An overview of clustering methods with guidelines for application in mental health research

Treatment journey clustering with a novel kernel k-means machine learning algorithm: a retrospective analysis of insurance claims in bipolar I disorder

In real-world psychiatric practice, patients may experience complex treatment journeys, including various diagnoses and lines of therapy. Insurance claims databases could potentially provide insight into outcomes of psychiatric treatment processes, but the diversity of event sequences restricts analyses with currently available methods. Here, we developed a novel kernel k-means clustering algorithm for event sequences that can accommodate highly diverse event types and sequence lengths. The approach, Divisive Optimized Clustering using Kernel K-means for Event Sequences (DOCKKES), also leverages a novel performance metric, the transition score, which measures sequence coherence in individual clusters. The performance of DOCKKES was evaluated in the context of bipolar I disorder, which is characterized by heterogeneous treatment journeys. We conducted a retrospective, observational analysis of a large sample (n = 31,578) of patients with bipolar I disorder from the MarketScan® Commercial Database. Using insurance claims, bipolar episode diagnoses and mental health-related lines of therapy were identified as events of interest for patient clustering. The dataset included 202,122 events; 75% of the cohort experienced unique treatment journeys. Based on an optimal run, DOCKKES identified 16 treatment journey clusters, which were evenly split for initial manic/mixed or depressive episodes (8 clusters each) and varied in sequence length and early lines of therapy. Variability across clusters was also observed for demographics, comorbidities, and mental health-related healthcare resource utilization and cost. This proof-of-concept study demonstrated the use of DOCKKES for integrating information from large datasets, enabling comparisons between patient clusters and evaluation of real-world treatment journeys in the context of evidence-based guidelines.

Understanding Dermatologists' Acceptance of Digital Health Interventions: Cross-Sectional Survey and Cluster Analysis

Background

Digital health interventions (DHIs) have the potential to enhance dermatological care by improving quality, patient empowerment, and efficiency. However, adoption remains limited, particularly in Germany.

Objective

This study explores German dermatologists' attitudes toward DHIs, clustering them by acceptance levels and analyzing differences in sociodemographics and current and future DHI use.

Methods

We conducted a cross-sectional survey, randomly inviting 1000 dermatologists in Germany to participate. The questionnaire consisted of Likert scale items rating the acceptability of DHIs from 1 to 5. Items on the current and future use of DHIs were also included. Exploratory factor analysis was used to identify factors and reduce data as input for a 2-step clustering algorithm.

Results

The survey with 170 dermatologists (mean age 50.8, SD 10.3 y; 74/167, 55.7% female) identified four factors through the exploratory factor analysis: (1) "Positive Expectancies and Acceptability of DHIs," (2) "Dermatologists' Digital Competencies," (3) "Negative Expectancies and Barriers," and (4) "Dermatologists' Perspectives on Patients' Acceptability and Competencies." The analysis identified three distinct clusters: (1) Indecisives (n=69)-moderate intentions to use DHIs and moderate negative expectations toward them; (2) Adopters (n=60)-high intentions to use DHIs and high digital competencies; and (3) Rejectors (n=26)-low intentions to use DHIs and low digital competencies. Adopters were significantly younger, more often based in urban centers, and exhibited the highest adoption rates of DHIs compared to the other clusters. Across all clusters, inadequate reimbursement and perceived structural barriers were cited as significant challenges to DHI adoption. Still, only one-third of the Adopters used DHIs including teledermatology or artificial intelligence.

Conclusions

Dermatologists in Germany exhibited varied levels of acceptance and readiness for DHIs, with demographic and structural factors influencing adoption. Addressing barriers such as reimbursement and investing in digital literacy could promote wider use, potentially reducing health inequalities by improving access to digital health care.

Clinical Staging for Psychiatry and Psychology

A global mental health crisis is threatening a generation of young people with a lifetime of symptoms that do not fit neatly into diagnostic systems. Optimal decisions regarding treatments, services, research, and policies are critically needed, yet such decisions are based on idiosyncratic categorization of clinical courses. This review suggests clinical staging approaches may unite mental health stakeholders around shared targets to reduce mental illness. It first presents key approaches to clinical staging and then outlines how clinical knowledge has been translated into a unified transdiagnostic staging heuristic and clinical service structure over the past 30 years. Directions for short-, medium-, and long-term action are recommended with global community engagement. With investment from the mental health community, staging could reduce suffering through the use of an ethical, organized, and targeted system of communication.

Effects of inflammatory endotypes on disease trajectory in chronic rhinosinusitis with nasal polyps

BACKGROUND

Although phenotypic features have traditionally guided treatment in chronic rhinosinusitis, recent research has favored categorization on the basis of inflammatory endotype. However, the impact of endotypic differeces on clinical outcomes remains largely unknown.

OBJECTIVE

We sought to compare disease trajectory, primarily time-to-polyp recurrence, between chronic rhinosinusitis with nasal polyp (CRSwNP) endotypes.

METHODS

Samples were obtained from patients with CRSwNP undergoing surgery between 2015 and 2023, and cytokine levels were measured using a multiplex bead assay. Principal-component analysis followed by hierarchical cluster analysis was used to identify endotype clusters. Clinical outcomes were subsequently compared between clusters.

RESULTS

Six CRSwNP disease clusters were identified among the 269 included patients. Cluster 1 (46.5%) was characterized by relatively low inflammation. Cluster 4 (13.3%) and cluster 6 (7.1%) also exhibited low inflammation but with elevated levels of IL-12/IL-21 and CCL5, respectively. Cluster 2 (4.5%) represented a mixed type 1/3 inflammatory endotype (IFN-γHigh/IL-4High/IL-17AHigh), and cluster 3 (10.0%) was characterized by an innate, proinflammatory response (IL-1βHigh/IL-6High/IL-8High). Cluster 5 (18.9%) exhibited type 2-dominant inflammation (IL-5High/IL-9High/IL-13High). When comparing disease trajectory, cluster 2 (IFN-γHigh/IL-4High/IL-17AHigh) and cluster 4 (IL-12High/IL-21High) had the shortest time-to-polyp recurrence, whereas cluster 3 (IL-1βHigh/IL-6High/IL-8High) demonstrated the longest time-to-recurrence (P < .001). Time-to-oral steroid course (P = .13) and time-to-biologic therapy (P = .43) were similar across clusters.

CONCLUSIONS

The study highlights the heterogeneous nature of CRSwNP and differences in disease trajectory between endotypes, notably that patients with mixed type 1 and type 3 inflammation demonstrate more recalcitrant disease. These findings suggest that therapies beyond traditional type 2 inflammation treatments may be needed to effectively reduce CRSwNP disease recurrence.

Inflammatory biomarkers and childhood maltreatment: A cluster analysis in patients with eating disorders

Eating Disorders (EDs) are severe psychiatric disorders, with growing evidence pointing towards the role of childhood maltreatment (CM) influencing their onset, severity, and response to treatment. Preliminary evidence showed that CM could be associated with an elevation of inflammatory biomarkers across the different EDs. The objective of the study was to elucidate the interplay between CM, ED-specific psychopathology, and inflammatory biomarkers. The study involved 198 female participants, comprising 70 patients with anorexia nervosa (AN), 56 patients with bulimia nervosa (BN), and 72 healthy controls (HCs). K-means clustering was used to assess the hypothesis that latent clusters could be described between patients affected by EDs based on serum levels of inflammatory biomarkers alone (CRP, IL-6, suPAR). Additionally, the analysis included a comparison between patients with and without history of childhood maltreatment. Patients with AN exhibited significantly higher suPAR levels than HCs, regardless of the severity of psychopathology. A direct association between CM and elevated levels of inflammatory biomarkers, particularly CRP, IL-6, and suPAR were found. Cluster analysis identified two distinct populations among patients with EDs, with the group showing elevated inflammatory biomarkers likely to report more severe CM. Even though preliminary, the results of the present study support the existence of a biologically grounded "maltreated eco-phenotype" in EDs. The present study also reports results on CRP, IL-6 and suPAR, in patients with EDs. These findings might suggest future potential tailored treatments and interventions designed to target specific subgroups of patients, and potentially improving treatment efficacy.

Biochemical differences based on sex and clusters of biomarkers in patients with COVID-19: analysis from the CARDIO COVID 19-20 registry

BACKGROUND

The inflammatory response associated with COVID-19 varies with sex, potentially affecting disease outcomes. Males have a higher risk of complications compared to females, requiring an evaluation of differences in inflammatory response severity based on sex.

OBJECTIVE

To compare clinical data, biochemical biomarkers, and outcomes among hospitalized COVID-19 patients in Latin America and the Caribbean (LA&C) based on sex and to perform a cluster analysis of biomarker profiles for both sexes.

METHODS

This prospective, multicenter observational registry made by the Inter-American Council of Heart Failure and Pulmonary Hypertension of the Inter-American Society of Cardiology included hospitalized COVID-19 patients from 44 hospitals in 14 countries in LA&C between May 1, 2020, and June 30, 2021.

RESULTS

Of 3,260 patients (1,201 females and 2,059 males), males had higher C-reactive protein and ferritin levels, while females had higher natriuretic peptides and d-dimer levels. Males had more cardiovascular complications (acute coronary syndrome [3.3% vs. 2.2%], decompensated heart failure [8.9% vs. 7.8%], pulmonary embolism [4.4% vs. 2.9%]), intensive care unit (ICU) admissions (56.9% vs. 47.7%), and overall mortality (27.5% vs. 22.1%). Cluster analysis identified three groups: one with normal-range biomarkers but elevated ferritin, one with coagulation abnormalities, and one with an inflammatory profile linked to renal injury and increased non-cardiovascular mortality.

CONCLUSIONS

In the LA&C population hospitalized with COVID-19, males had higher inflammatory biomarker levels, correlating with increased cardiovascular complications and mortality. The cluster with an inflammatory profile showed higher non-cardiovascular mortality, while clusters with elevated ferritin levels were associated with increased ICU admissions.

A systematic review of multimorbidity clusters in heart failure: Effects of methodologies

BACKGROUND

Clustering algorithms can identify distinct heart failure (HF) subgroups. The choice of algorithms, modelling process, and input variables can impact clustering outcomes. Therefore, we reviewed analytical methods and variables used in studies that performed clustering in patients with HF.

METHODS

We systematically searched CINAHL, COCHRANE, EMBASE, OVID Medline, and Web of Science for eligible articles between inception and April 2023. We included primary studies that identified distinct HF multimorbid subgroups and were appraised for risk-of-bias and against methodological recommendations for cluster analysis. A narrative synthesis was performed.

RESULTS

Our analysis included 43 studies, mostly following a cohort design (n = 34, 79 %) and conducted primarily in Europe (n = 15, 35 %) and North America (n = 13, 30 %). Model-based (n = 22, 48 %), centre-based (n = 10, 22 %), and hierarchical class clustering (n = 9, 20 %) were the most frequently employed algorithms, identifying a range of 2-10 multimorbid clusters. Most studies used a combination of multi-modal parameters (i.e., socio-demographics, biochemistry, clinical characteristics, comorbidities and risk factors, cardiac imaging, and biomarkers) (n = 27, 63 %), followed by disease-based parameters (i.e., comorbidities and risk factors) (n = 11, 26 %) as input variables for clustering. Notably, variables used for clustering reflected cardiovascular and metabolic conditions. The phenogroups identified differed by input variables and algorithms used for clustering. We found substantial quality gaps in developing clustering models, variable selection, reporting of modelling processes, and model validation.

CONCLUSION

Cluster analysis results differed based on the clustering algorithms used and input variables. This review found substantial gaps in analysis quality and reporting. Implementing a methodological framework to develop, validate, and report clustering analysis can improve the clinical utility and reproducibility of clustering outcomes.

Identifying High-Risk Populations for Sexually Transmitted Infections in Chinese Men Who Have Sex With Men: A Cluster Analysis

Background

This study aimed to identify subpopulations of Chinese men who have sex with men (MSM) with distinct sexual behavioral patterns and explore their correlations with sexually transmitted infections (STIs).

Methods

We recruited 892 eligible MSM in Xi'an, China, collecting sociodemographic, sexual behavior, and STI data. Cluster analysis identified distinct sexual behavioral patterns, allowing us to examine STI differences across clusters.

Results

Among the 892 MSM analyzed, 3 clusters were identified. Cluster 1 (n = 157) exhibited high-risk sexual behavioral patterns, including the highest median number of sexual partners (5 vs 1 in cluster 2 vs 3 in cluster 3, P < .001), lowest consistent condom use for insertive anal sex (0% vs 64.12% vs 99.76%, P = .004) and receptive anal sex (9.22% vs 67.71% vs 98.91%, P = .006), highest uncertainty of partners' STIs (77.07% vs 57.89% vs 64.5%, P < .001), all recent partners being casual, longest length of sequential sexual acts (6 vs 5 vs 5, P = .045), and highest rates of gonorrhea (20.38% vs 10.09% vs 14.99%, P = .019) and chlamydia (16.56% vs 8.33% vs 13.21%, P = .045). Cluster 2 (n = 228) showed the lowest engagement in high-risk behaviors and STIs, characterized by the fewest sexual partners, highest certainty of partner's STIs, and all recent partners being regular. Cluster 3 (n = 507) showed moderate levels of high-risk behaviors and STIs, with the highest consistent condom use during anal sex.

Conclusions

This study identified 3 subpopulations of Chinese MSM with distinct sexual behavioral patterns. Targeted public health interventions to the most at-risk subpopulations of MSM are essential for STI prevention.

Machine learning in the diagnosis, management, and care of patients with low back pain: a scoping review of the literature and future directions

BACKGROUND CONTEXT

Low back pain (LBP) remains the leading cause of disability globally. In recent years, machine learning (ML) has emerged as a potentially useful tool to aid the diagnosis, management, and prognostication of LBP.

PURPOSE

In this review, we assess the scope of ML applications in the LBP literature and outline gaps and opportunities.

STUDY DESIGN/SETTING

A scoping review was performed in accordance with the Preferred Reporting Items for Systematic reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) guidelines.

METHODS

Articles were extracted from the Web of Science, Scopus, PubMed, and IEEE Xplore databases. Title/abstract and full-text screening was performed by two reviewers. Data on model type, model inputs, predicted outcomes, and ML methods were collected.

RESULTS

In total, 223 unique studies published between 1988 and 2023 were identified, with just over 50% focused on low-back-pain detection. Neural networks were used in 106 of these articles. Common inputs included patient history, demographics, and lab values (67% total). Articles published after 2010 were also likely to incorporate imaging data into their models (41.7% of articles). Of the 212 supervised learning articles identified, 168 (79.4%) mentioned use of a training or testing dataset, 116 (54.7%) utilized cross-validation, and 46 (21.7%) implemented hyperparameter optimization. Of all articles, only 8 included external validation and 9 had publicly available code.

CONCLUSIONS

Despite the rapid application of ML in LBP research, a majority of articles do not follow standard ML best practices. Furthermore, over 95% of articles cannot be reproduced or authenticated due to lack of code availability. Increased collaboration and code sharing are needed to support future growth and implementation of ML in the care of patients with LBP.

Identification of Physician Concerns Regarding Implementation of the Nagasaki Acute Myocardial Infarction Secondary Prevention Clinical Pathway

Background

The Nagasaki Acute Myocardial Infarction Secondary Prevention Clinical Pathway (NASP) is a regional pathway that aims to standardize practices related to the treatment of acute myocardial infarction in order to improve patient prognoses. This study aimed to understand physician backgrounds and concerns regarding implementation of the NASP.

Methods and Results

This exploratory sequential mixed-methods study was developed around the RE-AIM (reach, effectiveness, adoption, implementation, and maintenance) framework. Following focus group interviews, the web-based, self-administered questionnaire survey with a cross-sectional study design was given to 62 physicians who practiced at acute care hospitals (ACHs), primary care hospitals (PCHs), or outpatient clinics (OCs) in the Nagasaki prefecture. Hayashi's quantitative theory type II analysis was used to assess the quantitative relationship between physician characteristics and their concerns. In addition, physicians were clustered based on the types of concerns they had. Our results demonstrated that specialists in cardiovascular disease held more concerns regarding implementation of the NASP. Furthermore, workload burden was found to be the most common concern among these physicians. Cooperation between physicians at ACHs and physicians at PCHs/OCs was also found to be vital for the NASP.

Conclusions

Interventions such as modifications to the NASP operation may assist in alleviating concerns regarding the NASP and allow for the development of tailored interventions and effective expansion of the pathway.

Clustering Methods in Rheumatic and Musculoskeletal Disease Research: An Educational Guide to Best Research Practices

Clinical manifestations and disease progression often exhibit significant variability among patients with rheumatic diseases, complicating diagnosis and treatment strategies. A better understanding of disease heterogeneity may allow for personalized treatment strategies. Cluster analysis is a class of statistical methods that aims to identify subgroups or patterns within a dataset. Cluster analysis is a type of unsupervised learning, meaning there are no outcomes or labels to guide the analysis (ie, there is no ground truth). This makes it difficult to assess the accuracy or validity of the identified clusters, and these methods therefore require thoughtful planning and careful interpretation. Here, we provide a high-level overview of clustering, including different types of clustering methods and important considerations when undertaking clustering, and review some examples from the rheumatology literature.

Professional reintegration of stroke survivors and their mental health, quality of life and community integration

PURPOSE

To assess the association between professional reintegration and mental health, quality of life (QoL) and community reintegration of stroke survivors.

METHODS

Using a cross-sectional study design, a structured questionnaire was administered to previously working stroke survivors, 18-24 months post-stroke. Data on sociodemographic characteristics, professional reintegration (prevalence of return to work (RTW), period of RTW, job placement, function at work, reintegration support, association of stroke with work and number of working hours), mental health (Hospital Anxiety and Depression Questionnaire), QoL (Stroke Specific Quality of Life Scale) and community integration (Community Integration Questionnaire) were reported by 553 stroke survivors.

RESULTS

Twenty months after stroke, 313 (56.6%; 95%CI 52.4-60.8) stroke survivors had return to work. RTW was positively associated with both global and sub-domains scores of Community Integration Questionnaire (CIQ) (global CIQ β = 3.50; 95%CI 3.30-3.79) and with depressive symptomatology (β = 0.63; 95%CI 0.20-1.46) measured by the Hospital Anxiety and Depression Scale. No significant differences were found regarding QoL, according to RTW status. For those who RTW, no significant associations were found between any of the professional reintegration determinants assessed and mental health, QoL and community integration scores.

CONCLUSIONS

RTW seems to be associated to better community integration after stroke, but appears to be negatively associated to stroke survivor's mental health, namely considering depression symptoms. Future studies should explore the barriers to stroke survivors' RTW and the challenges and strategies used to overcome them, to allow the development of professional reintegration policies.

Unsupervised Clustering in Neurocritical Care: A Systematic Review

Managing patients with acute brain injury in the neurocritical care (NCC) unit has become increasingly complex because of technological advances and increasing information derived from multiple data sources. Diverse data streams necessitate innovative approaches for clinicians to understand interactions between recorded variables. Unsupervised clustering integrates different data streams and could be supportive. Here, we provide a systematic review on the use of unsupervised clustering using NCC data. The primary objective was to provide an overview of clustering applications in NCC studies. As a secondary objective, we discuss considerations for future NCC studies. Databases (Medline, Scopus, Web of Science) were searched for unsupervised clustering in acute brain injury studies including traumatic brain injury (TBI), subarachnoid hemorrhage, intracerebral hemorrhage, acute ischemic stroke, and hypoxic-ischemic brain injury published until  March 13th 2024. We performed the systematic review in accordance with the Preferred Reporting Items for Systematic reviews and Meta-Analyses guidelines. We identified 18 studies that used unsupervised clustering in NCC. Predominantly, studies focused on patients with TBI (12 of 18 studies). Multiple research questions used a variety of resource data, including demographics,  clinical- and monitoring data, of which intracranial pressure was most often included (8 of 18 studies). Studies also covered various clustering methods, both traditional methods (e.g., k-means) and advanced methods, which are able to retain the temporal aspect. Finally, unsupervised clustering identified novel phenotypes for clinical outcomes in 9 of 12 studies. Unsupervised clustering can be used to phenotype NCC patients, especially patients with TBI, in diverse disease stages and identify clusters that may be used for prognostication. Despite the need for validation studies, this methodology could help to improve outcome prediction models, diagnostics, and understanding of pathophysiology.Registration number: PROSPERO: CRD4202347097676.

The Arabidopsis splicing factor PORCUPINE/SmE1 orchestrates temperature-dependent root development via auxin homeostasis maintenance

Appropriate abiotic stress response is pivotal for plant survival and makes use of multiple signaling molecules and phytohormones to achieve specific and fast molecular adjustments. A multitude of studies has highlighted the role of alternative splicing in response to abiotic stress, including temperature, emphasizing the role of transcriptional regulation for stress response. Here we investigated the role of the core-splicing factor PORCUPINE (PCP) on temperature-dependent root development. We used marker lines and transcriptomic analyses to study the expression profiles of meristematic regulators and mitotic markers, and chemical treatments, as well as root hormone profiling to assess the effect of auxin signaling. The loss of PCP significantly alters RAM architecture in a temperature-dependent manner. Our results indicate that PCP modulates the expression of central meristematic regulators and is required to maintain appropriate levels of auxin in the RAM. We conclude that alternative pre-mRNA splicing is sensitive to moderate temperature fluctuations and contributes to root meristem maintenance, possibly through the regulation of phytohormone homeostasis and meristematic activity.

Determinants of Citizens' Intention to Participate in Self-Led Contact Tracing: Cross-Sectional Online Questionnaire Study

BACKGROUND

Contact tracing (CT) is a key intervention to contain outbreaks of communicable diseases. During large-scale outbreaks, public health services may lack the resources required to perform CT effectively. One way of mitigating this issue is to shift some of the tasks in CT normally performed by public health services to cases and their contacts, supported by digital tools. We refer to this as "self-led CT." However, while the effectiveness of the self-led CT inherently depends on the willingness and skills of citizens to participate, the determinants of citizens' intention to participate in self-led CT are not yet fully understood.

OBJECTIVE

We aimed to identify determinants of Dutch citizens' intention to participate in self-led CT and assess their potential for behavioral change, so as to identify "behavior change targets," which may be used in the development and implementation of self-led CT to increase citizens' intention to participate.

METHODS

In March 2022, we performed an online cross-sectional questionnaire study. The questionnaire was developed based on findings from a previous exploratory semistructured interview study and distributed among a Dutch consumer panel. Using all questionnaire items as potential predictors, we performed a random forest analysis to identify determinants of citizens' intention to participate in self-led CT. We then performed an Agglomerative Hierarchical Cluster Analysis to identify groups of related determinants that may be considered overarching behavior change targets. Finally, we used Confidence Interval-Based Estimation of Relevance and calculated the Potential for Change Indices to compare the potential for behavioral change of the selected individual determinants and determinant clusters.

RESULTS

The questionnaire was completed by 3019 respondents. Our sample is representative of the Dutch population in terms of age, gender, educational level, and area of residence. Out of 3019 respondents, 2295 (76%) had a positive intention to participate in self-led CT. We identified 20 determinants of citizens' intention that we grouped into 9 clusters. In general, increasing citizens' trust in the digital tools developed for self-led CT has the highest potential to increase citizens' intention, followed by increasing the belief that using digital tools makes participating in self-led CT easier, reducing privacy-related concerns, and increasing citizens' willingness-and sense of responsibility-to cooperate in CT in general.

CONCLUSIONS

Overall, Dutch citizens are positive toward participating in self-led CT. Our results provide directions for the development and implementation of self-led CT, which may be particularly useful in preparing for future, large-scale outbreaks.

Capturing the clinical complexity in young people presenting to primary mental health services: a data-driven approach

AIMS

The specific and multifaceted service needs of young people have driven the development of youth-specific integrated primary mental healthcare models, such as the internationally pioneering headspace services in Australia. Although these services were designed for early intervention, they often need to cater for young people with severe conditions and complex needs, creating challenges in service planning and resource allocation. There is, however, a lack of understanding and consensus on the definition of complexity in such clinical settings.

METHODS

This retrospective study involved analysis of headspace's clinical minimum data set from young people accessing services in Australia between 1 July 2018 and 30 June 2019. Based on consultations with experts, complexity factors were mapped from a range of demographic information, symptom severity, diagnoses, illness stage, primary presenting issues and service engagement patterns. Consensus clustering was used to identify complexity subgroups based on identified factors. Multinomial logistic regression was then used to evaluate whether these complexity subgroups were associated with other risk factors.

RESULTS

A total of 81,622 episodes of care from 76,021 young people across 113 services were analysed. Around 20% of young people clustered into a 'high complexity' group, presenting with a variety of complexity factors, including severe disorders, a trauma history and psychosocial impairments. Two moderate complexity groups were identified representing 'distress complexity' and 'psychosocial complexity' (about 20% each). Compared with the 'distress complexity' group, young people in the 'psychosocial complexity' group presented with a higher proportion of education, employment and housing issues in addition to psychological distress, and had lower levels of service engagement. The distribution of complexity profiles also varied across different headspace services.

CONCLUSIONS

The proposed data-driven complexity model offers valuable insights for clinical planning and resource allocation. The identified groups highlight the importance of adopting a holistic and multidisciplinary approach to address the diverse factors contributing to clinical complexity. The large number of young people presenting with moderate-to-high complexity to headspace early intervention services emphasises the need for systemic change in youth mental healthcare to ensure the availability of appropriate and timely support for all young people.

Novel Sleep Phenotypic Profiles Associated With Incident Atrial Fibrillation in a Large Clinical Cohort

BACKGROUND

While sleep disorders are implicated in atrial fibrillation (AF), the interplay of physiologic alterations and symptoms remains unclear. Sleep-based phenotypes can account for this complexity and translate to actionable approaches to identify at-risk patients and therapeutic interventions.

OBJECTIVES

This study hypothesized discrete phenotypes of symptoms and polysomnography (PSG)-based data differ in relation to incident AF.

METHODS

Data from the STARLIT (sleep Signals, Testing, And Reports LInked to patient Traits) registry on Cleveland Clinic patients (≥18 years of age) who underwent PSG from November 27, 2004, to December 30,2015, were retrospectively examined. Phenotypes were identified using latent class analysis of symptoms and PSG-based measures of sleep-disordered breathing and sleep architecture. Phenotypes were included as the primary predictor in a multivariable-adjusted Cox proportional hazard models for incident AF.

RESULTS

In our cohort (N = 43,433, age 51.8 ± 14.5 years, 51.9% male, 74.9% White), 7.3% (n = 3,166) had baseline AF. Over a 7.6- ± 3.4-year follow-up period, 8.9% (n = 3,595) developed incident AF. Five phenotypes were identified. The hypoxia subtype (n = 3,245) had 48% increased incident AF (HR: 1.48; 95% CI: 1.34-1.64), the apneas + arousals subtype (n = 4,592) had 22% increased incident AF (HR: 1.22; 95% CI: 1.10-1.35), and the short sleep + nonrapid eye movement subtype (n = 6,126) had 11% increased incident AF (HR: 1.11; 95% CI: 1.01-1.22) compared with long sleep + rapid eye movement (n = 26,809), the reference group. The hypopneas subtype (n = 2,661) did not differ from reference (HR: 0.89; 95% CI: 0.77-1.03).

CONCLUSIONS

Consistent with prior evidence supporting hypoxia as an AF driver and cardiac risk of the sleepy phenotype, this constellation of symptoms and physiologic alterations illustrates vulnerability for AF development, providing potential value in enhancing our understanding of integrated sleep-specific symptoms and physiologic risk of atrial arrhythmogenesis.

Self-organising maps in the analysis of strains of human abdominal wall to identify areas of similar mechanical behaviour

The study refers to the application of a type of artificial neural network called the Self-Organising Map (SOM) for the identification of areas of the human abdominal wall that behave in a similar mechanical way. The research is based on data acquired during in vivo tests using the digital image correlation technique (DIC). The mechanical behaviour of the human abdominal wall is analysed during changing intra-abdominal pressure. SOM allow to study simultaneously three variables in four time/load steps. The variables refer to the principal strains and their directions. SOM classifies all the abdominal surface data points into clusters that behave similarly in accordance with the 12 variables. The analysis of the clusters provides a better insight into abdominal wall deformation and its evolution under pressure than when observing a single mechanical variable. The presented results may provide a better understanding of the mechanics of the living human abdominal wall. It might be particularly useful when selecting proper implants as well as for the design of surgical meshes for the treatment of abdominal hernias, which would be mechanically compatible with identified regions of the human anterior abdominal wall, and possibly open the way for patient-specific solutions.

Inter and intra-operator reliability of Lekholm and Zarb classification and proposal of a novel radiomic data-driven clustering for qualitative assessment of edentulous alveolar ridges

OBJECTIVES

The present study was conducted to evaluate the reproducibility of Lekholm and Zarb classification system (L&Z) for bone quality assessment of edentulous alveolar ridges and to investigate the potential of a data-driven approach for bone quality classification.

MATERIALS AND METHODS

Twenty-six expert clinicians were asked to classify 110 CBCT cross-sections according to L&Z classification (T0). The same evaluation was repeated after one month with the images put in a different order (T1). Intra- and inter-examiner agreement analyses were performed using Cohen's kappa coefficient (CK) and Fleiss' kappa coefficient (FK), respectively. Additionally, radiomic features extraction was performed from 3D edentulous ridge blocks derived from the same 110 CBCTs, and unsupervised clustering using 3 different clustering methods was used to identify patterns in the obtained data.

RESULTS

Intra-examiner agreement between T0 and T1 was weak (CK 0.515). Inter-examiner agreement at both time points was minimal (FK at T0: 0.273; FK at T1: 0.243). The three different unsupervised clustering methods based on radiomic features aggregated the 110 CBCTs in three groups in the same way.

CONCLUSIONS

The results showed low agreement among clinicians when using L&Z classification, indicating that the system may not be as reliable as previously thought. The present study suggests the possible application of a reproducible data-driven approach based on radiomics for the classification of edentulous alveolar ridges, with potential implications for improving clinical outcomes. Further research is needed to determine the clinical significance of these findings and to develop more standardized and accurate methods for assessing bone quality of edentulous alveolar ridges.

Artificial intelligence in perinatal mental health research: A scoping review

The intersection of Artificial Intelligence (AI) and perinatal mental health research presents promising avenues, yet uncovers significant challenges for innovation. This review explicitly focuses on this multidisciplinary field and undertakes a comprehensive exploration of existing research therein. Through a scoping review guided by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) framework, we searched relevant literature spanning a decade (2013-2023) and selected fourteen studies for our analysis. We first provide an overview of the main AI techniques and their development, including traditional methods across different categories, as well as recent emerging methods in the field. Then, through our analysis of the literature, we summarize the predominant AI and ML techniques adopted and their applications in perinatal mental health studies, such as identifying risk factors, predicting perinatal mental health disorders, voice assistants, and Q&A chatbots. We also discuss existing limitations and potential challenges that hinder AI technologies from improving perinatal mental health outcomes, and suggest several promising directions for future research to meet real needs in the field and facilitate the translation of research into clinical settings.

Association Between Sleep Quality and Deep Learning-Based Sleep Onset Latency Distribution Using an Electroencephalogram

To evaluate sleep quality, it is necessary to monitor overnight sleep duration. However, sleep monitoring typically requires more than 7 hours, which can be inefficient in termxs of data size and analysis. Therefore, we proposed to develop a deep learning-based model using a 30 sec sleep electroencephalogram (EEG) early in the sleep cycle to predict sleep onset latency (SOL) distribution and explore associations with sleep quality (SQ). We propose a deep learning model composed of a structure that decomposes and restores the signal in epoch units and a structure that predicts the SOL distribution. We used the Sleep Heart Health Study public dataset, which includes a large number of study subjects, to estimate and evaluate the proposed model. The proposed model estimated the SOL distribution and divided it into four clusters. The advantage of the proposed model is that it shows the process of falling asleep for individual participants as a probability graph over time. Furthermore, we compared the baseline of good SQ and SOL and showed that less than 10 minutes SOL correlated better with good SQ. Moreover, it was the most suitable sleep feature that could be predicted using early EEG, compared with the total sleep time, sleep efficiency, and actual sleep time. Our study showed the feasibility of estimating SOL distribution using deep learning with an early EEG and showed that SOL distribution within 10 minutes was associated with good SQ.

Applying Machine Learning Techniques to Implementation Science

Machine learning (ML) approaches could expand the usefulness and application of implementation science methods in clinical medicine and public health settings. The aim of this viewpoint is to introduce a roadmap for applying ML techniques to address implementation science questions, such as predicting what will work best, for whom, under what circumstances, and with what predicted level of support, and what and when adaptation or deimplementation are needed. We describe how ML approaches could be used and discuss challenges that implementation scientists and methodologists will need to consider when using ML throughout the stages of implementation.