Borderline personality disorder (BPD) and bipolar disorder (BD) are debilitating psychiatric illnesses with significant rates of misdiagnosis. This scoping review explores the potential of machine learning (ML) approaches in distinguishing individuals diagnosed with BD from those with BPD, reporting the performance metrics of various predictive models.

We searched Ovid MEDLINE, PubMed, Scopus, and Web of Science from inception to March 2025 for studies involving the terms "bipolar disorder," "borderline personality disorder," "machine learning", and "artificial intelligence." Peer-reviewed research was included without restriction on publication date or language. Of 60 studies screened, 5 met the inclusion criteria. The review followed the PCC framework, JBI Reviewer's Manual, and PRISMA guidelines.

This study identified five studies that applied predictive models to data from 591 participants to differentiate individuals with BD and BPD. Classification accuracy ranged from 61.7% to 89%. While ML models outperformed DSM-based categorical approaches overall, accuracy differed markedly by diagnosis: correctly 87.8% for BD compared with 57.7% for BPD, illustrating the persistent diagnostic challenges for BPD. Models were more accurate in distinguishing patients with both BD and BPD from those with BD alone (79.6%) than from those with BPD alone (61.7%). ML techniques based on brain imaging features achieved 80% accuracy, while mood ratings collected via smartphone enabled the differentiation of BD, BPD, and controls with 75% accuracy.

Currently, few predictive models have been developed to distinguish between BD and BPD. The findings of this review suggest that ML algorithms show moderate to good performance in clinical differentiation of BD and BPD. Further research is warranted to refine and validate predictive tools that aim to improve diagnostic precision in BD and BPD clinical practice.

Mental health concerns have been on the rise in recent years, especially in the context of the COVID-19 pandemic. Minoritized racial and ethnic groups are especially at risk for untreated mental health concerns. Mental health promotion via culturally relevant approaches in community settings serving minoritized groups can help to address mental health concerns among these populations. However, mental health promotion activities require resources reflective of organizational capacity. The current study investigates the role of organizational capacity and characteristics in barriers to mental health promotion among community organizations.

We surveyed 21 leaders of distinct community organizations (e.g., mosques, housing communities, and public libraries) in Maryland to assess their organizational capacity, characteristics, and barriers to mental health promotion. Data were analyzed using configurational analysis to identify characteristics that distinguished organizations with low, medium, and high levels of barriers to mental health promotion.

While no single variable explained the level of barriers to mental health promotion, characteristics differentiating the level of barriers experienced by organizations in the current sample included county in which the organization was located, leader interest in health promotion, number of members in the organization, organization type (e.g., mosque or housing community), presence of a health team, and budget for health promotion. Organizations reporting the fewest barriers to mental health promotion were located in an urban city and had an increase in leader interest in health promotion following the start of the COVID-19 pandemic or were medium-sized (100-399 active members) mosques. Organizations reporting the greatest number of barriers to mental health promotion were located in an urban city and had no increase in leader interest in health promotion following the start of the COVID-19 pandemic or were large organizations (400 + active members) with sufficient financial resources but no established health team.

Building community organization capacity for, and interest in, mental health promotion may be one way to encourage the availability and utilization of mental health supporting services among populations that may not otherwise receive such care.

Depression and loneliness are highly prevalent among older adults, yet access to timely and adequate mental health care remains limited in this population. Artificial intelligence-based conversational and socially assistive agents have emerged as a potentially scalable and cost-effective intervention; however, their effectiveness in alleviating depression and loneliness among older adults has not been comprehensively established. This systematic review and meta-analysis aimed to synthesize evidence from randomized controlled trials (RCTs) examining the effects of AI-based conversational and socially assistive agent interventions on depressive symptoms and loneliness in older adults.

A systematic search of five electronic databases was conducted from inception to November 15, 2025, to identify RCTs evaluating AI-based conversational and socially assistive agent interventions targeting depression and/or loneliness in older adults. Random-effects meta-analyses were performed using standardized mean differences. Statistical heterogeneity was assessed using the I² statistic and further explored through subgroup analyses. Risk of bias was evaluated using the Cochrane Risk of Bias 2 tool, and the certainty of evidence was appraised using the GRADE framework.

Eight RCTs comprising 611 participants met the inclusion criteria. Compared with control conditions, AI-based conversational and socially assistive agent interventions were associated with a statistically significant reduction in depressive symptoms (Hedges' g = - 0.25, 95% CI - 0.48 to - 0.02; I² = 10.7%). In contrast, no significant effect was observed for loneliness, and substantial heterogeneity was detected across studies (Hedges' g = - 0.67, 95% CI - 2.57 to 1.23; I² = 89%). Subgroup analyses suggested that interventions with a cognitive focus yielded more consistent effects than companionship-focused approaches, while no clear differences were observed between home-based and institutional settings.

AI-based conversational and socially assistive agent interventions appear to be effective in reducing depressive symptoms among older adults, whereas current evidence does not support a significant effect on loneliness. The effectiveness of these interventions may depend on their theoretical orientation and implementation characteristics. AI-based conversational and socially assistive agents may serve as a promising adjunct to conventional mental health care for older adults; however, further high-quality trials are needed to clarify their role in addressing loneliness and to optimize intervention design.

The protocol for this systematic review was registered in International Prospective Register of Systematic Reviews (PROSPERO identifier: CRD420261283098).

Adaptive deep brain stimulation (aDBS) dynamically modulates stimulation parameters in real time based on ongoing neural activity, potentially overcoming key limitations of conventional continuous DBS (cDBS) in Parkinson's disease (PD).

To assess the impact of switching from cDBS to aDBS on motor and non-motor symptoms in PD patients with suboptimal cDBS outcomes.

Four PD patients transitioned from optimized cDBS to aDBS using threshold algorithms tailored to beta-band activity. Motor and non-motor outcomes were assessed before surgery, under cDBS, and 9 months after aDBS using MDS-UPDRS parts I-IV, PDSS-2, and PDQ-39.

aDBS reduced motor complications (ΔUPDRS-IV: -40.4%), while maintaining motor benefit. Patients showed improved sleep (ΔPDSS-2: -52.9%), improvement in non-motor and motor aspects of daily-living (ΔUPDRS-I: -43.6%; ΔUPDRS-II:- 31.3%), and quality-of-life (ΔPDQ-39: -24.7%). No adverse effects occurred and all patients preferred aDBS over cDBS.

aDBS was feasible, well tolerated, and improved motor fluctuations and sleep quality compared with cDBS, supporting its role in personalized neuromodulation for PD.

Immune cells are vital to regeneration and repair processes in the nervous system. We previously reported that myeloid cells play a critical role in nerve regeneration and locomotor recovery after peripheral nerve injury (PNI) by facilitating the clearance of inhibitory myelin debris, promoting angiogenesis, and producing neurotrophins (NTs) such as nerve growth factor (NGF), brain-derived neurotrophic factor, NT-3 and NT-4/5. Here, we show that NTs are synthesized by various types of myeloid cells after PNI, including neutrophils, macrophages and dendritic cells. Notably, we found that within the first week of PNI in male and female mice, monocyte-derived Cd11b⁺Cd68⁺Ly6B⁺Ly6C^hi macrophages infiltrate the sciatic nerve in an interleukin (IL)-1-dependent but CCR2-independent manner, and then locally produce mature NGF. Accordingly, depletion of circulating monocytes using PLX73086, a CSF1R inhibitor unable to cross blood-nervous system barriers, reduced NGF mRNA levels in the sciatic nerve distal stump. When polarized toward a proinflammatory phenotype in vitro, mouse macrophages rapidly release the cleaved form of NGF. Further analysis revealed that systemic administration of an anti-NGF neutralizing antibody reduced mechanical pain in mice with PNI. Together with our previous work, these results suggest that infiltrating monocyte-derived macrophages release NGF, thereby promoting both peripheral nerve regeneration and pain following injury.Significance statement We unveil a pivotal role for nerve growth factor (NGF) in the modulation of pain following peripheral nerve injury (PNI). Our comprehensive investigation traces augmented NGF mRNA and protein levels during the initial week post-PNI, pinpointing the involvement of macrophages with a distinct immunophenotypic signature in NGF production. Strategic depletion experiments demonstrate monocyte-derived macrophage production of NGF in the nerve distal stump. Cultured macrophages, when polarized toward a proinflammatory phenotype, release mature NGF. Moreover, systemic NGF neutralization alleviates pain sensitivity post-PNI. This research identifies key molecular intricacies governing NGF expression and release by inflammatory macrophages, offering promising targets for therapeutics in pain management and peripheral nerve regeneration after injury.

Suicidality among U.S. veterans and servicemembers remains a critical public health crisis. Innovative, non-clinical interventions that overcome traditional barriers, including stigma and healthcare access, are urgently needed. Although non-clinical arts-based engagement offers a promising strategy, most studies to date are qualitative and few include military populations. To address this gap, we examined the impact of an online, poetry-based Community of Practice (CoP) on depression, loneliness, and suicidality, using validated psychometric scales in a military-connected population.

This pre-post pilot study enrolled 38 military servicemembers and veterans between February 2023 and February 2024. The intervention consisted of facilitated monthly virtual sessions incorporating shared reading, poetry writing, and peer-to-peer appreciative feedback. Changes in mental health scores were assessed using the Patient Health Questionnaire-9 (PHQ-9), DeJong Gierveld Loneliness Scale-6 (DJGLS-6), and the Columbia Suicide Severity Rating Scale (C-SSRS). Effect sizes were calculated using the matched-pairs rank-biserial correlation (rrb).

Participation was associated with a significant reduction in suicidal ideation severity (Z = 2.97, P = .003), representing a large effect size (rrb = 0.74, 95% CI: 0.50-0.88). Stratified analyses confirmed significant improvements for participants aged >45, those with health insurance, those living with others, and those attending ≥4 sessions (P-adj < .05). Changes in loneliness (DJGLS-6) and depression (PHQ-9) were not statistically significant (P = .095 and P = .186, respectively).

A virtual, poetry-based CoP may provide a scalable, non-clinical pathway to reduce suicidality among veterans and servicemembers. These findings suggest that structured creative engagement can improve high-risk mental health outcomes while fostering community, offering a promising upstream adjunct to traditional clinical care.

Many children and their families, especially those from priority populations, experience barriers to accessing high-quality early childhood health, education, social and legal services. Further, these families are often under-represented in service planning and research; hence innovations are not designed to meet their needs. Our aim is to codesign with families and the wider community, a Strength-based, Tiered, Accessible Resources and Supports for Kids (STARS for Kids) programme to optimise child development, parental mental well-being, and family psychosocial needs in the first 2000 days from pregnancy to start of school.

This study will employ a mixed methods design at three sites: (1) Fairfield, urban multicultural site in South-Western Sydney New South Wales; (2) Taree, a regional town with a large Indigenous community; and (3) The City of Wanneroo, a low socioeconomic area of Western Australia. The codesign process will involve five phases of the design thinking methodology informed by culturally safe, strengths-based, and trauma-informed practices. Codesign will involve families, service providers, and community leaders from priority groups such as multicultural stakeholders from South-Western Sydney and an Aboriginal Community Consultation Group with Biripi Elders and other local Indigenous representatives at Taree. Data collection will include semi-structured interviews, workshops, or focus groups as well as 'yarning' for the Aboriginal community. Qualitative data will be thematically analysed using Braun and Clarke's six-phase method of thematic approach.

Australian New Zealand Clinical Trials Registry - ACTRN12624000806561 (This protocol pertains only to the initial codesign phase, during which the STARS for Kids tiered care model will be finalised for subsequent implementation and evaluation in the next trial phase which is outlined in the trial registry).

The division of household labor within couples has been linked to mental and physical health and relationship quality, and may be a modifiable target of behavioral change. However, interventions that focus on domestic workload have received little empirical attention. Our pilot study evaluates the feasibility and acceptability of the commercially developed Fair Play Method online intervention among heterosexual, predominantly White mothers of preschool-aged children residing in the United States (n = 87). Just over half of participants (54%) completed the full intervention, indicating adequate but improvable feasibility, and participants rated the intervention as acceptable. Participants who completed a greater proportion of the intervention reported a significantly more egalitarian division of household labor at posttest. When women reported a more egalitarian balance of cognitive household labor after participating in the intervention, they reported lower posttest depression and stress and higher global mental and physical health, after adjusting for pretest levels. Women who reported a more egalitarian balance of physical household labor reported higher posttest relationship functioning and mental and physical health, after adjusting for pretest levels. In summary, this small, open-trial pilot study suggests that the division of labor can be modified in ways that benefit well-being, but our results warrant replication within a larger and more diverse sample.

Transdiagnostic psychotherapies have been gaining in popularity due in part to their hypothesized ability to address multiple emotional disorders via a single protocol. However, to date, most randomized clinical trials of these treatments have focused on patients with anxiety disorders, or in mixed samples, limiting their use in other diagnoses. The present study compared Transdiagnostic Behavior Therapy (TBT) to disorder-specific therapies (DSTs) in 304 veterans with principal diagnoses of major depressive disorder (MDD), posttraumatic stress disorder (PTSD), or panic disorder. Disorder-specific symptom outcomes and overall impairment were assessed at baseline, mid-treatment, and posttreatment. Structural equation modeling was used to evaluate changes in outcomes over time via latent growth curve models. Support for noninferiority between TBT and the matching DSTs was found in participants with PTSD and panic disorder for the disorder-specific symptom scales and for overall impairment scores across diagnoses. Similar effect sizes were obtained for MDD, although noninferiority could not be concluded because confidence intervals were wide. Similar scores for patient satisfaction and treatment completion also were observed across TBT and the comparison DSTs. Together, these findings suggest that TBT demonstrates comparable efficacy to the established DSTs in addressing symptoms of depressive disorders (MDD), trauma and stressor-related disorders (PTSD), and the anxiety disorders (panic disorder). Thus, ongoing dissemination and implementation efforts for TBT should be considered for a wider range of diagnoses, easing provider training burden and expanding treatment coverage in clinical settings.

Despite posttraumatic stress disorder's (PTSD) significant burden in China, accessible and effective treatments remain limited. This pilot randomized controlled trial examined the feasibility, acceptability, and preliminary efficacy of Written Exposure Therapy (WET) for PTSD in China and explored whether adding a mindfulness-based app (MBA) improved insomnia outcomes. Forty-six adults meeting Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) criteria for PTSD or subthreshold PTSD with associated insomnia were randomly assigned to WET (n = 15), WET + MBA (n = 15), or minimal contact control (MCC, n = 16). PTSD severity was measured using the PTSD Symptom Scale Interview (PSSI-5) and PTSD Checklist (PCL-5), and insomnia with the Insomnia Severity Index (ISI). Assessments occurred at baseline and posttreatment for all groups, with a 3-month follow-up for WET and WET + MBA. Effect sizes were measured with Hedges's g. Patient satisfaction was high for WET completers and moderate for WET + MBA completers. WET led to significantly greater reductions in PCL-5 (p = .022, g = -.66) and marginally greater reductions in PSSI-5 (p = .053, g = -.66) than MCC at posttreatment, with effects maintained at follow-up. While both conditions reduced insomnia (WET: g = -1.50; WET + MBA: g = -.67), WET alone produced significantly greater improvements than WET + MBA. WET appears to be a promising brief intervention for PTSD in Chinese samples. Exploring MBA clarifies the nuances of combining therapeutic approaches. Cultural adaptation might enhance engagement and inform optimized interventions for PTSD and associated insomnia.