Search

Type 1 diabetes mellitus is a chronic disease that requires those affected to self-administer insulin to control their blood glucose level. However, the estimation of the correct insulin dosage is not easy due to the complexity of glucose metabolism, which usually leads to blood glucose levels far from the optimal. This paper presents an adaptive and personalised basal insulin recommender system based on Kalman filter theory that can be used with or without continuous glucose monitoring systems. The proposed approach is tested with the UVa/PADOVA simulator with eleven virtual adult subjects. It has been tested in combination with two different bolus calculators, and the performance achieved has been compared with that obtained with the default basal doses of the simulator, which can be assumed as optimal. The achieved results demonstrate that the proposed system rapidly converges to the optimal basal dose, and it can be used with adaptive bolus calculators without the risk of instability This project has received funding from the grant of the University of Girona 2016-2018 (MPCUdG2016) and the European Union Horizon 2020 research and innovation programme under grant agreement No. 689810, www.pepper.eu.com/, PEPPER

The use of artificial intelligence techniques to find out which Single Nucleotide Polymorphisms (SNPs) promote the development of a disease is one of the features of medical research, as such techniques may potentially aid early diagnosis and help in the prescription of preventive measures. In particular, the aim is to help physicians to identify the relevant SNPs related to Type 2 diabetes, and to build a decision-support tool for risk prediction. Methods: We use the Random Forest (RF) technique in order to search for the most important attributes (SNPs) related to diabetes, giving a weight (degree of importance), ranging between 0 and 1, to each attribute. Support Vector Machines and Logistic Regression have also been used since they are two other machine learning techniques that are well-established in the health community. Their performance has been compared to that achieved by RF. Furthermore, the relevance of the attributes obtained through the use of RF has then been used to perform predictions with k-Nearest Neighbour method weighting attributes in the similarity measure according to the relevance of the attributes with RF. Results: Testing is performed on a set of 677 subjects. RF is able to handle the complexity of features' interactions, overfitting, and unknown attribute values, providing the SNPs' relevance with an up to 0.89 area under the ROC curve in terms of risk prediction. RF outperforms all the other tested machine learning techniques in terms of prediction accuracy, and in terms of the stability of the estimated relevance of the attributes. Conclusions: The Random Forest is a useful method for learning predictive models and the relevance of SNPs without any underlying assumption This work was supported by the European Unions Horizon 2020 research and innovation programme [grant number 689810, PEPPER]; the University of Girona [grant number MPCUdG2016]; and the Spanish MINECO [grant number DPI2013-47450-C21-R].

This special section of this issue of the Artificial Intelligence in Medicine (AIIM) journal originates from the First Workshop on Artificial intelligence for Diabetes (AID 2016) on 30th August 2016. The workshop was part of the 22nd European Conference on Artificial Intelligence (ECAI 2016) in The Hague, Holland. Authors with papers accepted for the workshop were subsequently invited to revise and extend their work for publication in this issue, and a wider call was also announced to attract work from other outstanding researchers in the area. Authors were invited to submit original contributions on the overarching theme of Artificial Intelligence-based solutions to problems associated with diabetes. In particular, papers were sought on topics including Intelligent solutions to empower citizens with self-management of health conditions; Intelligent systems for glucose prediction and alarm generation; clinical decision support tools to deal with the avalanche of data gathered by sensors; data mining approaches for risk prediction and prevention of diabetes comorbidities; as well as community tools, platforms to support research in this area and data sets for benchmarkin The workshop was supported by the European Union’s Horizon 2020 research and innovation programme under grant agreement No 689810 (PEPPER)

Clinical Decision Support System (CDSSs) should form an important part of the field of clinical knowledge management technologies through their capacity to support the clinical process and use of knowledge, including knowledge maintenance and continuous learning, from diagnosis and investigation through surgery, treatment and long-term care. The work presented shows a workflow-based CDSS designed to give case-specific assessment to clinicians during complex surgery or Minimally Invasive Surgery (MISs). Following a perioperative workflow, the designed software will use a Case-Based Reasoning (CBR) methodology to retrieve similar past cases from a case base to provide support at any particular point of the process. The graphical user interface allows easy navigation through the whole support progress, from the initial configuration steps to the final results organized as sets of experiments easily visualized in a user-friendly way. The eXiTCDSS tool is presented giving support to a recent complex minimally invasive surgery which is receiving growing attention lately, the Transcatheter Aortic Valve Implantation (TAVI). The results obtained are presented on a basis of a real TAVI case base of 82 patients operated at Rennes University Hospital We would like to give our special thanks to the Hospital Clinic of Barcelona, the Centre Hospitalier Universitaire de Rennes and the Laboratoire Traitement du Signal et de l'Image from the University of Rennes for helping us with the case model definition and giving us example cases of TAVI patients. This work has been financed by the Spanish Government Commission Ministerio de Industria, Turismo y Comercio (MITyC) under the project PLAN AVANZA 2 labeled by Information Technology for European Advancement 2 (ITEA2). Also, this research project has been partially funded through the project labeled DPI2011-24929

Objective: Medical applications have special features (interpretation of results in medical metrics, experiment reproducibility and dealing with complex data) that require the development of particular tools. The eXiT*CBR framework is proposed to support the development of and experimentation with new case-based reasoning (CBR) systems for medical diagnosis. Method: Our framework offers a modular, heterogeneous environment that combines different CBR techniques for different application requirements. The graphical user interface allows easy navigation through a set of experiments that are pre-visualized as plots (receiver operator characteristics (ROC) and accuracy curves). This user-friendly navigation allows easy analysis and replication of experiments. Used as a plug-in on the same interface, eXiT*CBR can work with any data mining technique such as determining feature relevance. Results: The results show that eXiT*CBR is a user-friendly tool that facilitates medical users to utilize CBR methods to determine diagnoses in the field of breast cancer, dealing with different patterns implicit in the data. Conclusions: Although several tools have been developed to facilitate the rapid construction of prototypes, none of them has taken into account the particularities of medical applications as an appropriate interface to medical users. eXiT*CBR aims to fill this gap. It uses CBR methods and common medical visualization tools, such as ROC plots, that facilitate the interpretation of the results. The navigation capabilities of this tool allow the tuning of different CBR parameters using experimental results. In addition, the tool allows experiment reproducibility.

In this work we propose a user-friendly medically oriented tool for prognosis development systems and experimentation under a case-based reasoning methodology. The tool enables health care collaboration practice to be mapped in cases where different doctors share their expertise, for example, or where medical committee composed of specialists from different fields work together to achieve a final prognosis. Each agent with a different piece of knowledge classifies the given cases through metrics designed for this purpose. Since multiple solutions for the same case are useless, agents collaborate among themselves in order to achieve a final decision through a coordinated schema. For this purpose, the tool provides a weighted voting schema and an evolutionary algorithm (genetic algorithm) to learn robust weights. Moreover, to test the experiments, the tool includes stratified cross-validation methods which take the collaborative environment into account. In this paper the different collaborative facilities offered by the tool are described. A sample usage of the tool is also provided This research project has been partially funded through the projects labelled TIN2008-04547, DPI2011-24929 and CTQ2008-06865-C02-02 and the grants UDG-BR10/18 and FPU-AP2009-2831