Knowledge-based approach to septic shock patient data using a neural network with trapezoidal activation functions

Paetz, Jürgen

doi:10.1016/s0933-3657(03)00057-5

Cited by 36 publications

(23 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We give an overview of the neuro-fuzzy approach based on [11,15,16] for our purposes. The basic network topology is a feedforward network with separate neurons for every class in the first layer.…”

Section: Knowledge-based Approach To Virtual Screeningmentioning

confidence: 99%

“…In Algorithm 2 we present NFALG. For more technical details refer to [11,16]. The shrink procedure of NFALG (used in step 4 and 5) is a heuristic mechanism.…”

Section: Knowledge-based Approach To Virtual Screeningmentioning

confidence: 99%

“…Having NFALG at hand we explain a feature selection method that is directly based on rule performance and rule features [16]. For this purpose the rule performance measures frequency and confidence are used as introduced in Ref.…”

Section: Knowledge-based Approach To Virtual Screeningmentioning

confidence: 99%

See 2 more Smart Citations

Descriptor vector redesign by neuro-fuzzy analysis

Paetz

2005

Soft Comput

View full text Add to dashboard Cite

Virtual screening is an important challenge for bioinformaticians and molecular biologists. Finding the most active ligands that bind to target molecules by intelligent methods reduces the computational efforts in the drug discovery process dramatically. Recently, we proposed a neurofuzzy based approach to virtual screening by generating interval rules, each one with a class assignment for enriched regions of active ligands. The underlying input data is based on so called descriptor vector calculations. Descriptors are designed by experts, and they represent properties of the molecules. It is not assumed that all descriptors in a usually high dimensional descriptor vector are needed for classification. We demonstrate the usefulness of a neuro-fuzzy system for rule-based classification and feature selection. On the one hand the classification results can be used for efficient virtual screening, and on the other hand the rules can be used by experts for the ongoing drug design process. By feature analysis of the rules descriptor vectors can be redesigned. In the future more tools are needed that are capable not only of achieving several analysis results but giving also rule-based hints to the experts of how to plan their further research activities.

show abstract

Section: Knowledge-based Approach To Virtual Screeningmentioning

confidence: 99%

“…In Algorithm 2 we present NFALG. For more technical details refer to [11,16]. The shrink procedure of NFALG (used in step 4 and 5) is a heuristic mechanism.…”

Section: Knowledge-based Approach To Virtual Screeningmentioning

confidence: 99%

See 1 more Smart Citation

Descriptor vector redesign by neuro-fuzzy analysis

Paetz

2005

Soft Comput

View full text Add to dashboard Cite

show abstract

“…A review of the literature highlights previous studies in this area that have applied knowledge-based soft computing techniques to various scenarios associated with septic shock [3,4,5,6]. In [5], however, feature selection is applied in order to reduce data dimensionality by removing inputs that do not improve the prediction performance of the model.…”

Section: Introductionmentioning

confidence: 99%

“…By using two sets of 12 and 28 selected features from the pre-processed MEDAN database [6] described in [4] and [5], respectively, results were compared to these studies.…”

Section: Introductionmentioning

confidence: 99%

Predicting septic shock outcomes in a database with missing data using fuzzy modeling: Influence of pre-processing techniques on real-world data-based classification

Pereira

Fialho

Cismondi

et al. 2011

2011 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE 2011)

View full text Add to dashboard Cite

Real-world databases often contain missing data and existing correction algorithms deliver varying performance. Also, most modeling techniques are not suitable to deal with them automatically. In this study we examine different approaches to predicting septic shock in the presence of missing data. Some preprocessing techniques for managing missing data include disregarding data, or replacing it with information that by design introduces bias. In this study, we show that predictive performance improves by employing a minimum pre-processing technique, the Zero-Order-Hold (ZOH) method, by applying a Fuzzy C-Means clustering technique based on the partial distance calculation strategy (FCM-PDS) and by computing the final classification regarding the samples from each patient. Performance improvements continue to occur where up to approximately 60% of the data is missing, though for higher percentage the classification performance still is statistically improved. We further validate this approach by making comparisons with previous studies.

show abstract

Supervised Neuro-fuzzy Clustering for Life Science Applications

Paetz

2006

Biological and Medical Data Analysis

View full text Add to dashboard Cite

Knowledge-based approach to septic shock patient data using a neural network with trapezoidal activation functions

Cited by 36 publications

References 30 publications

Descriptor vector redesign by neuro-fuzzy analysis

Descriptor vector redesign by neuro-fuzzy analysis

Predicting septic shock outcomes in a database with missing data using fuzzy modeling: Influence of pre-processing techniques on real-world data-based classification

Supervised Neuro-fuzzy Clustering for Life Science Applications

Contact Info

Product

Resources

About