Catálogo de publicaciones - libros

A number of calculators that compute the risk of atherosclerosis has been developed and made available on the Internet. They all are based on computing weighted sum of risk factors. We propose instead to use more flexible expert systems to estimate the risk. The goal of the AtherEx expert system is to classify patients according to their atherosclerosis risk into four groups. This application is based on the rule–based expert system shell. Knowledge for the AtherEx was obtained (using the machine learning algorithm KEX) from the data concerning a longitudial study of atherosclerosis risk factors and further refined by domain expert. AtherEx is available for consultations on web.

Willingness to use computerised decision support systems is often jeopardised by lack of effective integration into existing user interfaces for electronic patient record. Concepts illustrated in this paper stem from the need of developing a project for the comparison of the physicians’ compliance to a clinical practice guideline before and after an electronic version of the guideline was introduced. Before starting the implementation, we performed a deep users’ needs analysis. It was accomplished also on the basis of lesson learned on past guideline implementations. The new idea was to classify guideline suggestions on the basis of some attributes, whose values will determine the modality of presentation of the suggestion itself, and on a different management of non compliance advice.

Knowledge-based systems are widely used in many application areas, especially in health care and more recently in rehabilitation. The rehabilitation of cerebrovascular accident (CVA) victims can be a complex and demanding task. This research developed a Rehabilitation Expert System for Post-Stroke Patients (REPS) consisting of an assessment stage and a rehabilitation stage. The assessment is based on internationally validated assessment tools and widely accepted methods of rehabilitation. Both stages are based on the expertise and knowledge of physical therapists at the Fjórðungssjúkrahúsið á Akureyri (FSA) University Hospital in Akureyri, Iceland. This prototype demonstrates the feasibility of knowledge-based systems in the field of physical therapy, in general, and post-stroke rehabilitation, in particular.

The automatic application of computerized guidelines and protocols in intensive care is not simple, given the high volume of data which must be processed and the need to offer advice on a continuous basis. However most of this data is available automatically and there is therefore the real possibility of improving the quality of care by providing timely advice without placing any additional load on the clinicians. In this paper we describe a prototype system which demonstrates the feasibility of doing this. We then discuss specific issues which arise in applying guidelines for such environments.

In the objective of building care team guidelines from field observations, this paper introduces a representation method for describing the medical collaborative activities during an ICU patient management. An event-centered representation of medical activities is built during a 3-step procedure, successively involving an event-centered observation phase, an action extraction and coding phase, and an event and collaborative representation phase. This method has been used for analyzing the management of 24 cases of neurological and multiple traumas. We have represented the different actions of the medical team members (clinicians, nurses and outside medical consultants), underlining collaborative information management and the strong interaction between information management and medical actions. This method also highlights the difficulty of cases management linked to diagnosis severity, complexity of the situation and time constraints.

It is by now widely accepted that medical guidelines can help to significantly improve the quality of medical care. Unfortunately, constructing the required medical guidelines is a very labour intensive and costly process. The cost of guideline construction would decrease if guidelines could be built from a set of building blocks that can be reused across guidelines. Such reusable building blocks would also result in more standardised guidelines, facilitating their deployment. The goal of this paper is to identify a collection of patterns that can be used as guideline building blocks. We propose two different methods for finding such patterns We compare the collections of patterns obtained through these two methods, and experimentally validate some of the patterns by checking their usability in the actual modelling of a medical guideline for breastcancer treatment.

Currently, various guideline representation languages are available. However, these languages are too complex and algorithmic to be used by medical staff or guideline developers. Therefore, a big gap is between the informa tion represented in published guidelines by guideline developers and the formal representation of clinical guideline used in an execution model. We try to close this gap by analyzing existing clinical guidelines written in free text, tables, or flow chart notation with the target of detecting prototypical patterns in those guidelines.

In the management of chronic diseases, therapeutic decisions depend on previously administered therapies as well as patient answers to these prior treatments. To take into account the specific management of chronic diseases, the knowledge base of the guided mode of the system ASTI has been structured as a double level decision tree, a clinical level to characterize the clinical profile of a patient, and a therapeutic level to identify the new recommended treatment when taking into account the patient’s therapeutic history. We propose to automatically derive the therapeutic level of the decision tree from the formal expression of guideline-based therapeutic strategies. The method has been developed using Augmented Transition Networks. Preliminary results obtained with additive therapeutic strategies such as (, + , + + ) where , , and are therapeutic classes which can be substituted respectively by (′, ′′), (′, ′′), and (′, ′′) are promising. However, the method needs to be extended to take into account more complex patterns.

Supporting therapy selection is a fundamental task for a system for the computerized management of clinical guidelines (GL). The goal is particularly critical when no alternative is really better than the others, from a strictly clinical viewpoint. In these cases, decision theory appears to be a very suitable means to provide advice. In this paper, we describe how algorithms for calculating utility, and for evaluating the optimal policy, can be exploited to fit the GL management context.

Clinical Practice Guidelines (CPGs) are increasingly common in clinical medicine for prescribing a set of rules that a physician should follow. Recent interest is in accurate retrieval of CPGs at the point of care. Examples are the CPGs digital libraries National Guideline Clearinghouse (NGC) or Vaidurya, which are organized along predefined concept hierarchies. In this case, both browsing and concept-based search can be applied. However, mandatory step in enabling both ways to CPGs retrieval is manual classification of CPGs along the concepts hierarchy, which is extremely time consuming. Supervised learning approaches are usually not satisfying, since commonly too few or no CPGs are provided as training set for each class.

In this paper we apply for multiple classification. is an unsupervised model that supports the physician in the classification of CPGs along the concepts hierarchy, even when no labeled examples are available. This model exploits lexical and topological information on the hierarchy to elaborate a classification hypothesis for any given CPG. We argue that such a kind of unsupervised classification can support a physician to classify CPGs by recommending the most probable classes. An experimental evaluation on various concept hierarchies with hundreds of CPGs and categories provides the empirical evidence of the proposed technique.