Machine learning applied to second-trimester Down's syndrome screening

Research output: Chapter in BookChapterResearchpeer-review

Abstract

This chapter presents a new methodology to detect Down's syndrome during the second-trimester of gestation. We propose a Fuzzy Logic based classification method that is able of both improving the classification accuracy of the existing screening methods and extracting a set of intelligible rules that explain the system found. The accuracy is improved by reducing the False Positive rate up to 3%-4%, with a 60% of True Positive rate. Never before a screening method could reduce the FP rate below 5%, with a 60% of TP. The intelligibility is obtained by getting a reduced set of rules, expressed in a fuzzy way, which helps to explain the knowledge contained in the classification system found. Finally, conclude that the medical staff was able to understand the system expressed in the Fuzzy rules found, where its number of rules is just 10. © 2009 Nova Science Publishers, Inc. All rights reserved.
Original languageEnglish
Title of host publicationHandbook of Down Syndrome Research
Pages337-352
Number of pages15
Publication statusPublished - 1 Dec 2009

Fingerprint Dive into the research topics of 'Machine learning applied to second-trimester Down's syndrome screening'. Together they form a unique fingerprint.

  • Cite this

    Soler, V., & Prim, M. (2009). Machine learning applied to second-trimester Down's syndrome screening. In Handbook of Down Syndrome Research (pp. 337-352)