Clinical Training Research Articles

Utilizing clinical assessment databases to enhance clinical training and research in neuropsychology: benefits, challenges, and practical strategies

ABSTRACT Research serves as the foundation of clinical neuropsychology, and strengthening research training pipelines is critical to sustaining innovation and scientific rigor within the field. Clinical assessment databases are one promising tool for achieving this goal. Herein, we describe the benefits and challenges of implementing clinical assessment databases across various neuropsychology training settings. First, we highlight their benefits, including the promotion of a standardized approach to clinical training, improvements in patient care, and increased opportunities for research. Next, we describe how clinical databases can be implemented across various training settings, including university-affiliated psychology training clinics, academic medical centers, Veterans Affairs hospitals, and private practice settings. The potential roles of faculty, trainees, and administrative staff in supporting database development and use are discussed. Lastly, we address practical strategies for overcoming implementation challenges, such as balancing both clinical and research demands, obtaining institutional review board approval, selecting appropriate data storage platforms, and ensuring data quality and consistency. Specific recommendations regarding platform storage are provided along with cost considerations. By integrating clinical assessment databases, neuropsychology training clinics can enhance their ability to provide evidence-based clinical training, deliver high quality care, and enhance research training. Although implementing clinical databases presents technical, logistical, and ethical challenges, these can be managed with careful planning.

Privacy-preserving Federated Learning and Uncertainty Quantification in Medical Imaging.

"Just Accepted" papers have undergone full peer review and have been accepted for publication in Radiology: Artificial Intelligence. This article will undergo copyediting, layout, and proof review before it is published in its final version. Please note that during production of the final copyedited article, errors may be discovered which could affect the content. Artificial Intelligence (AI) has demonstrated strong potential in automating medical imaging tasks, with potential applications across disease diagnosis, prognosis, treatment planning, and posttreatment surveillance. However, privacy concerns surrounding patient data remain a major barrier to the widespread adoption of AI in clinical practice, as large and diverse training datasets are essential for developing accurate, robust, and generalizable AI models. Federated Learning offers a privacy-preserving solution by enabling collaborative model training across institutions without sharing sensitive data. Instead, model parameters, such as model weights, are exchanged between participating sites. Despite its potential, federated learning is still in its early stages of development and faces several challenges. Notably, sensitive information can still be inferred from the shared model parameters. Additionally, postdeployment data distribution shifts can degrade model performance, making uncertainty quantification essential. In federated learning, this task is particularly challenging due to data heterogeneity across participating sites. This review provides a comprehensive overview of federated learning, privacy-preserving federated learning, and uncertainty quantification in federated learning. Key limitations in current methodologies are identified, and future research directions are proposed to enhance data privacy and trustworthiness in medical imaging applications. ©RSNA, 2025.

The effect of an educational orientation tour on anxiety of nursing students before their first clinical training: a quasi-experimental study

BackgroundThe first clinical experience in a hospital setting can be highly stressful for nursing students, often leading to significant anxiety. Addressing this issue requires effective interventions to help students transition smoothly into their clinical practice. This study aimed to evaluate the impact of an educational tour on preclinical anxiety among first-year nursing students.MethodsThis quasi-experimental study was conducted in 2024 in Tehran. A total of 72 s-semester nursing students were selected through convenience sampling and assigned to either the control group (n = 37) or the intervention group (n = 35). Before the start of clinical training, the intervention group participated in an educational tour of the hospital, conducted by the researcher, while the control group received no intervention. Data were collected using the Spielberger State-Trait Anxiety Inventory (STAI) one week before the clinical training and after its completion.ResultsAfter completing the clinical training, the mean trait anxiety score was 34.28 ± 7.89 in the control group and 31.30 ± 6.70 in the intervention group. However, an independent t-test revealed that the difference in post-intervention trait anxiety scores between the two groups was not statistically significant (p = 0.089). In contrast, the mean state anxiety score after the intervention was 36.14 ± 7.38 in the control group and 31.21 ± 6.86 in the intervention group, demonstrating a significant reduction in anxiety levels among students who participated in the educational tour (p = 0.005).ConclusionThe findings suggest that an educational tour can effectively reduce preclinical anxiety among first-year nursing students, making it a valuable strategy for improving their transition into hospital-based clinical training.

Development of an automated artificial intelligence-based tool for reticulin fibrosis assessment in bone marrow biopsies.

Bone marrow fibrosis plays a critical role in the diagnosis, prognosis, and management of haematological disorders, particularly myeloproliferative neoplasms like primary myelofibrosis. Accurate assessment of fibrosis, typically graded through histochemical techniques such as reticulin and trichrome staining, is essential but remains highly dependent on the pathologist's experience. To address the challenges of variability in interpretation and the increasing demand for standardized evaluations, we developed a digital pathology system for automated bone marrow reticulin fibrosis grading. This study utilized 86 bone marrow biopsy specimens from patients diagnosed with Philadelphia chromosome-negative myeloproliferative neoplasms, collected between 2018 and 2023. A fully convolutional network based on the InceptionV3 architecture was trained to assess fibrosis grades (MF0-MF3) from whole slide images of reticulin-stained sections. The model was trained using 3814 annotated images and validated using a separate set of 40 BMBs. The algorithm's performance was evaluated by comparing its fibrosis grading to expert hematopathologists' assessments, yielding a Cohen's kappa coefficient of 0.831, indicating excellent agreement. The algorithm showed strong concordance in fibrosis grading, especially for MF0 (k = 0.918) and MF3 (k = 0.886), and substantial agreement for intermediate grades (MF1 and MF2). Further validation across multiple institutions and scanning platforms confirmed the algorithm's robustness, with an overall agreement of 0.816. These results demonstrate the potential of digital pathology tools to provide standardized, reproducible fibrosis grading, thereby aiding pathologists in clinical decision-making and training.

The impact of scenario-based simulation training on nursing students’ knowledge and performance in patient care after coronary artery surgery

BackgroundCoronary artery bypass graft (CABG) surgery is one of the most effective treatments for coronary artery disease. Despite its many benefits, this surgery can also lead to potential complications. Studies have shown that nurses play an important role in managing these complications. Studies have shown that nurses play an important role in controlling these complications. Conventional nursing education often fails to prepare graduates for employment in complex and evolving clinical environments like cardiac wards. Meanwhile, scenario-based learning has recently emerged as a valuable educational method in the clinical training of healthcare professionals. Therefore, this study aimed to determine the effect of scenario-based learning on nursing students’ knowledge and performance in patient care after coronary artery surgery at Kashan University of Medical Sciences.MethodsThe present study used a pretest-posttest design in 2024(April to December), with the research population consisting of fifty nursing students of Kashan University of Medical Sciences. All students were included in the study using convenience sampling method and were randomly assigned to two groups: one receiving conventional training methods and the other participating in scenario-based simulation training. First, the scenarios were written and validated. Data were collected using a demographic questionnaire, a knowledge assessment questionnaire, and a performance assessment checklist. The validity and reliability of the instruments were investigated and confirmed. After data collection, the analysis was conducted using SPSS 16 software, with a significance level of 0.05.ResultsAccording to the study results, a significant increase was observed in the performance scores of nursing students in patient care after coronary artery surgery after receiving the training (P < 0.05). However, the change in the knowledge score was not significantly different between the two groups (P < 0.05). The results of the independent T-test showed no significant difference in the change in knowledge score between the two groups. But the increase in performance score between the two groups was significant (P < 0.05). The results of the ANCOVA test showed that the type of training had no effect on the change in knowledge scores (P = 0.301) but significantly affected the change in performance scores (P = 0.036).ConclusionGiven the effectiveness of scenario-based simulation training, it is recommended that this method be used in the clinical education of nursing students and nurses.

Advancing education in interventional psychiatry: scoping review of simulation training and the future of virtual reality-based learning

ObjectivesInterventional psychiatric procedures such as electroconvulsive therapy (ECT) and repetitive transcranial magnetic stimulation (rTMS) have become increasingly important therapeutic options for managing severe or treatment-resistant mental illnesses. However, research suggests that gaps in training students in these techniques represent a rate-limiting step for their further dissemination and accessibility for the public. Studies have shown that the majority of psychiatry residents lack necessary competency and self-confidence in performing these treatments. Simulation based training has served as a gold standard for training procedural skills in medicine. Simulation-based training environments, particularly immersive reality technology (e.g., virtual reality [VR]), represent a promising novel avenue for trainees to develop the necessary skills for delivering these treatments. This scoping review discusses the current training in interventional psychiatry and how simulation-based training, specifically VR, can improve pedagogy in this area.MethodsIn this scoping review, a literature search was conducted on the PubMed database using specific search terms such as “simulat*”, “training”, “ECT”, “TMS”, “neuromodulation”, and “interventional psychiatry”. The search was limited to studies with language in English from 1980 to 2023.ResultsThe initial search yielded 2094 articles, of which 4 evaluated the effectiveness of simulation approaches for ECT and were included in this review. No published studies were identified regarding VR-based education in ECT or rTMS.ConclusionsThis scoping review provides an overview of the current landscape of pedagogical methods in interventional psychiatry and highlights the identified gaps in both the existing literature and the potential application of simulation-based environments, including VR, within this field. Considering the ongoing shift in medical education towards competency-based training, this review discusses the needs and benefits of VR-based simulators as an avenue to enhance competency in interventional psychiatry. Leveraging existing experience in the use of VR-based simulators in procedural skill acquisition in surgery and anesthesia, as well as recommendations on how to translate this approach to clinical training in psychiatry, are also discussed.

Evaluating the Educational Impact of Video Tutorials on Coproparasitological Diagnostic Techniques in Veterinary Parasitology: A Cross-Sectional Study

(1) Background: Coproparasitological techniques are fundamental in veterinary medicine for diagnosing intestinal parasitic infections and form a core part of clinical training. Due to their procedural nature, teaching these techniques can benefit from scalable, visual tools that support skill acquisition and self-directed learning. This study aimed to evaluate the impact of instructional videos on students’ understanding and perceptions of coproparasitological methods. (2) Methods: A cross-sectional study was conducted with 110 veterinary students who viewed instructional videos covering 11 coproparasitological techniques. Their knowledge was assessed using a 17-item multiple-choice exam. Additionally, a structured opinion questionnaire was used to gather student feedback on the clarity and usefulness of the videos. (3) Results: Fourteen of the seventeen exam items were answered correctly by more than 80% of participants, with one item reaching 96.4% accuracy. Regarding perceptions, 94% of students rated the videos as “very clear”, and 94% as “very useful”, highlighting strong acceptance and satisfaction. (4) Conclusions: Instructional videos significantly supported students’ comprehension and were perceived as effective learning tools. Their integration into veterinary parasitology curricula is recommended to reinforce technical training, improve learning outcomes, and address limitations in access to hands-on instruction, especially in resource-constrained educational settings. Overall, instructional videos represent a valuable strategy to strengthen practical competencies in veterinary parasitology education.

EXPLORING THE CLINICAL COMPETENCES REGARDING INFECTION CONTROL MEASURES AMONG NURSING STUDENTS AT PEOPLE'S NURSING SCHOOL JAMSHORO

Background: Infection is one of the most important problems in health care services worldwide; also, it constitutes one of the most important causes of morbidity and mortality associated with clinical, diagnostic and therapeutic procedures. Nursing students are more exposed to infections during their clinical training, so they need to improve their performance related to infection control measures. Standard precautions are set of measures formulated to prevent transmission of blood borne pathogens when providing health care. Objective: The objective of this study was to assess nursing students' knowledge regarding infection control measures and to identify factors affecting adherence to infection control guidelines. Methodology: A cross-sectional study was conducted among 60 graduates at People’s Nursing School Liaquat University Jamshoro. Data were collected using a structured questionnaire consisting of 18 multiple-choice questions related to Infection and Infection Control measures. Data were analyzed using SPSS version 26. Descriptive statistics, including percentages, were used to summarize the data. Results: The study results show 60 participants, with the majority being male 56.67% (n=34), were male, aged between 22–27 years and are in 4thYear of BS Nursing Generic. Out of 60 respondents 34 participants (56.7%) fell within the Good category, 15 participants (25.0%) demonstrated a Moderate level of knowledge and 11 participants (18.3%) were classified in the Poor category. However, notable gaps existed in understanding Infection and Infection Control Measure and recognizing unhygienic practices as a key contributing factor. Conclusion: This study indicates that nursing students generally demonstrate a Good level of competency in infection control practices. However, certain knowledge gaps remain, particularly in areas such as the reporting of needlestick injuries and adherence to consistent hand hygiene protocols. These shortcomings suggest a need for further emphasis on both theoretical understanding and practical application of infection control measures. Addressing these issues is essential to ensure that students are not only knowledgeable about infection control guidelines but are also capable of implementing them effectively in clinical settings.

Exploring user profiles and preferences for mobile apps promoting active lifestyles during pregnancy and postpartum: cross-sectional study

BackgroundThe proliferation of mobile health applications (apps) presents promising opportunities for promoting maternal-fetal health outcomes. While numerous pregnancy-related apps exist, their alignment with user needs and evidence-based recommendations remains understudied. This study aimed to analyze the usage profile, characteristics, and preferences regarding mobile apps for promoting a healthy lifestyle during pregnancy and postpartum, with particular emphasis on physical activity.MethodsA cross-sectional retrospective study was conducted using a questionnaire titled “Active Pregnancy App– Promoting an active and healthy lifestyle during pregnancy and postpartum,” consisting of 36 questions in digital format via the Google Forms platform. The participants included 235 pregnant or postpartum women, with a mean age of 36 ± 4.67 years, with babies born between 2021 and 2023.ResultsAmong the women surveyed, 80% engaged in physical activity during pregnancy (39% 1–2 times/week), and 63% in the postpartum period. Walking was the predominant activity (90% of health/wellness activities). Most participants (87%) had never used pregnancy-specific fitness apps, despite 53% using general fitness apps. The majority considered the existence of a specific application for physical activity during pregnancy and postpartum to be important or very important. The main preferences regarding the app were: access to recommendations on an active and healthy lifestyle during pregnancy and postpartum; direct interaction with health and exercise professionals; the ability to record health and clinical parameters, physical activity, and training logs; access to guidelines on postpartum preparation and recovery programs; and workouts to perform at home and outdoors.ConclusionsUser perspective is an actual trend for focusing on end users’ point of view and preferences, as they are the people for whom the software is designed. The results reinforce and highlight the relevance of building a specific app for physical activity and other lifestyle parameters during pregnancy and postpartum that includes reliable and updated information, allows interaction with health and exercise professionals for monitoring, and enables the recording of progress. These results will be used in the creation and development of the “Active Pregnancy App” which can thus better meet the needs and demands of pregnant and postpartum women.

Motivational interviewing to facilitate goal setting in rehabilitation: a feasibility study

Purpose To investigate the feasibility of using embedded motivational interviewing (MI) to develop patient-centred goals in rehabilitation. Method Sixty adults (mean age 68 years, 60% female) referred with any health condition for community rehabilitation and four MI trained clinicians participated to inform feasibility of embedding motivational interviewing in goal setting to facilitate patient-centred discussions. Feasibility domains of acceptability, demand, implementation (including MI fidelity), practicality and limited efficacy were evaluated. Results Over the 14-month recruitment period, 70 patients were eligible and 60 agreed to participate (86% uptake). Patient participants reported high levels of acceptance (median 10/10, IQR 9 to 10) and identified a median of 2 (IQR 2 to 4) patient-centred goals, of which 69% were achieved at discharge. MI goal setting took a median of 20 mins (IQR 17 to 24) and most commonly occurred during the second rehabilitation session (n = 28, 47%). There were no adverse events and no instances where goal setting was incomplete. Clinicians proficiently integrated MI into clinical practice and supported the application of MI within routine rehabilitation goal setting. Conclusion Integrating motivational interviewing into rehabilitation goal setting was a feasible way to elicit patient-centred goals, which were accepted by patients and rehabilitation clinicians.

Perceptions and Challenges Faced by Dental Students During Their Transition from Pre-Clinical to Clinical Training in the Conservative Department: A Cross-Sectional Study

A high level of stress among dentistry students is attributable to the fact that they need to obtain multiple abilities, such as theoretical knowledge, practical competencies, and interpersonal skills. The purpose of this cross-sectional study was to examine the perspectives and obstacles that fourth-year dental students have as they shift to clinical practice in conservative dentistry. Data was collected during the academic year of November 2024- February 2025 among fourth-year dental students at Tripoli University, Libya, and responses were automatically recorded using a Google Forms questionnaire. The results showed that most students were anxious about starting their first case (74%). This anxiety was more evident in the conservative treatment cases (21.7%), while most preferred to work with patients in the middle age group (20 to 40) (74%) to reduce anxiety and preferred to work with first-degree relatives. The primary concern expressed by students was a fear of damaging the patient's tooth (38.3%). There were also major problems with posture (33.2%) and handpiece use (29.8%). 47.2% of students expressed fear of unnecessary root canal treatment, while 43.8% were concerned about short-term filling failures. To reduce worry, 37% of students proposed the continual presence of supervisors, while 23.4% suggested higher work grades. Other suggested solutions included extending the preclinical phase (13.6%) and increasing case demonstrations (17.4%). The students were nervous about starting their first clinical cases, particularly the conservative ones. Most students appeared to be wary of working with children and the elderly, preferring to work with first-degree relatives. Their fear in the conservative area was that they might cause further damage to the tooth, such as requiring an unneeded root canal treatment, fracturing a filling, and needing to replace it, or exacerbating the patient's exhausting pain. They also mentioned concerns about using the handpiece or not being able to see clearly due to their location. Initiatives such as enhanced supervisory support and prolonged preclinical training can assist with this transition.

Self-Care for Psychosocial and Palliative Care Clinicians: Stakeholder-Informed Recommendations for Medical Education and Clinical Training.

BackgroundDespite overwhelming evidence for work-related stress and burnout, health care clinicians receive little training in self-care.ObjectivesWe explored training and current self-care satisfaction of psychosocial and palliative care clinicians.DesignForty-one psychosocial and palliative care clinicians (18 physicians, 16 social workers, and 7 others [nurse practitioners, psychologists, pharmacists, and physician assistants]) who care for adult oncology patients at a large U.S. academic cancer center, completed an online survey about well-being, including their prior training, current satisfaction, and barriers to self-care.ResultsThis cross-sectional mixed-methods study found that clinicians felt that their graduate training did not prepare them very well to look after themselves in their professional roles (m = 1.71 [SD = 1.25]), where zero corresponded to "not well at all" and 4 to "extremely well." Open-ended responses highlighted potential gaps in self-care training: (1) Institutional support; (2) Information and education; (3) Self-care techniques and support; (4) Expectations; and (5) Managing boundaries. Clinicians rated their satisfaction with their current self-care practices as "moderately satisfied" (m = 2.10 [SD = 0.92]). Participants also noted barriers to self-care: (1) Time; (2) Competing demands and priorities between work and home; (3) Work culture, including pace and load; (4) Energy, motivation, and awareness; and (5) New methods and tools.ConclusionsThe findings highlight gaps in clinical education and training about self-care practices for health care clinicians, especially for those who care for seriously ill and dying patients. We discuss training implications and propose possible interventions, to strengthen the existing models of self-care for health care clinicians.

Impact of insurance type on outpatient mental health treatment of US adults

The mental health treatment gap in the US continues to be a major public health challenge. Even individuals with health insurance face substantial barriers to care, including high costs, insufficient coverage and inaccurate provider directories. Policies to address the treatment gap require updated population-based information about whether treatment rates vary by type of insurance. The current study aimed to compare past-year outpatient mental health treatment across insurance types (private, Medicare, Medicaid, other, or none), in the household sample of non-elderly adults in the Mental and Substance Use Disorder Prevalence Study (MDPS), (n = 4,640). MDPS, fielded October 2020 through October 2022, identified 12-month prevalence of mental disorders and rates of treatment among US adults from interviews by trained clinicians using the Structured Clinical Interview for DSM-5. Logistic regressions estimated odds of treatment among participants with a past-year MDPS diagnosis across insurance types, after adjusting for age, sex, race/ethnicity, income level, diagnosis, and functional impairment. Analyses were weighted to reflect the US adult population. 60.2% of the 1,833 participants with an MDPS mental disorder received outpatient treatment in the past year. Compared to participants with private insurance, those with no insurance had lower odds of outpatient treatment (AOR = 0.37 [0.16-0.87]). Participants with Medicare had higher odds of treatment (AOR = 4.25 [1.56-11.64]), suggesting that individuals with complex and disabling illness were least likely to have treatment disruptions during the early phases of the pandemic. Differences between groups decreased as the pandemic progressed, but utilization of services only significantly increased among individuals with private insurance. Persisting mental health treatment gaps in the US vary by type of health insurance, which warrants extensive policy reforms.

Understanding Complex Care ThroughNarrative Medicine: AQualitative Study.

Current training models do not adequately prepare pediatricians to care for children with medical complexity (CMC) as part of a team. Narrative medicine may foster mutual understanding between clinicians and families with colearning as a foundation for collaborative care. In this study, we implemented family-led interprofessional narrative medicine training and explored participants' perspectives in complex care. We performed a qualitative study of narrative medicine training grounded in entrustable professional activities (EPAs) in complex care. We recruited clinicians and families who care for CMC to participate in a series of 6 workshops. Each workshop involved close reading of a text and discussion related to a clinical activity in complex care. Participants defined the EPA before and after each workshop. We performed thematic analysis of EPA definitions and workshop transcripts and synthesized findings into a conceptual model. We recruited 3 interprofessional cohorts (35 total participants) with a mean participation rate of 80%. Each cohort included at least 2 family partners. We analyzed EPA definitions and transcripts from 18 workshops across all cohorts. Four themes emerged involving shifts in perspective about complex care: fragmented to holistic care, intractable to navigable complexity, transactional to relational connection, and caring in isolation to caring in community. Findings aligned with elements of patient- and family-centered care. Clinician and family participants in narrative medicine training described changes in their perspectives on key clinical activities in complex care. Further research should explore colearning models for pediatricians and families that foster team-based patient- and family-centered care.

Rapid cycle deliberate practice simulation with standardized prebriefing and video based formative feedback in advanced cardiac life support

Rapid-cycle deliberate practice (RCDP) is a training method involving brief, repetitive practice cycles with immediate feedback. This study examined the effectiveness of RCDP simulations in Advanced Cardiac Life Support (ACLS) training compared to traditional methods. A nonequivalent control group design was applied. Nursing students were divided into an experimental group (n = 30), which received Ludlow’s standardized three-stage prebriefing, video-based formative feedback, repeated ACLS simulations, and debriefing, and a control group (n = 29) that had traditional ACLS simulation. Data collected from April 29 to June 5, 2024, were analyzed using SPSS/WIN 25.0. The results showed significant improvements in ACLS self-efficacy, knowledge, performance skills, and team communication in the experimental group compared to the control group. The findings suggest that RCDP simulation training, with its focus on repetition and immediate feedback, enhances practical skills and confidence in emergency resuscitation scenarios. This approach may be more effective than traditional simulation methods in medical education, particularly in improving clinical performance and patient outcomes. Beyond medical education, RCDP holds potential for clinical training by enhancing proficiency in high-stakes situations such as cardiac resuscitation. Continuous skill refinement through immediate correction in fast-paced environments can improve individual performance and team dynamics, ultimately leading to better patient outcomes.

The role of artificial intelligence in enhancing personalized learning pathways and clinical training in dental education

The role of artificial intelligence in enhancing personalized learning pathways and clinical training in dental education

Deep learning-based evaluation of the severity of mitral regurgitation in canine myxomatous mitral valve disease patients using digital stethoscope recordings

BackgroundMyxomatous mitral valve disease (MMVD) represents the most prevalent cardiac disorder in dogs, frequently resulting in mitral regurgitation (MR) and congestive heart failure. Although echocardiography is the gold standard for diagnosis, it is an expensive tool that involves significant clinical training to ensure consistent application. Deep learning models offer an innovative approach to assessing MR using digital stethoscopic recordings, enabling early screening and precise prediction. Thus, in this study, we evaluated the effectiveness of a convolutional neural network 6 (CNN6) in providing an objective alternative to traditional methods for assessing MR. This study, conducted at the Seoul National University Veterinary Medicine Teaching Hospital, included 460 dogs with MMVD, classified according to the American College of Veterinary Internal Medicine guidelines. Phonocardiogram signals were recorded using digital stethoscopes and analyzed using the deep models CNN6, patch-mix audio spectrogram transformer (PaSST), and residual neural network (ResNET38), which were trained to categorize MR severity into mild, moderate, and severe based on MINE score. Performance metrics were calculated to evaluate model effectiveness.ResultsThe CNN6-Fbank model achieved an accuracy of 94.12% [95% confidence interval (CI): 94.11–93.12], specificity of 97.30% (95% CI: 97.30–97.34), sensitivity of 94.12% (95% CI: 93.74–94.50), precision of 92.63% (95% CI: 92.29–92.97), and F1 score of 93.32% (95% CI: 93.05–93.59), outperforming the PaSST and ResNet38 models overall and demonstrating robust performance across most metrics.ConclusionsDeep learning models, particularly CNN6, can effectively assess MR severity in dogs with MMVD using digital stethoscope recordings. This approach provides a rapid, noninvasive, and reliable adjunct to echocardiography, potentially enhancing diagnosis and outcomes. Future studies should focus on broader clinical validation and real-time application of this technology.

Machine learning of blood haemoglobin and haematocrit levels via smartphone conjunctiva photography in Kenyan pregnant women: a clinical study protocol.

Anaemia during pregnancy is a widespread health burden globally, especially in low- and middle-income countries, posing a serious risk to both maternal and neonatal health. The primary challenge is that anaemia is frequently undetected or is detected too late, worsening pregnancy complications. The gold standard for diagnosing anaemia is a clinical laboratory blood haemoglobin (Hgb) or haematocrit (Hct) test involving a venous blood draw. However, this approach presents several challenges in resource-limited settings regarding accessibility and feasibility. Although non-invasive blood Hgb testing technologies are gaining attention, they remain limited in availability, affordability and practicality. This study aims to develop and validate a mobile health (mHealth) machine learning model to reliably predict blood Hgb and Hct levels in Black African pregnant women using smartphone photos of the conjunctiva. This is a single-centre, cross-sectional and observational study, leveraging existing antenatal care services for pregnant women aged 15 to 49 years in Kenya. The study involves collecting smartphone photos of the conjunctiva alongside conventional blood Hgb tests. Relevant clinical data related to each participant's anaemia status will also be collected. The photo acquisition protocol will incorporate diverse scenarios to reflect real-world variability. A clinical training dataset will be used to refine a machine learning model designed to predict blood Hgb and Hct levels from smartphone images of the conjunctiva. Using a separate testing dataset, comprehensive analyses will assess its performance by comparing predicted blood Hgb and Hct levels with clinical laboratory and/or finger-prick readings. This study is approved by the Moi University Institutional Research and Ethics Committee (Reference: IREC/585/2023 and Approval Number: 004514), Kenya's National Commission for Science, Technology, and Innovation (NACOSTI Reference: 491921) and Purdue University's Institutional Review Board (Protocol Number: IRB-2023-1235). Participants will include emancipated or mature minors. In Kenya, pregnant women aged 15 to 18 years are recognised as emancipated or mature minors, allowing them to provide informed consent independently. The study poses minimal risk to participants. Findings and results will be disseminated through submissions to peer-reviewed journals and presentations at the participating institutions, including Moi Teaching and Referral Hospital and Kenya's Ministry of Health. On completion of data collection and modelling, this study will demonstrate how machine learning-driven mHealth technologies can reduce reliance on clinical laboratories and complex equipment, offering accessible and scalable solutions for resource-limited and at-home settings.

Comparative Analysis of Machine Learning Approaches for Fetal Movement Detection with Linear Acceleration and Angular Rate Signals.

Reduced fetal movement (RFM) can indicate that a fetus is at risk, but current monitoring methods provide only a "snapshot in time" of fetal health and require trained clinicians in clinical settings. To improve antenatal care, there is a need for continuous, objective fetal movement monitoring systems. Wearable sensors, like inertial measurement units (IMUs), offer a promising data-driven solution, but distinguishing fetal movements from maternal movements remains challenging. The potential benefits of using linear acceleration and angular rate data for fetal movement detection have not been fully explored. In this study, machine learning models were developed using linear acceleration and angular rate data from twenty-three participants who wore four abdominal IMUs and one chest reference while indicating perceived fetal movements with a handheld button. Random forest (RF), bi-directional long short-term memory (BiLSTM), and convolutional neural network (CNN) models were trained using hand-engineered features, time series data, and time-frequency spectrograms, respectively. The results showed that combining accelerometer and gyroscope data improved detection performance across all models compared to either one alone. CNN consistently outperformed other models but required larger datasets. RF and BiLSTM, while more sensitive to signal noise, offered reasonable performance with smaller datasets and greater interpretability.

Dutch post-graduate training in Global Health and Tropical Medicine: a qualitative study on graduates’ perspectives

IntroductionThe Dutch Medical Doctor-Global Health (MD-GH) prepares to work in low-resource settings (LRS) by completing a hybrid postgraduate training program of 2 years and 9 months, with clinical and public health exposure in the Netherlands and a Global Health residency in LRS. The objectives of the program include acquiring clinical skills to work as a physician in a setting with different (often more severe) pathology and limited resources. In public health teaching, emphasis is given, among other, to adapting to a culturally different environment. After graduation, MD-GH work in a wide variety of countries and settings for variable time. As part of a curriculum review, this study examines MD-GHs' perception of the quality of the training program and provides recommendations for improvement.MethodsA qualitative study was performed. Thematic analysis was applied to semi-structured interviews with 23 MD-GH who graduated between 2017 and 2021.ResultsMD-GHs predominantly worked as clinicians; several were (also) involved in management or capacity building. The clinical training program adequately addressed general skills, but did not sufficiently prepare for locally encountered, often severe, pathology. During the training, adequate supervision with clear learning goals was found pivotal to a positive learning experience. Gaps included clinical training in Internal Medicine (particularly infectious diseases and non-communicable diseases) and Paediatrics. Public Health teaching as well as cultural awareness should be intensified and introduced earlier in the program. The Global Health residency was considered important, but tasks and learning outcomes varied. Teaching, supervision, and capacity building were considered increasingly important key elements of working in LRS. Consensus favoured the current duration of the training program without extension.DiscussionWhile the generalist nature of the MD-GH training was appreciated, the program would benefit from additional clinical training in infectious diseases, non-communicable diseases, and Paediatrics. Moving forward, emphasis should be placed on structured mentorship, enhanced public health teaching, and standardized residency programs with clearly delineated objectives to better equip MD-GH professionals for their multifaceted roles in LRS. Moreover, future revisions of the training program should incorporate the perspectives of host institutes in LRS and tailor the training needs.