Rich Data Research Articles

Self-Supervised Marine Noise Learning with Sparse Autoencoder Network for Generative Target Magnetic Anomaly Detection

As an effective physical field feature to perceive ferromagnetic targets, magnetic anomaly is widely used in covert marine surveillance tasks. However, its practical usability is affected by the complex marine magnetic noise interference, making robust magnetic anomaly detection (MAD) quite a challenging task. Recently, learning-based detectors have been widely studied for the discrimination of magnetic anomaly signal and achieve superior performance than traditional rule-based detectors. Nevertheless, learning-based detectors require abundant data for model parameter training, which are difficult to access in practical marine applications. In practice, target magnetic anomaly data are usually expensive to acquire, while rich marine magnetic noise data are readily available. Thus, there is an urgent need to develop effective models to learn discriminative features from the abundant marine magnetic noise data for newly appearing target anomaly detection. Motivated by this, in this paper we formulate MAD as a single-edge detection problem and develop a self-supervised marine noise learning approach for target anomaly classification. Specifically, a sparse autoencoder network is designed to model the marine noise and restore basis geomagnetic field from the collected noisy magnetic data. Subsequently, reconstruction error of the network is used as a statistical decision criterion to discriminate target magnetic anomaly from cluttered noise. Finally, we verify the effectiveness of the proposed approach on real sea trial data and compare it with seven state-of-the-art MAD methods on four numerical indexes. Experimental results indicate that it achieves a detection accuracy of 93.61% and has a running time of 21.06 s on the test dataset, showing superior MAD performance over its counterparts.

Rapid assessment of human–elephant conflict: a crime science approach

An interdisciplinary approach has the potential not only to help solve conservation-centric problems but also to enrich and improve evidence-based scientific research. Crime science, an offshoot of criminology, provides a comprehensive, solution-oriented approach that transcends disciplinary boundaries and bridges science and practice for developing effective conservation interventions to real-life problems such as Human Elephant Conflict (HEC). This paper focuses on HEC as a conservation concern, but the resultant behaviors toward elephants, people, and their property are criminology’s concern. Using the Action Research paradigm, a rapid assessment of human–elephant conflict (HEC) in India was undertaken to identify contextual solutions. This study utilized problem-oriented field research methods that enabled the gathering of data on elephant habitat-landscape, villagers’ lifestyle (habitat) in the fringe areas, their current approaches in dealing with the conflict, the challenges forest officials face to mitigate HEC, and the assistance provided by district administrators to protect villagers and their corps and HEC-related deaths. The qualitative inquiry, including observation of village-forest fringe areas, focus group discussions with villagers, and interviews with forest officers and rangers, and district administrators/collectors who are handlers, guardians, and managers of the conflict space, provided rich data in identifying situational practical measures and underscored the role of crime science in providing a conceptual framework to gather evidence in addressing HEC in forest areas. The findings of the research suggest that human–animal convergence space is the source (or location) of conflict and criminology-driven situational crime prevention measures, including increasing effort, risks, reducing rewards and provocations, and removing excuses might mitigate the conflict, requiring coordinated efforts by villagers, forest and district administrators, and local law enforcers.

Multilevel Longitudinal Functional Principal Component Model.

Sensor devices, such as accelerometers, are widely used for measuring physical activity (PA). These devices provide outputs at fine granularity (e.g., 10-100Hz or minute-level), which while providing rich data on activity patterns, also pose computational challenges with multilevel densely sampled data, resulting in PA records that are measured continuously across multiple days and visits. On the other hand, a scalar health outcome (e.g., BMI) is usually observed only at the individual or visit level. This leads to a discrepancy in numbers of nested levels between the predictors (PA) and outcomes, raising analytic challenges. To address this issue, we proposed a multilevel longitudinal functional principal component analysis (mLFPCA) model to directly model multilevel functional PA inputs in a longitudinal study, and then implemented a longitudinal functional principal component regression (FPCR) to explore the association between PA and obesity-related health outcomes. Additionally, we conducted a comprehensive simulation study to examine the impact of imbalanced multilevel data on both mLFPCA and FPCR performance and offer guidelines for selecting optimal methods.

Model predictive complex system control from observational and interventional data

Complex systems, characterized by intricate interactions among numerous entities, give rise to emergent behaviors whose data-driven modeling and control are of utmost significance, especially when there is abundant observational data but the intervention cost is high. Traditional methods rely on precise dynamical models or require extensive intervention data, often falling short in real-world applications. To bridge this gap, we consider a specific setting of the complex systems control problem: how to control complex systems through a few online interactions on some intervenable nodes when abundant observational data from natural evolution is available. We introduce a two-stage model predictive complex system control framework, comprising an offline pre-training phase that leverages rich observational data to capture spontaneous evolutionary dynamics and an online fine-tuning phase that uses a variant of model predictive control to implement intervention actions. To address the high-dimensional nature of the state-action space in complex systems, we propose a novel approach employing action-extended graph neural networks to model the Markov decision process of complex systems and design a hierarchical action space for learning intervention actions. This approach performs well in three complex system control environments: Boids, Kuramoto, and Susceptible-Infectious-Susceptible (SIS) metapopulation. It offers accelerated convergence, robust generalization, and reduced intervention costs compared to the baseline algorithm. This work provides valuable insights into controlling complex systems with high-dimensional state-action spaces and limited intervention data, presenting promising applications for real-world challenges.

Major impacts of the US Renewable Fuel Standard on corn and soybean cultivation

AbstractWe use rich spatial data on observed crop choices and ethanol production to estimate the local impacts of ethanol plants and the overall impact of the 2007 US Renewable Fuel Standard (the RFS2) on crop patterns. We find that the RFS2 increased total corn area by 9.7%, corn monoculture by 16.8%, and soybean area by 6.7%, with notable spatial heterogeneity in both the overall impact and the local effects of ethanol plants. The results indicate an RFS2‐induced shift in the geography of corn and soybean cultivation within the Corn Belt—intensification in the core, and significant expansion near the periphery.

Experimental investigation and statistical analysis of the microwave drying process on the physical properties of Y2O3–Al2O3–ZrO2 composite ceramic powder

Y2O3–Al2O3–ZrO2 composite ceramic powders are an important structural and functional material. The agglomeration problem in the ceramic powder drying process must be resolved to prepare high-quality ceramic powder, and choosing the right drying techniques can significantly lessen ceramic powder agglomeration. Y2O3–Al2O3–ZrO2 ceramic powder was dried using microwave drying. The drying kinetics of the powders were examined with their initial mass, moisture content, and microwave power. As microwave power, beginning moisture content, and initial mass rose, the average drying rate also increased, according to the experimental data. To better characterize the microwave drying process, the drying data were fitted and analyzed using four thin-layer drying kinetic models, namely Quadratic Model, Modified Page, Wang and Singh, and Two-term exponential. The outcomes demonstrate the good fitting effect of the Modified Page model and the compliance of the fitting parameters with the law. The samples before and after drying were characterized by Fourier transform infrared spectroscopy. Calculating the diffusion coefficient by Fick's second law shows that the effective diffusion coefficient of Y2O3–Al2O3–ZrO2 composite ceramic powder is 5.6 × 10−11 m2/s when the initial moisture content is 5%, the microwave heating power is 560 W, and the initial mass is 25 g. The activation energy for microwave drying of Y2O3–Al2O3–ZrO2 ceramic powders was calculated to be 22.83 W/g according to the Arrhenius formula. This paper aims to provide a theoretical basis and rich experimental data for the study of microwave drying of Y2O3–Al2O3–ZrO2 composite ceramic powder.

Dropping out of learners with hearing impairment from inclusive classes in urban schools in Bulawayo

This study explored factors contributing to the dropping out of learners with a hearing impairment from inclusive classes. It used the interpretivism theory to extract rich data and the qualitative research approach to conduct the study. A case study was used as a research design. In this study, participants were purposively selected. Semi-structured interviews, observations, and document analysis were used to collect data. The findings from the study revealed the challenges encountered in the teaching and learning of learners with hearing impairment. These include curriculum adaptation, social interaction with peers, negative attitudes of teachers and hearing learners, parental involvement, communication barriers, lack of interpreters, high learner ratio, and lack of resources. The findings further revealed that some opportunities are being gained by learners with hearing impairment in inclusive classes, such as access to a wider curriculum, social interaction, enhanced communication, and learning in local schools. However, challenges seemed to outweigh the opportunities gained by learners with hearing impairment, compelling them to drop out of inclusive classes. Therefore, the study gave some recommendations to help curb the dropping out of learners with hearing impairment from inclusive classes. These include allocating more resources in inclusive classes to address the problem of inadequate materials and the fact that government policies on inclusion should be made available to all educators and fully implemented. In addition, supervision, monitoring, support from parents, and evaluation of the system should be done to facilitate the learning of learners with hearing impairment in inclusive classes.

Women improving nutrition through self-help groups in India: Does nutrition information help?

Women’s self-help groups (SHGs) are an important platform for reaching poor women in India. Despite SHGs' women-focused programming, evidence of the impact of SHG-based interventions on nutrition outcomes is limited, and most evaluations of nutrition interventions have not examined intermediate outcomes along the impact pathways or outcomes for women themselves. This paper evaluates the effectiveness of an integrated agriculture-nutrition intervention delivered through women’s SHGs in five states in central and eastern India. The interventions involved the delivery of nutrition behavior change communication to groups through participatory approaches, community engagement around key issues, and the strengthening of collective organizations. Our analysis is based on three rounds of rich panel data on close to 2700 rural women and their households from eight districts in these five states and qualitative work from an accompanying process evaluation. Using difference-in-difference models with nearest neighbor matching methods, we present results on women’s anthropometry and diet-related outcomes.We do not observe any improvements in women’s BMI or overall dietary diversity. Although more women in the nutrition intensification arm consumed animal source foods, nuts and seeds, and fruits, this was not enough to increase overall dietary diversity scores or the proportion of women achieving minimum dietary diversity. We measure intermediate outcomes along the program’s impact pathways and find improvements in household incomes, cultivation of home gardens, and utilization of government schemes but not in women’s empowerment. The lack of improvement in anthropometry and diets despite changes in some intermediate outcomes can be attributed to several factors such as low implementation intensity, poor facilitator capacity and incentives, the lack of relevance of the BCC topics to the average SHG member, and resource and agency constraints to adoption of recommended practices. Although we do not have data to test the parallel trends assumption and so do not interpret our results as causal, these findings do suggest that optimism about using group-based platforms needs to be tempered in resource-poor contexts.

Are genomic language models all you need? Exploring genomic language models on protein downstream tasks.

Large language models, trained on enormous corpora of biological sequences, are state-of-the-art for downstream genomic and proteomic tasks. Since the genome contains the information to encode all proteins, genomic language models (gLMs) hold the potential to make downstream predictions not only about DNA sequences, but also about proteins. However, the performance of gLMs on protein tasks remains unknown, due to few tasks pairing proteins with the coding DNA sequences (CDS) that can be processed by gLMs. In this work, we curated five such datasets and used them to evaluate the performance of gLMs and proteomic language models (pLMs). We show that gLMs are competitive and even outperform their pLMs counterparts on some tasks. The best performance was achieved using the retrieved CDS compared to sampling strategies. We found that training a joint genomic-proteomic model outperforms each individual approach, showing that they capture different but complementary sequence representations, as we demonstrate through model interpretation of their embeddings. Lastly, we explored different genomic tokenization schemes to improve downstream protein performance. We trained a new Nucleotide Transformer (50M) foundation model with 3mer tokenization that outperforms its 6mer counterpart on protein tasks while maintaining performance on genomics tasks. The application of gLMs to proteomics offers the potential to leverage rich CDS data, and in the spirit of the central dogma, the possibility of a unified and synergistic approach to genomics and proteomics. We make our inference code, 3mer pre-trained model weights and datasets available.

Identification of immunity- and ferroptosis-related signature genes as potential design targets for mRNA vaccines in AML patients.

Immune-associated ferroptosis plays an important role in the progression of acute myeloid leukemia (AML); however, the targets that play key roles in this process are currently unknown. This limits the development of mRNA vaccines based on immune-associated ferroptosis for clinical therapeutic applications. In this study, based on the rich data resources of the TCGA-LAML cohort, we analyzed the tumor mutational burden (TMB), gene mutation status, and associations between immune and ferroptosis genes to reveal the disease characteristics of AML patients. To gain a deeper understanding of differentially expressed genes, we applied the Limma package for differential expression analysis and integrated data sources such as ImmPort Shared Data and FerrDb V2. Moreover, we established gene modules related to TMB according to weighted gene coexpression network analysis (WGCNA) and explored the functions of these modules in AML and their relationships with TMB. We focused on the top 30 most frequent genes through a detailed survey of missense mutations and single nucleotide polymorphisms (SNPs) and selected potentially critical gene targets for subsequent analysis. Based on the expression of these genes, we successfully subgrouped AML patients and found that the subgroups associated with TMB (C1 and C2) exhibited significant differences in survival. The differences in the tumor microenvironment and immune cells between C1 and C2 patients were investigated with the ESTIMATE and MCP-counter algorithms. A predictive model of TMB-related genes (TMBRGs) was constructed, and the validity of the model was demonstrated by categorizing patients into high-risk and low-risk groups. The differences in survival between the high-risk patients and high-TMB patients were further investigated, and potential vaccine targets were identified via immune cell-level analysis. The identification of immunity- and ferroptosis-associated signature genes is an independent predictor of survival in AML patients and provides new information on immunotherapy for AML.

Robot Operating Systems–You Only Look Once Version 5–Fleet Efficient Multi-Scale Attention: An Improved You Only Look Once Version 5-Lite Object Detection Algorithm Based on Efficient Multi-Scale Attention and Bounding Box Regression Combined with Robot Operating Systems

This paper primarily investigates enhanced object detection techniques for indoor service mobile robots. Robot operating systems (ROS) supply rich sensor data, which boost the models’ ability to generalize. However, the model’s performance might be hindered by constraints in the processing power, memory capacity, and communication capabilities of robotic devices. To address these issues, this paper proposes an improved you only look once version 5 (YOLOv5)-Lite object detection algorithm based on efficient multi-scale attention and bounding box regression combined with ROS. The algorithm incorporates efficient multi-scale attention (EMA) into the traditional YOLOv5-Lite model and replaces the C3 module with a lightweight C3Ghost module to reduce computation and model size during the convolution process. To enhance bounding box localization accuracy, modified precision-defined intersection over union (MPDIoU) is employed to optimize the model, resulting in the ROS–YOLOv5–FleetEMA model. The results indicated that relative to the conventional YOLOv5-Lite model, the ROS–YOLOv5–FleetEMA model enhanced the mean average precision (mAP) by 2.7% post-training, reduced giga floating-point operations per second (GFLOPS) by 13.2%, and decreased the params by 15.1%. In light of these experimental findings, the model was incorporated into ROS, leading to the development of a ROS-based object detection platform that offers rapid and precise object detection capabilities.

Community-Engaged Survey Approach to Pandemic Impacts on Marginalized Communities, Massachusetts, 2020-2021.

Objectives. To describe how an innovative, community-engaged survey illuminated previously unmeasured pandemic inequities and informed health equity investments. Methods. The methodological approach of Massachusetts' COVID-19 Community Impact Survey, a cross-sectional online survey, was driven by key health equity principles: prioritizing community engagement, gathering granular and intersectional data, capturing root causes, elevating community voices, expediting analysis for timeliness, and creating data-to-action pathways. Data collection was deployed statewide in 11 languages from 2020 to 2021. Results. The embedded equity principles resulted in a rich data set and enabled analyses of populations previously undescribed. The final sample included 33 800 respondents including unprecedented numbers of populations underrepresented in traditional data sources. Analyses indicated that pandemic impacts related to basic needs, discrimination, health care access, workplace protections, employment, and mental health disproportionately affected these priority populations, which included Asian American/Pacific Islanders and parents. Conclusions. Equity-centered data approaches allow for analyses of populations previously invisible in surveillance data, enable more equitable public health action, and are both possible and necessary to deploy in state health departments. (Am J Public Health. 2024;114(S7):S599-S609. https://doi.org/10.2105/AJPH.2024.307800).

Candidate selection and its behavioral consequences

The rules by which parties select their candidates are important, not only because they determine the policy and leaders of a party. Candidate selection, the identity of those taking the final decision (selectorate), has been found to impact the behavior of parliamentarians around the world (e.g. Preece, 2014; Sieberer, 2006). However, there is disagreement among scholars as to which kind of effect we can expect, under which conditions and why. In light of an interesting, previously unstudied case – Switzerland – and on the basis of rich data relating to many different political activities we re-investigate the question of how candidate selection affects the political behavior. Our results indicate that in highly personalized political systems like the Swiss, MPs selected by a more inclusive selectorate do not differ significantly from other MPs in party loyalty but are more likely to engage in some activities that are more visible to their cantonal selectorate.

Integrating PCA with deep learning models for stock market Forecasting: An analysis of Turkish stocks markets

Financial data such as stock prices are rich time series data that contain valuable information for investors and financial professionals. Analysis of such data is critical to understanding market behaviour and predicting future price movements. However, stock price predictions are complex and difficult due to the intense noise, non-linear structures, and high volatility contained in this data. While this situation increases the difficulty of making accurate predictions, it also creates an important area for investors and analysts to identify opportunities in the market. One of the effective methods used in predicting stock prices is technical analysis. Multiple indicators are used to predict stock prices with technical analysis. These indicators formulate past stock price movements in different ways and produce signals such as buy, sell, and hold. In this study, the most frequently used ten different indicators were analyzed with PCA (Principal Component Analysis. This study aims to investigate the integration of PCA and deep learning models into the Turkish stock market using indicator values and to assess the effect of this integration on market prediction performance. The most effective indicators used as input for market prediction were selected with the PCA method, and then 4 different models were created using different deep learning architectures (LSTM, CNN, BiLSTM, GRU). The performance values of the proposed models were evaluated with MSE, MAE, MAPE and R2 measurement metrics. The results obtained show that using the indicators selected by PCA together with deep learning models improves market prediction performance. In particular, it was observed that one of the proposed models, the PCA-LSTM-CNN model, produced very successful results.

Unlocking the Potential of Free Text in Electronic Health Records with Large Language Models (LLM): Enhancing Patient Safety and Consultation Interactions.

Computer-mediated clinical consultation, involving clinicians, electronic health record (EHR) systems, and patients, yield rich narrative data. Despite advancements in Natural Language Processing (NLP), these narratives remain underutilised. Free text recording in EHRs allows expressivity, complements structured data from clinical coding systems, and facilitates collaborative care. Large language models (LLMs) excel in understanding and generating natural language, enabling complex dialogue processing. Integrating LLM tools into consultations could harness the untapped potential of free text to identify patient safety concerns, support diagnosis and provide content to enhance clinical-patient interactions. Tailoring LLMs for specific consultation tasks through pre-training and fine-tuning is viable. This paper outlines approaches for adopting LLMs in primary care and suggests that using fine-tuned LLMs with prompt engineering could enhance computer-mediated clinical consultation cost-effectively.

Rolling Through the Complications With the Clinical Judgment Model.

Nurse educators have been challenged to creatively implement the clinical judgment model (CJM) across the curriculum. Opportunities exist to utilize rich patient data in clinical debriefing. Rolling Through the Complications with the CJM, a scripted clinical debriefing tool, was created to help students think beyond the clinical day and apply clinical judgment to potential complications. In a pilot study, five medical-surgical clinical instructors implemented the clinical debriefing tool. Survey responses from 29 students following the clinical debriefing expressed the activity was a positive experience encouraging focused analysis beyond typical post-conference discussions. Comments focused on critical thinking, collaborative learning, application to future practice, and holistic care. Clinical post-conferences offer excellent opportunities to implement scripted debriefing encompassing the CJM. [J Nurs Educ. 2024;63(X):XXX-XXX.].

GenAI-Assisted Database Deployment for Heterogeneous Indigenous–Native Ethnographic Research Data

In ethnographic research, data collected through surveys, interviews, or questionnaires in the fields of sociology and anthropology often appear in diverse forms and languages. Building a powerful database system to store and process such data, as well as making good and efficient queries, is very challenging. This paper extensively investigates modern database technology to find out what the best technologies to store these varied and heterogeneous datasets are. The study examines several database categories: traditional relational databases, the NoSQL family of key-value databases, graph databases, document databases, object-oriented databases and vector databases, crucial for the latest artificial intelligence solutions. The research proves that when it comes to field data, the NoSQL lineup is the most appropriate, especially document and graph databases. Simplicity and flexibility found in document databases and advanced ability to deal with complex queries and rich data relationships attainable with graph databases make these two types of NoSQL databases the ideal choice if a large amount of data has to be processed. Advancements in vector databases that embed custom metadata offer new possibilities for detailed analysis and retrieval. However, converting contents into vector data remains challenging, especially in regions with unique oral traditions and languages. Constructing such databases is labor-intensive and requires domain experts to define metadata and relationships, posing a significant burden for research teams with extensive data collections. To this end, this paper proposes using Generative AI (GenAI) to help in the data-transformation process, a recommendation that is supported by testing where GenAI has proven itself a strong supplement to document and graph databases. It also discusses two methods of vector database support that are currently viable, although each has drawbacks and benefits.

Information preferences of patients with chronic blood cancer: A qualitative investigation.

Haematological malignancies (blood cancers) often follow chronic trajectories that can span many months or years. Management of these diseases typically involves periods of 'Watch & Wait', whereby patients are monitored by the hospital and only treated at progression, if or when this occurs, which might be never or on multiple occasions. Such remitting-relapsing pathways are unpredictable and can cause anxiety and distress. This study aimed to generate evidence about information preferences, with a view to underpinning future patient-facing resources; potentially mitigating psycho-social difficulties and promoting effective shared decision-making. A qualitative study was conducted, set within a UK population-based cohort of patients with haematological malignancies. Sampling was purposive, based on age (initially around the median age of diagnosis) and disease subtype (chronic lymphocytic leukaemia, follicular lymphoma, marginal zone lymphoma and myeloma); and in-depth interviews took place with 35 patients (10 with relatives). Analysis drew on qualitative description and thematic content analysis and included critical reading and annotation of transcripts, identification of common and rare phenomena, generation of codes and coding of material, and theme development. Patients discussed their preferences and experiences at length and rich data were generated from diagnosis onwards, across diagnostic subtypes. The overarching theme identified was 'Variations in preferences' with needs seen to differ from person to person; as well as changing over time for individuals. Five sub-themes were identified: 1) To know or not to know? 2) Needs are dynamic; 3) The polarising issue of prognosis; 4) Preferred sources; and 5) Differences in content, depth and presentation. Varied, dynamic information preferences indicate that resources should be developed in a way that provides maximum choice, enabling patients to select relevant material at different time-points on their trajectory. The development of blood cancer subtype-specific "real-world clinical scenarios" could improve patient experiences and inform shared decision-making.

Exploring perceptions of anti-vaping message themes: A qualitative study of Australian adolescents and adults.

Minimising electronic cigarette (e-cigarette) uptake and encouraging vaping cessation have become important components of public health agendas. Given the success of well-designed anti-smoking campaigns, attention has turned to developing health communications that target vaping. Although prior work has identified a range of potentially effective campaign messages, several gaps in the literature remain. We aimed to (i) identify message themes perceived to be effective at minimising e-cigarette use and (ii) elicit opinion on how these message themes could be optimised in terms of message execution and delivery. Focus groups (n = 16) were conducted with adolescents (14-15 and 16-17 year olds), young adults (18-24 year olds), and adults (25-39 year olds). The groups reported on message themes they believed would assist with minimising e-cigarette use. They then provided feedback on 14 message themes developed by the research team. Data were analysed using reflexive thematic analysis. Groups discussed a range of message themes they believed would be effective. These included themes focused on the health harms associated with vaping, the chemical ingredients in e-cigarettes, use dependence, and social attitudes to use. Groups discussed the importance of using evidence-based personal testimonies, and reported that messages would be most effectively communicated via social media, in public places, and at point of sale. Results build on existing research by identifying message themes perceived to be most effective, and how these would best be executed. Results have the potential to inform the development of co-designed health communication campaigns addressing youth vaping. Using a stratified focus group methodology, we collected rich and detailed qualitative data on Australians' perceptions of anti-vaping message themes. Themes considered most effective were those that addressed the health harms of vaping, the chemical ingredients in e-cigarettes, social consequences of use and the impacts of dependence. We recommend further work to develop and trial a range of specific anti-vaping messages based on the themes identified as having the most potential. This process should include iterative co-design with the target audience to optimise future health communication campaigns.

Acceptance and commitment coaching for music performance anxiety in adolescent singers.

Most of the anxiety disorders, particularly social anxiety, seem to develop either during childhood or adolescence. Adolescent singers who experience physical, mental and emotional changes along with voice change are particularly prone to the development of MPA. However, adolescence also seems to be an opportune time to instil healthy behaviours in singing students as they are more likely to remember these coping strategies, owing to the release of dopamine and the 'reminiscence bump'. As this period of a singer's life is wrought with inevitable anxiety development, the additional aim of the study was to develop a coaching framework which can be used by singing teachers in their practice. Mindset training for adolescents seems crucial to help them continue singing through puberty so they do not drop out of singing lessons or choir during voice change and identity development. The study aimed to determine if Acceptance and Commitment coaching could be used with adolescent singers with MPA and to record and analyse rich qualitative data in the form of semi-structured participant interviews and questionnaires. When their perception of MPA symptoms and physiological and psychological arousal before a performance changed, their behaviour and reactions changed accordingly. The change took place over a period of time, which was characterised by discoveries about themselves, confusion in understanding new concepts and letting go of old habits. Interestingly, by the end of the coaching sessions, their preoccupation with pleasing the audience and appearing perfect on stage was replaced by a new-found delight in pursuing values and goals related to their singing. Along with this came the acceptance of themselves and others as individuals with the potential for growth and change and the capacity to learn from mistakes. This study marks the first investigation into the effectiveness of using of ACC as an intervention for MPA in adolescent singers undergoing puberty by a singing teacher. The results are promising and suggest that ACC is an effective MPA intervention for adolescent singers to cope with inevitable development-related anxieties and keep them engaged in the activity of singing during their pubertal years.