270 research outputs found
Automated Essay Writing: An AIED Opinion
No full textThis opinion piece emerged from research for the book, Story Machines: How Computers Have Become Creative Writers, by Mike Sharples and Rafael Pérez y Pérez, published by Routledge. While thinking of ways to promote the book, I realised that students could employ AI story generators to write essays. That led me to research automated essay writing, write a Twitter thread that has garnered 43,000 engagements, and author a blog article (Sharples, 2022). The current piece is a revised and expanded version of that article
Multi-State Models for Panel Data: The msm Package for R
No full textPanel data are observations of a continuous-time process at arbitrary times, for example, visits to a hospital to diagnose disease status. Multi-state models for such data are generally based on the Markov assumption. This article reviews the range of Markov models and their extensions which can be fitted to panel-observed data, and their implementation in the msm package for R. Transition intensities may vary between individuals, or with piecewise-constant time-dependent covariates, giving an inhomogeneous Markov model. Hidden Markov models can be used for multi-state processes which are misclassified or observed only through a noisy marker. The package is intended to be straightforward to use, flexible and comprehensively documented. Worked examples are given of the use of msm to model chronic disease progression and screening. Assessment of model fit, and potential future developments of the software, are also discussed.
Mapping of the EQ-5D index from clinical outcome measures and demographic variables in patients with coronary heart disease
No full textThe official published version of the article can be found at the link below.Background: The EuroQoL 5D (EQ-5D) is a questionnaire that provides a measure of utility for cost-effectiveness analysis. The EQ-5D has been widely used in many patient groups, including those with coronary heart disease. Studies often require patients to complete many questionnaires and the EQ-5D may not be gathered. This study aimed to assess whether demographic and clinical outcome variables, including scores from a disease specific measure, the Seattle Angina Questionnaire (SAQ), could be used to predict, or map, the EQ-5D index value where it is not available.Methods: Patient-level data from 5 studies of cardiac interventions were used. The data were split into two groups approximately 60% of the data were used as an estimation dataset for building models, and 40% were used as a validation dataset. Forward ordinary least squares linear regression methods and measures of prediction error were used to build a model to map to the EQ-5D index. Age, sex, a proxy measure of disease stage, Canadian Cardiovascular Society (CCS) angina severity class, treadmill exercise time (ETT) and scales of the SAQ were examined.Results: The exertional capacity (ECS), disease perception (DPS) and anginal frequency scales (AFS) of the SAQ were the strongest predictors of the EQ-5D index and gave the smallest root mean square errors. A final model was chosen with age, gender, disease stage and the ECS, DPS and AFS scales of the SAQ. ETT and CCS did not improve prediction in the presence of the SAQ scales. Bland-Altman agreement between predicted and observed EQ-5D index values was reasonable for values greater than 0.4, but below this level predicted values were higher than observed. The 95% limits of agreement were wide (-0.34, 0.33).Conclusions: Mapping of the EQ-5D index in cardiac patients from demographics and commonly measured cardiac outcome variables is possible; however, prediction for values of the EQ-5D index below 0.4 was not accurate. The newly designed 5-level version of the EQ-5D with its increased ability to discriminate health states may improve prediction of EQ-5D index values.Funding for this project by EuroQoL grou
Multi-state modelling of repeated hospitalisation and death in patients with heart failure: The use of large administrative databases in clinical epidemiology.
No full textIn chronic diseases like heart failure (HF), the disease course and associated clinical event histories for the patient population vary widely. To improve understanding of the prognosis of patients and enable health care providers to assess and manage resources, we wish to jointly model disease progression, mortality and their relation with patient characteristics. We show how episodes of hospitalisation for disease-related events, obtained from administrative data, can be used as a surrogate for disease status. We propose flexible multi-state models for serial hospital admissions and death in HF patients, that are able to accommodate important features of disease progression, such as multiple ordered events and competing risks. Fully parametric and semi-parametric semi-Markov models are implemented using freely available software in R. The models were applied to a dataset from the administrative data bank of the Lombardia region in Northern Italy, which included 15,298 patients who had a first hospitalisation ending in 2006 and 4 years of follow-up thereafter. This provided estimates of the associations of age and gender with rates of hospital admission and length of stay in hospital, and estimates of the expected total time spent in hospital over five years. For example, older patients and men were readmitted more frequently, though the total time in hospital was roughly constant with age. We also discuss the relative merits of parametric and semi-parametric multi-state models, and model assessment and comparison
Contemporary roles of registries in clinical cardiology: When do we need randomized trials?
No full textClinical registries are established as tools for auditing clinical standards and benchmarking quality improvement initiatives. They also have an emerging role (as electronic health records) in cardiovascular research and, in particular, the conduct of RCTs. While the RCT is accepted as the most robust experimental design, observational data from clinical registries has become increasingly valuable for RCTs. Data from clinical registries may be used to augment results from RCTs, identify patients for recruitment and as an alternative when randomization is not practically possible or ethically desirable. Here the authors appraise the advantages and disadvantages of both methodologies, with the aim of clarifying when their joint use may be successful
Clinical and cost-effectiveness analysis of an open label, single-centre, randomised trial of spinal cord stimulation (SCS) versus percutaneous myocardial laser revascularisation (PMR) in patients with refractory angina pectoris: The SPiRiT trial
No full textBackground: Patients with refractory angina have significant morbidity. This study aimed to
compare two of the treatment options, Spinal Cord Stimulation (SCS) and Percutaneous
Myocardial Laser Revascularisation (PMR) in terms of clinical outcomes and cost-effectiveness.
Methods: Eligible patients were randomised to PMR or SCS and followed up for exercise
tolerance time (ETT), Canadian Cardiovascular Society (CCS) classification and the quality of life
measures SF-36, Seattle Angina Questionnaire and the EuroQoL at 3, 12 and 24 months. Utilities
were calculated using the EQ-5D and these and costs were compared between groups. The
incremental cost-effectiveness ratio (ICER) per QALY for SCS compared to PMR was also
calculated.
Results: At 24 months post-randomisation, patients that had SCS and PMR had similar ETT (mean
difference 0.05, 95% CI -2.08, 2.18, p = 0.96) and there was no difference in CCS classification or
quality of life outcomes. The difference in overall mean costs when comparing SCS to PMR was
GBP5,520 (95% CI GBP1,966 to GBP8,613; p < 0.01) and the ICER of using SCS was GBP46,000
per QALY.
Conclusion: Outcomes after SCS did not differ appreciably from those after PMR, with the former
procedure being less cost-effective as currently applied. Larger studies could clarify which patients
would most benefit from SCS, potentially increasing cost-effectiveness.
Trial registration: Current Controlled Trials ISRCTN0964895
Advances in Editors for IMS LD in the TENCompetence Project
No full textThe TENCompetence project has made the first release of IMS LD tools which provide significant improvements in usability and integration. The project and the user group are introduced, and the principal improvements described. The results of initial evaluation are outlined, and the direction of future work is indicated
Using Widgets to Provide Portable Services for IMS Learning Design
No full textSharples, P., Griffiths, D., & Scott, W. (2008). Using Widgets to Provide Portable Services for IMS Learning Design. In R. Koper, K. Stefanov & D. Dicheva (Eds.), Proceedings of the 5th International TENCompetence Open Workshop "Stimulating Personal Development and Knowledge Sharing" (pp. 57-60). October, 30-31, 2008, Sofia, Bulgaria: TENCompetence Workshop. [For the whole proceedings please see also http://hdl.handle.net/1820/1961 ]Since the publication of the IMS LD specification it has been recognised that the lack of a rich set of runtime
services is a major barrier to adoption. The approaches taken to resolving this problem are reviewed, and their
strengths and limitations identified. A generic widget server developed by the authors is described. Integration of
the widget server with the IMS LD runtime system provides an extensible set of services. This has been demonstrated
with the creation of widgets for forums, messaging, vote, and Google maps which are provided for users within the
context of their role and activities in a Unit of Learning. Authoring and administration are described, showing how
the system is both extensible and portable.The work on this publication has been sponsored by the TENCompetence Integrated Project that is funded by the European Commission's 6th Framework Programme, priority IST/Technology Enhanced Learning. Contract 027087 [http://www.tencompetence.org
Extensión de la especificación IMS Learning Design desde la adaptación e integración de unidades de aprendizaje
No full textIMS Learning Design (IMS-LD) representa una corriente actual en aprendizaje online y blended que se caracteriza porque: a) Es una especificación que pretende estandarizar procesos de aprendizaje, así como reutilizarlos en diversos contextos b) Posee una expresividad pedagógica más elaborada que desarrollos anteriores o en proceso c) Mantiene una relación cordial y prometedora con Learning Management Systems (LMSs), herramientas de autoría y de ejecución d) Existe una amplia variedad de grupos de investigación y proyectos europeos trabajando sobre ella, lo que augura una sostenibilidad, al menos académica Aun así, IMS Learning Design es un producto inicial (se encuentra en su primera versión, de 2003) y mejorable en diversos aspectos, como son la expresividad pedagógica y la interoperabilidad. En concreto, en esta tesis nos centramos en el aprendizaje adaptativo o personalizado y en la integración de Unidades de Aprendizaje, como dos de los pilares que definen la especificación, y que al mismo tiempo la potencian considerablemente. El primero (aprendizaje adaptativo) hace que se puedan abordar itinerarios individuales personalizados de estudio, tanto en flujo de aprendizaje como en contenido o interfaz; el segundo (integración) permite romper el aislamiento de los paquetes de información o cursos (Unidades de Aprendizaje, UoL) y establecer un diálogo con otros sistemas (LMSs), modelos y estándares, así como una reutilización de dichas UoLs en diversos contextos. En esta tesis realizamos un estudio de la especificación desde la base, analizando su modelo de información y cómo se construyen Unidades de Aprendizaje. Desde el Nivel A al Nivel C analizamos y criticamos la estructura de la especificación basándonos en un estudio teórico y una investigación práctica fruto del modelado de Unidades de Aprendizaje reales y ejecutables que nos proporcionan una información muy útil de base, y que mayormente adjuntamos en los anexos, para no interferir en el flujo de lectura del cuerpo principal. A partir de este estudio, analizamos la integración de Unidades de Aprendizaje con otros sistemas y especificaciones, abarcando desde la integración mínima mediante un enlace directo hasta la compartición de variables y estados que permiten una comunicación en tiempo real de ambas partes. Exponemos aquí también las conclusiones de diversos casos de estudio basados en adaptación que se anexan al final de la tesis y que se vuelven un instrumento imprescindible para lograr una solución real y aplicable. Como segundo pilar de la tesis complementario a la integración de Unidades de Aprendizaje, estudiamos el aprendizaje adaptativo: Los tipos, los avances y los enfoques y restricciones de modelado dentro de IMS-LD. Por último, y como complemento de la investigación teórica, a través de diversos casos prácticos estudiamos la manera en que IMS-LD modela la perzonalización del aprendizaje y hasta qué punto. Este primer bloque de análisis (general, integración y aprendizaje adaptativo) nos permite realizar una crítica estructural de IMS-LD en dos grandes apartados: Modelado y Arquitectura. Modelado apunta cuestiones que necesitan mejora, modificación, extensión o incorporación de elementos de modelado dentro de IMS-LD, como son procesos, componentes y recursos de programación. Arquitectura engloba otras cuestiones centradas en la comunicación que realiza IMS-LD con el exterior y que apuntan directamente a capas estructurales de la especificación, más allá del modelado. Aunque se encuentra fuera del núcleo de esta tesis, también se ha realizado una revisión de aspectos relacionados con Herramientas de autoría, por ser este un aspecto que condiciona el alcance del modelado y la penetración de la especificación en los distintos públicos objetivo. Sobre Herramientas, no obstante, no realizamos ninguna propuesta de mejora. La solución desarrollada, se centra en las diversas cuestiones sobre Modelado y Arquitectura encontradas en el análisis. Esta solución se compone de un conjunto de propuestas de estructuras, nuevas o ya existentes y modificadas, a través de las que se refuerza la capacidad expresiva de la especificación y la capacidad de interacción con un entorno de trabajo ajeno. Esta investigación de tres años ha sido llevada a cabo entre 2004 y 2007, principalmente con colegas de The Open University of The Netherlands, The University of Bolton, Universitat Pompeu Fabra y del departamento Research & Innovation de ATOS Origin, y ha sido desarrollada parcialmente dentro de proyectos europeos como UNFOLD, EU4ALL y ProLearn. La conclusión principal que se extrae de esta investigación es que IMS-LD necesita una reestructuración y modificación de ciertos elementos, así como la incorporación de otros nuevos, para mejorar una expresividad pedagógica y una capacidad de integración con otros sistemas de aprendizaje y estándares eLearning, si se pretenden alcanzar dos de los objetivos principales establecidos de base en la definición de esta especificación: La personalización del proceso de aprendizaje y la interoperabilidad real. Aun así, es cierto que la implantación de la especificación se vería claramente mejorada si existieran unas herramientas de más alto nivel (preferiblemente con planteamiento visual) que permitieran un modelado sencillo por parte de los usuarios finales reales de este tipo de especificaciones, como son los profesores, los creadores de contenido y los pedagogos-didactas que diseñan la experienicia de aprendizaje. Este punto, no obstante, es ajeno a la especificación y afecta a la interpretación que de la misma realizan los grupos de investigación y compañías que desarrollan soluciones de autoría. _____________________________________________IMS Learning Design (IMS-LD) is a current asset in eLearning and blended learning, due
to several reasons:
a) It is a specification that points to standardization and modeling of learning processes,
and not just content; at the same time, it is focused on the re-use of the information
packages in several contexts;
b) It shows a deeper pedagogical expressiveness than other specifications, already
delivered or in due process
c) It is integrated at different levels into well-known Learning Management Systems
(LMSs)
d) There are a huge amount of European research projects and groups working with it,
which aims at sustainability (in academia, at least)
Nevertheless, IMS-LD is roughly an initial outcome (be aware that we are still working
with the same release, dated on 2003). Therefore, it can and must be improved in
several aspects, i.e., pedagogical expressiveness and interoperability. In this thesis, we
concentrate on Adaptive Learning (or Personalised Learning) and on the Integration of
Units of Learning (UoLs). They both are core aspects which the specification is built upon.
They also can improve it significantly. Adaptation makes personalised learning itineraries,
adapted to every role, to every user involved in the process, and focus on several
aspects, i.e., flow, content and interface. Integration fosters the re-use of IMS-LD
information packages in different contexts and connects both-ways UoLs with other
specifications, models and LMSs. In order to achive these goals we carry out a threephase
analysis. First, analysis of IMS-LD in several steps: foundations, information
model, construction of UoLs. From Level A to Level C, we analyse and review the
specification structure. We lean on a theoretical frameword, along with a practical
approach, coming from the actual modeling of real UoLs which give an important report
back. Out of this analysis we get a report on the general structure of IMS-LD.
Second, analysis and review of the integration of UoLs with several LMSs, models and
specifications: we analyse three different types of integration: a) minimal integration,
with a simple link between parts; b) embedded integration, with a marriage of both parts
in a single information package; and d) full integration, sharing variables and states
between parts. In this step, we also show different case studies and report our partial
conclusions.
And third, analysis and review of how IMS-LD models adaptive learning: we define,
classify and explain several types of adaptation and we approach them with the specificacion. A key part of this step is the actual modeling of UoLs showing adaptive
learning processes. We highlight pros and cons and stress drawbacks and weak points
that could be improved in IMS-LD to support adaptation, but also general learning
processes
Out of this three-step analysis carried out so far (namely general, integration,
adaptation) we focus our review of the IMS-LD structure and information model on two
blocks: Modeling and Architecture. Modeling is focused on process, components and
programming resources of IMS-LD. Architecture is focused on the communication that
IMS-LD establishes outside, both ways, and it deals with upper layers of the specification,
beyong modeling issues. Modeling and Architecture issues need to be addressed in order
to improve the pedagogical expressiveness and the integration of IMS-LD. Furthermore,
we provide an orchestrated solution which meets these goals. We develop a structured
and organized group of modifications and extensions of IMS-LD, which match the
different reported problems issues. We suggest modifications, extensions and addition of
different elements, aiming at the strength of the specification on adaptation and
integration, along with general interest issues.
The main conclusion out of this research is that IMS-LD needs a re-structure and a
modification of some elements. It also needs to incorporate new ones. Both actions
(modification and extension) are the key to improve the pedagogical expressiveness and
the integration with other specifications and eLearning systems. Both actions aim at two
clear objectives in the definition of IMS-LD: the personalisation of learning processes,
and a real interoperability. It is fair to highlight the welcome help of high-level visual
authoring tools. They can support a smoother modeling process that could focus on
pedagogical issues and not on technical ones, so that a broad target group made of
teachers, learning designers, content creators and pedagogues could make use of the
specification in a simpler way. However, this criticism is outside the specification, so
outside the core of this thesis too.
This three-year research (2004-2007) has been carried out along with colleagues from
The Open University of The Netherlands, The University of Bolton, Universitat Pompeu
Fabra and from the Department of Research & Innovation of ATOS Origin. In addition, a
few European projects, like UNFOLD, EU4ALL and ProLearn, have partially supported it
Dynamic Prediction, Mediation and Communication for Survival Outcomes, with applications to Cystic Fibrosis
No full textPatients with chronic diseases and their clinicians want accurate and up-to-date information about risk, prognosis, and survival. The overall aim of this thesis is to advance statistical methods available for providing such information. The methods are motivated by analysis of cystic fibrosis (CF), a genetic life-shortening disease, and illustrated using longitudinal data from the UK CF Registry. First, dynamic models, that update predicted survival probabilities as new measurements become available, are studied. Although machine learning methods are established for prediction problems, they have not been widely used in dynamic survival prediction. Here, the combination of a machine learning ensemble with the landmarking approach is developed. Predictive performance of this method is compared to that of the most commonly-used statistical techniques: joint modelling and landmarking. A simulation study investigates cases where a machine learning ensemble may improve predictive accuracy. This thesis then provides a review of literature on communicating survival predictions, focusing on preferred graphical formats, comprehension by a broad audience, and best practices in survival communication. Based on this literature and semi-structured interviews conducted by qualitative research partners, an online tool was created. This provides life expectancy information sensitively and according to an individual's characteristics. In the final part of the thesis, CF-related diabetes (CFRD), a common comorbidity of CF, and its role in survival are investigated. Using multi-state models, the relationship between CFRD and survival is described. Mechanisms through which CFRD affects survival are explored using two methods that can accommodate longitudinal mediators for survival outcomes. Each method is applied to a stacked dataset, constructed in similar fashion to a landmark dataset, designed to maximally use the longitudinal registry data. A simulation study investigates the sensitivity of these two methods to model misspecification and data availability issues
- …
