement and no guilt.

To capture what worked, the research team used an Aboriginal Participatory Action Research approach with a strengths-based lens, treating lived experience as expert knowledge rather than deficit. Researchers lived at the camp for five days, participating in every activity and recording daily debriefs, an approach they describe as ethnographic observation paired with relational engagement, in which knowledge emerges through collaboration rather than detached observation. Their observations were validated by Aboriginal co-researchers during debriefs and on subsequent visits to Fitzroy Crossing. On the last day of each camp, an Aboriginal researcher and MWRC staff member invited parents to yarning sessions, conducted in English with Kimberley Kriol translation where needed. All seven parents at the 2024 camp and at least one parent from six of nine families in 2025 took part, with findings fed back to the community in person for validation.

The results paint a striking picture of change. Parents described their children as more confident, more independent and calmer, with one reporting that her son slept better and his behaviour was much improved, and another saying camp helped her find ways to make him calmer. Shy children made friends, sometimes for the first time, and siblings shared experiences that brought families closer together. One mother recounted that her son made her toast the morning after camp, a small domestic gesture she read as a genuine shift in their relationship. In one family of a child with epilepsy, parents perceived a reduction in night-time seizures during the activity-filled week, though the researchers caution this was not verified by a health professional and the cause remains unclear.

Parents themselves emerged as powerful agents of the camp’s success. They arrived seeking help with behaviour, speech, emotion regulation and social skills, and left with concrete strategies framed within kinship values and collective care rather than deficit models. Many said the knowledge about brain development helped them understand their children and parent more effectively, and some wanted to carry that knowledge back to schools so teachers could better support their children. Respite proved to be a strategic outcome in its own right: held away from the Fitzroy Valley, the camp freed parents from kinship obligations and daily pressures, allowing them to sit, yarn and connect in ways they said were impossible at home. One parent described arriving and feeling her body simply go loose and relaxed. Fathers, whose engagement in parenting programs is uncommon in the region, actively joined workshops and sports, an outcome both staff and parents celebrated, though the team noted the absence of male staff in 2025 and plans to involve more fathers and young men in future.

The study’s fourth theme may be its most consequential finding: two-way teams work. Parents repeatedly emphasised that the combination of an Aboriginal staff member who speaks the local language and understands the community with a non-Indigenous clinician who is professional at what she does made a big difference, because the message came from an Aboriginal woman too. Aboriginal team members connected clinical knowledge about brain development, trauma and communication to families’ histories and experiences, and explained how colonisation and intergenerational trauma continue to shape parenting and family structures in the valley. Trust built through prior contact with Marurra-U services proved decisive; one mother said she does not leave her children with anybody, but the camp team’s kindness made her feel her kids were safe. The 2025 camp even employed Aboriginal parents who had attended the previous year, which the team credited with deepening trust and boosting engagement.

The authors are careful about the limits of their evidence. Yarning sessions included only parents and carers, so children’s voices are absent, participant numbers were small, and no quantitative measures or formal assessments of language or motor skills were used, consistent with the Indigenous research methodology. Longitudinal data and culturally appropriate evaluation tools co-designed with families are needed to capture specific therapeutic outcomes and effects on education. Still, the study concludes that family-centred therapeutic camps, delivered through two-way teams with trauma-informed practice and local Aboriginal women’s leadership, show genuine promise for improving emotion regulation, social skills, confidence and behaviour in children while empowering parents. Set within a broader wraparound model that includes telehealth and service-provider training, the Marurra-U camp suggests a replicable blueprint: when clinical expertise is woven together with cultural authority, community leadership and relationship, healing follows, and the researchers ask what health systems might look like if every program were built this way.

Subject of Research: Family-centred therapeutic camps for Aboriginal children with complex neurodevelopmental needs and early life trauma in remote Australia

Article Title: Family-centred, therapeutic camps for Aboriginal children with complex neurodevelopmental needs and early life trauma in remote Australia: a qualitative study of the Marurra-U family camp

Article References: Pickard, A., Stubbs, T., Davies, J., Napier-Raman, S., Carter, E., Thomas, S., Berringal, V., Johnson, Z., Pindan, J., Hickey, C., Evans, R., Casburn, K., Esgin, T., Elliott, E. J., & Martiniuk, A. (2026). Family-centred, therapeutic camps for Aboriginal children with complex neurodevelopmental needs and early life trauma in remote Australia: a qualitative study of the Marurra-U family camp. The Lancet Regional Health – Western Pacific, 74, Article 101974. https://doi.org/10.1016/j.lanwpc.2026.101974

Image Credits: AI Generated

DOI: 10.1016/j.lanwpc.2026.101974

Keywords: Aboriginal health, family camp, neurodevelopmental disorders, early life trauma, Fitzroy Valley, fetal alcohol spectrum disorder, two-way partnership, trauma-informed care, social and emotional wellbeing, Indigenous research, parenting support, remote Australia