Application of data mining to predict the dosage of vancomycin as an outcome variable in a teaching hospital population

Chan, Allen; Chen, J.-X.; Wang, H.-Y.

doi:10.5414/cpp44533

Cited by 4 publications

(3 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is worth mentioning the wide variety of studies that use Decision Trees as a predictive method for medical problems. The algorithm C4.5 has been successfully applied on numerous occasions: (Chan et al 2006); (Tahir, Bouridane, 2006) and (Polat et al 2005). It is also frequent to find Decision Tree applications that are improved with the use of Genetic Algorithms in order to optimize the generated Rule Set, (Podgorelec et al 2005), and even in combination with the abovementioned Rough Set theory, (Kusiak et al 2004), in order to generate Decision Trees based on data with a lot of inconsistencies.…”

Section: Predictive Methods In Medicinementioning

confidence: 99%

See 1 more Smart Citation

SELECTING THE MOST ACCURATE FORECASTING METHOD FOR MEDICAL DIAGNOSIS. BREAST CANCER DIAGNOSIS - A Case Study

Almiñana¹,

Rabasa²,

Santamaría³

et al. 2010

Proceedings of the Third International Conference on Health Informatics

View full text Add to dashboard Cite

Different methods are usually applied for medical diagnosis problems. Most of them are only based on expert knowledge and the results are provided by model-driven methods and they are built from inflexible mathematical expressions. In this paper we suggest a Data-Driven perspective to facilitate the medical expert labour on diagnosis tasks. Furthermore, this paper offers a step by step procedure to select the most accurate forecasting method depending on the nature of the variables and the structure problem constraints. To validate such a selecting procedure, we apply it to a breast cancer diagnosis problem as a real case study.

show abstract

Section: Predictive Methods In Medicinementioning

confidence: 99%

“…Attribute selection (Wang et al 2006). Cancer forecasting DECISION TREES (DT) AND RULE SYSTEMS (RS) (Chan et al 2006). C4.5 forecasting posology (Gamberger et al 2002).…”

Section: Theory Studiesmentioning

confidence: 99%

SELECTING THE MOST ACCURATE FORECASTING METHOD FOR MEDICAL DIAGNOSIS. BREAST CANCER DIAGNOSIS - A Case Study

Almiñana¹,

Rabasa²,

Santamaría³

et al. 2010

Proceedings of the Third International Conference on Health Informatics

View full text Add to dashboard Cite

show abstract

“…It is worth mentioning the wide range of papers that show Decision Trees (themselves and their corresponding Rule Sets) as the most suitable predictive method for medical problems. The algorithm C4.5 has been successfully applied on numerous occasions: (Chan et al [12]); (Tahir and Bouridane [7]) and (Polat et al [9]) for numerical target variables; or ID3 for categorical antecedents and class variables.…”

Section: Part Ii: Selecting and Designing The Most Suitable Forecastimentioning

confidence: 99%

Data management for an anaesthesiology department optimization

Rabasa

Compañ

Agulló³

et al. 2013

WIT Transactions on Information and Communication Technologies

View full text Add to dashboard Cite

This paper is part of a Study in Progress which presents the Data Management main guidelines for a live project designed to optimize the anaesthesiology department of a public hospital in Spain. The main goal is the detection of dispensable pre-surgery tests (e.g. chest radiography, blood test or electrocardiogram), if there is a low correlation in the contrast between a patient's clinical history and incidences in the operating theatre. The paper's contribution consists of two parts. Part I shows the development of a data base web application to facilitate data entry of both the anaesthesiology and the operation data itself. The design of the web data base forms and the expert (anaesthetist) specifications are shown. Part II shows how to choose and design the classification rules as the most appropriate Data Mining method for this goal and explains how such a method must be parameterized. Thus, from the historical and clinical data of each new patient, the system could advise about the suitability (or not) for certain expensive and invasive pre-surgery tests (as chest radiography can be). By applying this software, the Hospital will be able to improve the scheduling of critical resources that are often very expensive and cause delays in operation waiting lists. Finally, as future research line, this research aims to extend the proposed method to other medical departments in order to achieve lower costs, avoid invasive tests and waiting lists, in conditions where it is statistically possible.

show abstract

Advanced Bioinformatics Tools and Strategies

Childs¹,

Ouyang²

2010

Principles and Practices of Plant Genomics, Volume 3

View full text Add to dashboard Cite

Application of data mining to predict the dosage of vancomycin as an outcome variable in a teaching hospital population

Cited by 4 publications

References 0 publications

SELECTING THE MOST ACCURATE FORECASTING METHOD FOR MEDICAL DIAGNOSIS. BREAST CANCER DIAGNOSIS - A Case Study

SELECTING THE MOST ACCURATE FORECASTING METHOD FOR MEDICAL DIAGNOSIS. BREAST CANCER DIAGNOSIS - A Case Study

Data management for an anaesthesiology department optimization

Advanced Bioinformatics Tools and Strategies

Contact Info

Product

Resources

About