Pasar al contenido principal

Resultados relacionados

Resultados totales (Incluyendo duplicados): 4
Encontrada(s) 1 página(s)

Digibug. Repositorio Institucional de la Universidad de Granada
Publicación Artículo científico (JournalArticle)

An empirical assessment of deep learning approaches to task-oriented dialog management

  • Mateju, Lukas
  • Griol Barres, David
  • Callejas Carrión, Zoraida|||0000-0001-8891-5237
  • Molina, José Manuel
  • Sanchis, Araceli
Deep learning is providing very positive results in areas related to conversational interfaces, such as speech recognition, but its potential benefit for dialog management has still not been fully studied. In this paper, we perform an assessment of different configurations for deep-learned dialog management with three dialog corpora from different application domains and varying in size, dimensionality and possible system responses. Our results have allowed us to identify several aspects that can have an impact on accuracy, including the approaches used for feature extraction, input representation, context consideration and the hyper-parameters of the deep neural networks employed., This work was supported by the Student Grant Scheme 2020 of the Technical University in Liberec, the European Union’s Horizon 2020 research and innovation programme under grant agreement No 823907 (MENHIR project: https://menhir-project.eu), and by the Spanish project TEC2017-84593-C2-1-R.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

I-PALIA, An algorithm for discovering BPMN processes with duplicated tasks

  • Fernández Llatas, Carlos|||0000-0002-2819-5597
  • Burattin, Andrea
[EN] Process mining encompasses a range of methods designed to analyze event logs. Among these methods, control-flow discovery algorithms are particularly significant, as they enable the identification of real-world process models, known as in-vivo processes, in contrast to anticipated models. An obstacle faced by control-flow discovery algorithms is their limited ability to recognize duplicated activities, which are activities that occur in multiple locations within a process. This issue is particularly relevant in the healthcare sector, where numerous instances of duplicated activities exist in processes but remain undetected by conventional algorithms. This article introduces a novel concept for a control-flow discovery algorithm capable of effectively revealing duplicated activities. The effectiveness of this technique is demonstrated through experimentation on a synthetic dataset. Moreover, the algorithm has been implemented and its source code is available as open-source software, accessible both as a ProM plugin and a Java Maven dependency., Open Access funding provided thanks to the CRUE-CSIC agreement with Springer Nature. This research has been partially funded by projects MINE-GUIDE(PID2020-113723RB-C21) funded by MCIN/ AEI/10.13039/501100011033 and LIFECHAMPS(Grant Agreement No 875329) under European Union s Horizon 2020 research and innovation program. Funding for open access charge: CRUE-Universitat Politècnica de València
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Personalised Risk Modelling for Older Adult Cancer Survivors: Combining Wearable Data and Self-Reported Measures to Address Time-Varying Risks

  • Valero Ramon, Zoe|||0000-0002-4903-3896
  • Gema Ibanez-Sanchez|||0000-0003-1824-281X
  • Martinez-Millana, Antonio|||0000-0003-1056-5067
  • Fernández Llatas, Carlos|||0000-0002-2819-5597
[EN] Recent advancements in wearable devices have significantly enhanced remote patient monitoring, enabling healthcare professionals to evaluate conditions within home settings. While electronic health records (EHRs) offer extensive clinical data, they often lack crucial contextual information about patients¿ daily lives and symptoms. By integrating continuous self-reported outcomes related to vulnerability, anxiety, and depression from older adult cancer survivors with objective data from wearables, we can develop personalised risk models that address time-varying risk factors in cancer care. Our study combines real-world data from wearable devices with self-reported information, employing process mining techniques to analyse dynamic risk models for vulnerability and anxiety. Unlike traditional static assessments, this approach recognises that risk factors evolve. Collaborating with healthcare professionals, we analysed data from the LifeChamps study to create two dynamic risk models. This collaborative effort revealed how activity and sleep patterns influence self-reported vulnerability and anxiety among participants. It underscored the potential of wearable sensors and artificial intelligence techniques for deeper analysis and understanding, making us all part of a larger effort in cancer care. Overall, patients with prolonged sedentary activity had a higher risk of vulnerability, while those with highly dynamic sleep patterns were more likely to report anxiety and depression. Prostate-metastatic patients showed an increased risk of vulnerability compared to other cancer types., This research was partially funded by the European Union s Horizon 2020 Research and Innovation Programme under Grant Agreement no. 875329.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Analytical exploratory tool for healthcare professionals to monitor cancer patients&apos, progress

  • Valero Ramon, Zoe|||0000-0002-4903-3896
  • Fernández Llatas, Carlos|||0000-0002-2819-5597
  • Traver Salcedo, Vicente|||0000-0003-1806-8575
  • Collantes, Gonzalo
  • Valdivieso, Fernando
  • Billis, Antonis
  • Bamidis, Panagiotis
[EN] Cancer is a primary public concern in the European continent. Due to the large case numbers and survival rates, a significant population is living with cancer needs. Consequently, health professionals must deal with complex treatment decision-making processes. In this context, much data is collected during cancer care delivery. Once collected, these data are complex for health professionals to access to support clinical decision-making and performance review. There is a need for innovative tools that make clinical data more accessible to support cancer health professionals in these activities. One approach for providing health professionals with access to clinical data is to create the infrastructure and interface for a clinical tool to make data accessible in a relevant manner. In this sense, results should be understandable and valuable for cancer experts to evaluate and optimize cancer processes. This work aims to collect and report the process of developing an exploratory analytical Interactive Process Mining tool with clinical relevance for healthcare professionals for monitoring cancer patients¿ care processes in the context of the LifeChamps project. Following a co-creation and interactive approach thanks to the Interactive Process Mining paradigm, the tool presents patients¿ progress over time for different clinical models and a graphical and navigable Process Indicator in the context of prostate cancer patients., This work was partially funded by the European Union¿s Horizon 2020 research and innovation program under Grant Agreement No 875329.