Ventilator has been widely used to support the breathing needs of patients, and weaning is the process of removing ventilator from them. So far there is no positive answer about whether it will be successful to wean a patient off a ventilator. It may be of help if we develop an intelligent system to assist clinicians in making such a decision. In this paper, we apply an artificial neuromolecular system (ANM system), which is a selforganizing learning system, to a database of 189 weaned patients. The ANM system is a multi-level evolutionary learning architecture that captures the gradual transformability feature of structure–function relationship embedded in biological systems. Our experiments with the model show that the integrated system achieves a satisfactory result in separating those patients who have successful weaning results from those who do not, based on the 27 parameters that may affect their weaning results. Our parameter analysis shows that most of the parameters identified as significant by the system are the same as those by clinicians, but some are not. The finding of the latter should provide clinicians another dimension of information, in particular the effectiveness of each parameter in determining weaning results for patients.
Abstract.We have developed a vertical information processing model, motivated from physiological evidence, which integrates intra-and interneuronal information processing. Information processing at the intraneuronal levels is to create a repertoire of pattern processing neurons. Information processing at the interneuronal levels is to group appropriate pattern processing neurons to constitute an effective pattern processing system. The system was applied to a database of the weaning results of ventilator-dependent patients. Ventilator has been used to support the breathing need of patients, and weaning is the gradual process of removing it from ventilator-dependent patients. Experiments with the model show that the integrated system is able to learn to differentiate data in an autonomous manner, separating those patients who have successful weaning results from those who do not. Our parameter analysis shows that most of the parameters identified as significant by the system are the same as those by physicians, but some are not.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.