RESUMO
Background: As artificial intelligence (AI) continues to advance with breakthroughs in natural language processing (NLP) and machine learning (ML), such as the development of models like OpenAI's ChatGPT, new opportunities are emerging for efficient curation of electronic health records (EHR) into real-world data (RWD) for evidence generation in oncology. Our objective is to describe the research and development of industry methods to promote transparency and explainability. Methods: We applied NLP with ML techniques to train, validate, and test the extraction of information from unstructured documents (e.g., clinician notes, radiology reports, lab reports, etc.) to output a set of structured variables required for RWD analysis. This research used a nationwide electronic health record (EHR)-derived database. Models were selected based on performance. Variables curated with an approach using ML extraction are those where the value is determined solely based on an ML model (i.e. not confirmed by abstraction), which identifies key information from visit notes and documents. These models do not predict future events or infer missing information. Results: We developed an approach using NLP and ML for extraction of clinically meaningful information from unstructured EHR documents and found high performance of output variables compared with variables curated by manually abstracted data. These extraction methods resulted in research-ready variables including initial cancer diagnosis with date, advanced/metastatic diagnosis with date, disease stage, histology, smoking status, surgery status with date, biomarker test results with dates, and oral treatments with dates. Conclusion: NLP and ML enable the extraction of retrospective clinical data in EHR with speed and scalability to help researchers learn from the experience of every person with cancer.
RESUMO
EDIFY (Eating Disorders: Delineating Illness and Recovery Trajectories to Inform Personalised Prevention and Early Intervention in Young People) is an ambitious research project aiming to revolutionise how eating disorders are perceived, prevented and treated. Six integrated workstreams will address key questions, including: What are young people's experiences of eating disorders and recovery? What are the unique and shared risk factors in different groups? What helps or hinders recovery? How do the brain and behaviour change from early- to later-stage illness? How can we intervene earlier, quicker and in a more personalised way? This 4-year project, involving over 1000 participants, integrates arts, design and humanities with advanced neurobiological, psychosocial and bioinformatics approaches. Young people with lived experience of eating disorders are at the heart of EDIFY, serving as advisors and co-producers throughout. Ultimately, this work will expand public and professional perceptions of eating disorders, uplift under-represented voices and stimulate much-needed advances in policy and practice.
RESUMO
Distances are commonly underperceived in virtual reality (VR), and this finding has been documented repeatedly over more than two decades of research. Yet, there is evidence that perceived distance is more accurate in modern compared to older head-mounted displays (HMDs). This meta-analysis, based on 137 samples from 61 publications, describes egocentric distance perception across 20 HMDs and examines the relationship between perceived distance and technical HMD characteristics. Judged distance was positively associated with HMD field of view (FOV), positively associated with HMD resolution, and negatively associated with HMD weight. The effects of FOV and resolution were more pronounced among heavier HMDs. These findings suggest that future improvements in these technical characteristics may be central to resolving the problem of distance underperception in VR.
Assuntos
Óculos Inteligentes , Realidade Virtual , Percepção de Distância , Gráficos por ComputadorRESUMO
Despite the wide adoption of insertable cardiac monitors (ICMs), high false-positive rates, suboptimal signal quality, limited ability to detect atrial flutter, and lack of remote programming remain challenging. The LUX-Dx PERFORM study was designed to evaluate novel technologies engineered to address these issues. Here, we present preliminary results from the trial focusing on the safety of ICM insertion, remote monitoring rates, and the feasibility of remote programming. LUX-Dx PERFORM is a multicenter, prospective, single-arm, post-market, observational study with planned enrollment of up to 827 patients from 35 sites in North America. A preliminary cohort consisting of the first 369 patients who were enrolled between March and October 2021 was selected for analysis. Three hundred sixty-three (363) patients had ICM insertions across inpatient and outpatient settings. The mean time followed was 103.4 ± 61.8 days per patient. The total infection rate was 0.8% (3/363). Interim results show high levels of remote monitoring with a median 94% of days with data transmission (interquartile range: 82-99). Thirteen (13) in-clinic and 24 remote programming sessions were reported in 34 subjects. Reprogramming examples are presented to highlight signal quality, the ability to detect atrial flutter, and the positive impact of remote programming on patient management. Interim results from LUX-Dx PERFORM study demonstrate the safety of insertion, high data transmission rates, the ability to detect atrial flutter, and the feasibility of remote programming to optimize arrhythmia detection and improve clinical workflow. Future results from LUX-Dx PERFORM will further characterize improvements in signal quality and arrhythmia detection.
Assuntos
Fibrilação Atrial , Flutter Atrial , Humanos , Fibrilação Atrial/diagnóstico , Eletrocardiografia Ambulatorial/métodos , Estudos Prospectivos , Pacientes AmbulatoriaisRESUMO
Virtual reality (VR) allows users to walk to explore the virtual environment (VE), but this capability is constrained by real obstacles. Teleporting interfaces overcome this constraint by allowing users to select a position, and sometimes orientation, in the VE before being instantly transported without self-motion cues. This study investigated whether individual differences in navigation performance when teleporting correspond to characteristics of the individual, including spatial ability. Participants performed triangle completion (traverse two outbound path legs, then point to the path origin) within VEs differing in visual landmarks. Locomotion was accomplished using three interfaces: walking, partially concordant teleporting (teleport to change position, rotate the body to change orientation), and discordant teleporting (teleport to change position and orientation). A latent profile analysis identified three profiles of individuals: those who performed well overall and improved with landmarks, those who performed poorly without landmarks but improved when available, and those who performed poorly even with landmarks. Characteristics of individuals differed across profiles, including gender, self-reported spatial ability, mental rotation, and perspective-taking; but only perspective-taking significantly distinguished all three profiles. This work elucidates spatial cognitive correlates of navigation and provides a framework for identifying susceptibility to disorientation in VR. (PsycInfo Database Record (c) 2023 APA, all rights reserved).
Assuntos
Navegação Espacial , Realidade Virtual , Humanos , Individualidade , Sinais (Psicologia) , Caminhada , Percepção EspacialRESUMO
Ingestible robotic capsules with locomotion capabilities and on-board sampling mechanism have great potential for non-invasive diagnostic and interventional use in the gastrointestinal tract. Real-time tracking of capsule location and operational state is necessary for clinical application, yet remains a significant challenge. To this end, we propose an approach that can simultaneously determine the mechanism state and in-plane 2D pose of millimeter capsule robots in an anatomically representative environment using ultrasound imaging. Our work proposes an attention-based hierarchical deep learning approach and adapts the success of transfer learning towards solving the multi-task tracking problem with limited dataset. To train the neural networks, we generate a representative dataset of a robotic capsule within ex-vivo porcine stomachs. Experimental results show that the accuracy of capsule state classification is 97%, and the mean estimation errors for orientation and centroid position are 2.0 degrees and 0.24 mm (1.7% of the capsule's body length) on the hold-out test set. Accurate detection of the capsule while manipulated by an external magnet in a porcine stomach and colon is also demonstrated. The results suggest our proposed method has the potential for advancing the wireless capsule-based technologies by providing accurate detection of capsule robots in clinical scenarios.
Assuntos
Aprendizado Profundo , Robótica , Animais , Suínos , UltrassonografiaRESUMO
The wide availability of consumer-oriented virtual reality (VR) equipment has enabled researchers to recruit existing VR owners to participate remotely using their own equipment. Yet, there are many differences between lab environments and home environments, as well as differences between participant samples recruited for lab studies and remote studies. This paper replicates a lab-based experiment on VR locomotion interfaces using a remote sample. Participants completed a triangle-completion task (travel two path legs, then point to the path origin) using their own VR equipment in a remote, unsupervised setting. Locomotion was accomplished using two versions of the teleporting interface varying in availability of rotational self-motion cues. The size of the traveled path and the size of the surrounding virtual environment were also manipulated. Results from remote participants largely mirrored lab results, with overall better performance when rotational self-motion cues were available. Some differences also occurred, including a tendency for remote participants to rely less on nearby landmarks, perhaps due to increased competence with using the teleporting interface to update self-location. This replication study provides insight for VR researchers on aspects of lab studies that may or may not replicate remotely.
Assuntos
Gráficos por Computador , Realidade Virtual , Sinais (Psicologia) , Humanos , Locomoção , Movimento (Física)RESUMO
AIMS AND OBJECTIVES: In order to retain and develop pre-doctoral and doctoral staff, we sought to establish and implement a model for supporting postdoctoral clinical academic careers at our hospital. BACKGROUND: Doctorally-prepared nurses can raise clinical practice standards, but there is a deficit of career opportunities and post-doctoral positions. This will inevitably impact delivering the evidence for improving patient outcomes and service delivery. It is therefore imperative that post-doctoral nursing and allied professional post-doctoral clinical academic career opportunities are developed and embedded in practice. DESIGN: We describe the development and implementation of a model to provide and embed post-doctoral clinical academic roles at our hospital. METHODS: A multidisciplinary group devised the model which was approved by the Hospital Executive Board. RESULTS: The model includes having a bespoke plan for each individual, planning the post-doctoral role at time of PhD planning, providing 1 day a week protected research time for a minimum of 12 months after PhD completion, having an appropriate honorary academic contract at a partner University, and providing integrated clinical and academic supervision throughout. Other key components include senior clinical and Executive Board support and an existing vibrant research culture at our organisation. CONCLUSIONS: The St Bartholomew's Hospital model, focusing on individualised posts integrating clinical and academic roles to address service needs, offers a novel approach to supporting post-doctoral clinical academic careers in a clinical setting. RELEVANCE TO CLINICAL PRACTICE: Since doctorally-prepared nurses can raise clinical practice standards we developed a sustainable infrastructure to increase our postdoctoral nursing and allied professional workforce. This model could provide a framework for other hospitals to embed postdoctoral clinical academic careers in practice.
Assuntos
Educação de Pós-Graduação em Enfermagem , Médicos , Hospitais , Humanos , UniversidadesRESUMO
BACKGROUND: Evidence suggests specialist eating disorders services for children and adolescents with anorexia nervosa have the potential to improve outcomes and reduce costs through reduced hospital admissions. This study aimed to evaluate the cost-effectiveness of assessment and diagnosis in community-based specialist child and adolescent mental health services (CAMHS) compared to generic CAMHS for children and adolescents with anorexia nervosa. METHOD: Observational, surveillance study of children and adolescents aged 8 to 17, in contact with community-based CAMHS in the UK or Republic of Ireland for a first episode of anorexia nervosa. Data were reported by clinicians at baseline, 6 and 12-months follow-up. Outcomes included the Children's Global Assessment Scale (CGAS) and percentage of median expected body mass for age and sex (%mBMI). Service use data included paediatric and psychiatric inpatient admissions, outpatient and day-patient attendances. A joint distribution of incremental mean costs and effects for each group was generated using bootstrapping to explore the probability that each service is the optimal choice, subject to a range of values a decision-maker might be willing to pay for outcome improvements. Uncertainty was explored using cost-effectiveness acceptability curves. RESULTS: Two hundred ninety-eight children and adolescents met inclusion criteria. At 12-month follow-up, there were no significant differences in total costs or outcomes between specialist eating disorders services and generic CAMHS. However, adjustment for pre-specified baseline covariates resulted in observed differences favouring specialist services, due to significantly poorer clinical status of the specialist group at baseline. Cost-effectiveness analysis using CGAS suggests that the probability of assessment in a specialist service being cost-effective compared to generic CAMHS ranges from 90 to 50%, dependent on willingness to pay for improvements in outcome. CONCLUSIONS: Assessment in a specialist eating disorders service for children and adolescents with anorexia nervosa may have a higher probability of being cost-effective than assessment in generic CAMHS. TRIAL REGISTRATION: ISRCTN12676087 . Date of registration 07/01/2014.
Specialist eating disorders services may improve outcomes and reduce hospitalisations for children and adolescents with anorexia nervosa. Reductions in hospitalisation could save money for the NHS and are better for young people because hospitalisation disrupts their home life, social life and education. This study evaluated outcomes and costs of specialist eating disorders services compared to general child and adolescent mental health services (CAMHS) for children and adolescents with anorexia nervosa.Children and adolescents were identified by contacting child and adolescent psychiatrists in the UK and Ireland and asking them to report any new cases of anorexia nervosa. These psychiatrists identified 298 young people aged 8 to 17 with an anorexia nervosa diagnosis for the first time. The psychiatrists provided information on the health services these young people used and how they were doing when they were first diagnosed and 6 months and 1 year later.Children and adolescents in specialist services were more severely ill than those in CAMHS when they were first diagnosed. Despite this, care for the young people in specialist services cost about the same as for those diagnosed in CAMHS, and their outcomes after 1 year were similar. This work showed that specialist services may be better value for money than CAMHS.
RESUMO
BACKGROUND: Knowledge of Medicare reimbursement is essential for plastic surgeons providing care to Medicare beneficiaries. The authors sought to evaluate changes in Medicare reimbursement for common plastic surgery procedures from 2010 to 2020. METHODS: The authors assessed the Physician Fee Schedule of the Centers for Medicare and Medicaid Services website. Rates of work-, facility-, or malpractice-related relative value units and total monetary units for 26 common plastic surgery procedures between 2010 and 2020 were evaluated. Descriptive statistics were used to calculate relative differences and to compare observed changes over time with the rate of inflation. RESULTS: For the selected procedures, the authors found an average relative difference in terms of monetary units of an increase by 2.02 percent. However, after adjusting for inflation, the average relative difference was a decrease by 14.31 percent. The authors' analysis indicates that, on average, there was a 1.55 percent decrease in physician relative value units between 2010 and 2020. CONCLUSIONS: Medicare reimbursement rates have changed significantly over the past decade. However, these changes did not keep pace with the rate of inflation. Plastic surgeons should be aware of these trends and advocate for more fair reimbursement rates.
Assuntos
Reembolso de Seguro de Saúde/tendências , Medicare , Procedimentos de Cirurgia Plástica/economia , Humanos , Estados UnidosRESUMO
Research about vection (illusory self-motion) has investigated a wide range of sensory cues and employed various methods and equipment, including use of virtual reality (VR). However, there is currently no research in the field of vection on the impact of floating in water while experiencing VR. Aquatic immersion presents a new and interesting method to potentially enhance vection by reducing conflicting sensory information that is usually experienced when standing or sitting on a stable surface. This study compares vection, visually induced motion sickness, and presence among participants experiencing VR while standing on the ground or floating in water. Results show that vection was significantly enhanced for the participants in the Water condition, whose judgments of self-displacement were larger than those of participants in the Ground condition. No differences in visually induced motion sickness or presence were found between conditions. We discuss the implication of this new type of VR experience for the fields of VR and vection while also discussing future research questions that emerge from our findings.
RESUMO
In this study we seek to determine the efficacy of using automated mapping methods to reduce the manual mapping burden of laboratory data to LOINC(r) on a nationwide electronic health record derived oncology specific dataset. We developed novel encoding methodologies to vectorize free text lab data, and evaluated logistic regression, random forest, and knn machine learning classifiers. All machine learning models did significantly better than deterministic baseline algorithms. The best classifiers were random forest and were able to predict the correct LOINC code 94.5% of the time. Ensemble classifiers further increased accuracy, with the best ensemble classifier predicting the same code 80.5% of the time with an accuracy of 99%. We conclude that by using an automated laboratory mapping model we can both reduce manual mapping time, and increase quality of mappings, suggesting automated mapping is a viable tool in a real-world oncology dataset.
Assuntos
Logical Observation Identifiers Names and Codes , Aprendizado de Máquina , Algoritmos , Registros Eletrônicos de Saúde , Humanos , LaboratóriosRESUMO
OBJECTIVE: This systematic review assesses the average duration of untreated eating disorder (DUED) in help-seeking populations at the time of first eating disorder (ED) treatment and investigates the relationship between DUED and symptom severity/clinical outcomes. METHOD: PRISMA guidelines were followed throughout. Selected studies provided information on either: (i) length of DUED, (ii) components of DUED, (iii) cross-sectional associations between DUED and symptom severity, (iv) associations between DUED and clinical outcomes, or (v) experimental manipulation of DUED. Study quality was assessed. RESULTS: Fourteen studies from seven countries were included. Across studies, average DUED weighted by sample size was 29.9 months for anorexia nervosa, 53.0 months for bulimia nervosa and 67.4 months for binge eating disorder. A younger age at time of first treatment was indicative of shorter DUED. Retrospective studies suggest that a shorter DUED may be related to a greater likelihood of remission. Manipulation of DUED by shortening service-related delays may improve clinical outcomes. CONCLUSIONS: Data on length of DUED provide a benchmark for early intervention in EDs. Preliminary evidence suggests DUED may be a modifiable factor influencing outcomes in EDs. To accurately determine the role of DUED, definition and measurement must be uniformly operationalised. Highlights This systematic review is the first to examine duration of untreated eating disorder (DUED) across different eating disorders. Definitions and measurement of DUED and its components vary considerably between studies. Across different eating disorders average DUED weighted by sample size ranges from approximately two and a half years (for anorexia nervosa) to nearly 6 years (for binge eating disorder). DUED appears to be related to age such that younger patients have shorter DUED.
Assuntos
Anorexia Nervosa , Transtorno da Compulsão Alimentar , Bulimia Nervosa , Transtornos da Alimentação e da Ingestão de Alimentos , Anorexia Nervosa/diagnóstico , Anorexia Nervosa/terapia , Transtorno da Compulsão Alimentar/diagnóstico , Transtorno da Compulsão Alimentar/terapia , Bulimia Nervosa/terapia , Estudos Transversais , Transtornos da Alimentação e da Ingestão de Alimentos/terapia , Humanos , Estudos RetrospectivosRESUMO
Virtual reality systems typically allow users to physically walk and turn, but virtual environments (VEs) often exceed the available walking space. Teleporting has become a common user interface, whereby the user aims a laser pointer to indicate the desired location, and sometimes orientation, in the VE before being transported without self-motion cues. This study evaluated the influence of rotational self-motion cues on spatial updating performance when teleporting, and whether the importance of rotational cues varies across movement scale and environment scale. Participants performed a triangle completion task by teleporting along two outbound path legs before pointing to the unmarked path origin. Rotational self-motion reduced overall errors across all levels of movement scale and environment scale, though it also introduced a slight bias toward under-rotation. The importance of rotational self-motion was exaggerated when navigating large triangles and when the surrounding environment was large. Navigating a large triangle within a small VE brought participants closer to surrounding landmarks and boundaries, which led to greater reliance on piloting (landmark-based navigation) and therefore reduced-but did not eliminate-the impact of rotational self-motion cues. These results indicate that rotational self-motion cues are important when teleporting, and that navigation can be improved by enabling piloting.
RESUMO
Teleporting is a popular interface to allow virtual reality users to explore environments that are larger than the available walking space. When teleporting, the user positions a marker in the virtual environment and is instantly transported without any self-motion cues. Five experiments were designed to evaluate the spatial cognitive consequences of teleporting and to identify environmental cues that could mitigate those costs. Participants performed a triangle completion task by traversing 2 outbound path legs before pointing to the unmarked path origin. Locomotion was accomplished via walking or 2 common implementations of the teleporting interface distinguished by the concordance between movement of the body and movement through the virtual environment. In the partially concordant teleporting interface, participants teleported to translate (change position) but turned the body to rotate. In the discordant teleporting interface, participants teleported to translate and rotate. Across all 5 experiments, discordant teleporting produced larger errors than partially concordant teleporting which produced larger errors than walking, reflecting the importance of translational and rotational self-motion cues. Furthermore, geometric boundaries (room walls or a fence) were necessary to mitigate the spatial cognitive costs associated with teleporting, and landmarks were helpful only in the context of a geometric boundary. (PsycInfo Database Record (c) 2020 APA, all rights reserved).
Assuntos
Sinais (Psicologia) , Movimento , Navegação Espacial , Realidade Virtual , Adulto , Meio Ambiente , Feminino , Humanos , Masculino , Caminhada , Adulto JovemRESUMO
BACKGROUND: Perspectives of young people with eating disorders and their parents on helpful aspects of care should be incorporated into evidence-based practice and service design, but data are limited.AimsTo explore patient and parent perspectives on positive and negative aspects of care for young people with eating disorders. METHOD: Six online focus groups with 19 young people aged 16-25 years with existing or past eating disorders and 11 parents. RESULTS: Thematic analysis identified three key themes: the need to (a) shift from a weight-focused to a more holistic, individualised and consistent care approach, with a better balance in targeting psychological and physical problems from an early stage; (b) improve professionals' knowledge and attitude towards patients and their families at all levels of care from primary to 'truly specialist'; (c) enhance peer and family support. CONCLUSIONS: Young people and parents identified an array of limitations in approaches to care for young people with eating disorders and raised the need for change, particularly a move away from a primarily weight-focused treatment and a stronger emphasis on psychological needs and individualised care.Declaration of interestNone.
RESUMO
PURPOSE: Skeletal-muscle function can be evaluated using force-times curves generated via the isometric midthigh pull (IMTP). Various sampling frequencies (500-1000 Hz) have been used for IMTP assessments; however, no research has investigated the influence of sampling frequency on IMTP kinetics. Therefore, the purpose of this study was to investigate the influence of sampling frequency on kinetic variables during the IMTP, including peak force, time-specific force values (100, 150, and 200 ms), and rate of force development (RFD) at 3 time bands (0-100, 0-150, 0-200 ms). METHODS: Academy rugby league players (n = 30, age 17.5 ± 1.1 y, height 1.80 ± 0.06 m, mass 85.4 ± 10.3 kg) performed 3 IMTP trials on a force platform sampling at 2000 Hz, which was subsequently down-sampled to 1500, 1000, and 500 Hz for analysis. RESULTS: Intraclass correlation coefficients (ICC) and coefficients of variation (CV) demonstrated high within-session reliability for all force and RFD variables across all sampling frequencies (ICC ≥ .80, CV ≤ 14.4%) except RFD 0-100 and 0-150, which demonstrated slightly greater levels of variance (CV = 18.0-24.1%). Repeated-measures analysis of variance revealed no significant differences (P > .05, Cohen d ≤ 0.0171) in kinetic variables between sampling frequencies. Overall, high reliability was observed across all sampling frequencies for peak force, time-specific force, and RFD 0- to 200-ms variables, with no significant differences (P > .05) for each kinetic variable across sampling frequencies. CONCLUSIONS: Practitioners and scientists may consider sampling as low as 500 Hz when measuring peak force, time-specific force values, and RFD at predetermined time bands during the IMTP for accurate and reliable data.
RESUMO
The accuracy of machine learning tasks critically depends on high quality ground truth data. Therefore, in many cases, producing good ground truth data typically involves trained professionals; however, this can be costly in time, effort, and money. Here we explore the use of crowdsourcing to generate a large number of training data of good quality. We explore an image analysis task involving the segmentation of corn tassels from images taken in a field setting. We investigate the accuracy, speed and other quality metrics when this task is performed by students for academic credit, Amazon MTurk workers, and Master Amazon MTurk workers. We conclude that the Amazon MTurk and Master Mturk workers perform significantly better than the for-credit students, but with no significant difference between the two MTurk worker types. Furthermore, the quality of the segmentation produced by Amazon MTurk workers rivals that of an expert worker. We provide best practices to assess the quality of ground truth data, and to compare data quality produced by different sources. We conclude that properly managed crowdsourcing can be used to establish large volumes of viable ground truth data at a low cost and high quality, especially in the context of high throughput plant phenotyping. We also provide several metrics for assessing the quality of the generated datasets.
Assuntos
Produtos Agrícolas/fisiologia , Crowdsourcing/métodos , Processamento de Imagem Assistida por Computador/métodos , Aprendizado de Máquina , Algoritmos , Confiabilidade dos Dados , Abastecimento de Alimentos , Humanos , Internet , Fenótipo , Projetos PilotoRESUMO
Spatial memories are often hierarchically organized with different regions of space represented in unique clusters within the hierarchy. Each cluster is thought to be organized around its own microreference frame selected during learning, whereas relationships between clusters are organized by a macroreference frame. Two experiments were conducted in order to better understand important characteristics of macroreference frames. Participants learned overlapping spatial layouts of objects within a room-sized environment before performing a perspective-taking task from memory. Of critical importance were between-layout judgments thought to reflect the macroreference frame. The results indicate that (1) macroreference frames characterize overlapping spatial layouts, (2) macroreference frames are used even when microreference frames are aligned with one another, and (3) macroreference frame selection depends on an interaction between the global macroaxis (defined by characteristics of the layout of all learned objects), the relational macroaxis (defined by characteristics of the two layouts being related on a perspective-taking trial), and the learning view. These results refine the current understanding of macroreference frames and document their broad role in spatial memory.