Improved multimodal biomarkers for Alzheimer's disease and mild cognitive impairment diagnosis - Data from ADNI

Antonio Martinez-Torteya, Víctor Treviño-Alvarado, José Tamez-Peña

Resultado de la investigación

6 Citas (Scopus)

Resumen

The accurate diagnosis of Alzheimer's disease (AD) and mild cognitive impairment (MCI) confers many clinical research and patient care benefits. Studies have shown that multimodal biomarkers provide better diagnosis accuracy of AD and MCI than unimodal biomarkers, but their construction has been based on traditional statistical approaches. The objective of this work was the creation of accurate AD and MCI diagnostic multimodal biomarkers using advanced bioinformatics tools. The biomarkers were created by exploring multimodal combinations of features using machine learning techniques. Data was obtained from the ADNI database. The baseline information (e.g. MRI analyses, PET analyses and laboratory essays) from AD, MCI and healthy control (HC) subjects with available diagnosis up to June2012 was mined for case/controls candidates. The data mining yielded 47 HC, 83 MCI and 43 AD subjects for biomarker creation. Each subject was characterized by at least 980 ADNI features. A genetic algorithm feature selection strategy was used to obtain compact and accurate cross-validated nearest centroid biomarkers. The biomarkers achieved training classification accuracies of 0.983, 0.871 and 0.917 for HC vs. AD, HC vs. MCI and MCI vs. AD respectively. The constructed biomarkers were relatively compact: from 5 to 11 features. Those multimodal biomarkers included several widely accepted univariate biomarkers and novel image and biochemical features. Multimodal biomarkers constructed from previously and non-previously AD associated features showed improved diagnostic performance when compared to those based solely on previously AD associated features. © 2013 SPIE.
Idioma originalEnglish
DOI
EstadoPublished - 5 jun 2013
Publicado de forma externa
EventoProceedings of SPIE - The International Society for Optical Engineering -
Duración: 5 jun 2013 → …

Conference

ConferenceProceedings of SPIE - The International Society for Optical Engineering
Período5/6/13 → …

Huella dactilar

Alzheimer's Disease
biomarkers
impairment
Biomarkers
Diagnostics
Disease control
data mining
machine learning
Case-control
Bioinformatics
Centroid
genetic algorithms
centroids
Magnetic resonance imaging
Feature Selection
Univariate
Data mining
Learning systems
Feature extraction
Baseline

All Science Journal Classification (ASJC) codes

  • Electronic, Optical and Magnetic Materials
  • Condensed Matter Physics
  • Computer Science Applications
  • Applied Mathematics
  • Electrical and Electronic Engineering

Citar esto

Martinez-Torteya, A., Treviño-Alvarado, V., & Tamez-Peña, J. (2013). Improved multimodal biomarkers for Alzheimer's disease and mild cognitive impairment diagnosis - Data from ADNI. Papel presentado en Proceedings of SPIE - The International Society for Optical Engineering, . https://doi.org/10.1117/12.2008100
Martinez-Torteya, Antonio ; Treviño-Alvarado, Víctor ; Tamez-Peña, José. / Improved multimodal biomarkers for Alzheimer's disease and mild cognitive impairment diagnosis - Data from ADNI. Papel presentado en Proceedings of SPIE - The International Society for Optical Engineering, .
@conference{5c6b8c832c5348309e28effb32922953,
title = "Improved multimodal biomarkers for Alzheimer's disease and mild cognitive impairment diagnosis - Data from ADNI",
abstract = "The accurate diagnosis of Alzheimer's disease (AD) and mild cognitive impairment (MCI) confers many clinical research and patient care benefits. Studies have shown that multimodal biomarkers provide better diagnosis accuracy of AD and MCI than unimodal biomarkers, but their construction has been based on traditional statistical approaches. The objective of this work was the creation of accurate AD and MCI diagnostic multimodal biomarkers using advanced bioinformatics tools. The biomarkers were created by exploring multimodal combinations of features using machine learning techniques. Data was obtained from the ADNI database. The baseline information (e.g. MRI analyses, PET analyses and laboratory essays) from AD, MCI and healthy control (HC) subjects with available diagnosis up to June2012 was mined for case/controls candidates. The data mining yielded 47 HC, 83 MCI and 43 AD subjects for biomarker creation. Each subject was characterized by at least 980 ADNI features. A genetic algorithm feature selection strategy was used to obtain compact and accurate cross-validated nearest centroid biomarkers. The biomarkers achieved training classification accuracies of 0.983, 0.871 and 0.917 for HC vs. AD, HC vs. MCI and MCI vs. AD respectively. The constructed biomarkers were relatively compact: from 5 to 11 features. Those multimodal biomarkers included several widely accepted univariate biomarkers and novel image and biochemical features. Multimodal biomarkers constructed from previously and non-previously AD associated features showed improved diagnostic performance when compared to those based solely on previously AD associated features. {\circledC} 2013 SPIE.",
author = "Antonio Martinez-Torteya and V{\'i}ctor Trevi{\~n}o-Alvarado and Jos{\'e} Tamez-Pe{\~n}a",
year = "2013",
month = "6",
day = "5",
doi = "10.1117/12.2008100",
language = "English",
note = "Proceedings of SPIE - The International Society for Optical Engineering ; Conference date: 05-06-2013",

}

Improved multimodal biomarkers for Alzheimer's disease and mild cognitive impairment diagnosis - Data from ADNI. / Martinez-Torteya, Antonio; Treviño-Alvarado, Víctor; Tamez-Peña, José.

2013. Papel presentado en Proceedings of SPIE - The International Society for Optical Engineering, .

Resultado de la investigación

TY - CONF

T1 - Improved multimodal biomarkers for Alzheimer's disease and mild cognitive impairment diagnosis - Data from ADNI

AU - Martinez-Torteya, Antonio

AU - Treviño-Alvarado, Víctor

AU - Tamez-Peña, José

PY - 2013/6/5

Y1 - 2013/6/5

N2 - The accurate diagnosis of Alzheimer's disease (AD) and mild cognitive impairment (MCI) confers many clinical research and patient care benefits. Studies have shown that multimodal biomarkers provide better diagnosis accuracy of AD and MCI than unimodal biomarkers, but their construction has been based on traditional statistical approaches. The objective of this work was the creation of accurate AD and MCI diagnostic multimodal biomarkers using advanced bioinformatics tools. The biomarkers were created by exploring multimodal combinations of features using machine learning techniques. Data was obtained from the ADNI database. The baseline information (e.g. MRI analyses, PET analyses and laboratory essays) from AD, MCI and healthy control (HC) subjects with available diagnosis up to June2012 was mined for case/controls candidates. The data mining yielded 47 HC, 83 MCI and 43 AD subjects for biomarker creation. Each subject was characterized by at least 980 ADNI features. A genetic algorithm feature selection strategy was used to obtain compact and accurate cross-validated nearest centroid biomarkers. The biomarkers achieved training classification accuracies of 0.983, 0.871 and 0.917 for HC vs. AD, HC vs. MCI and MCI vs. AD respectively. The constructed biomarkers were relatively compact: from 5 to 11 features. Those multimodal biomarkers included several widely accepted univariate biomarkers and novel image and biochemical features. Multimodal biomarkers constructed from previously and non-previously AD associated features showed improved diagnostic performance when compared to those based solely on previously AD associated features. © 2013 SPIE.

AB - The accurate diagnosis of Alzheimer's disease (AD) and mild cognitive impairment (MCI) confers many clinical research and patient care benefits. Studies have shown that multimodal biomarkers provide better diagnosis accuracy of AD and MCI than unimodal biomarkers, but their construction has been based on traditional statistical approaches. The objective of this work was the creation of accurate AD and MCI diagnostic multimodal biomarkers using advanced bioinformatics tools. The biomarkers were created by exploring multimodal combinations of features using machine learning techniques. Data was obtained from the ADNI database. The baseline information (e.g. MRI analyses, PET analyses and laboratory essays) from AD, MCI and healthy control (HC) subjects with available diagnosis up to June2012 was mined for case/controls candidates. The data mining yielded 47 HC, 83 MCI and 43 AD subjects for biomarker creation. Each subject was characterized by at least 980 ADNI features. A genetic algorithm feature selection strategy was used to obtain compact and accurate cross-validated nearest centroid biomarkers. The biomarkers achieved training classification accuracies of 0.983, 0.871 and 0.917 for HC vs. AD, HC vs. MCI and MCI vs. AD respectively. The constructed biomarkers were relatively compact: from 5 to 11 features. Those multimodal biomarkers included several widely accepted univariate biomarkers and novel image and biochemical features. Multimodal biomarkers constructed from previously and non-previously AD associated features showed improved diagnostic performance when compared to those based solely on previously AD associated features. © 2013 SPIE.

U2 - 10.1117/12.2008100

DO - 10.1117/12.2008100

M3 - Paper

ER -

Martinez-Torteya A, Treviño-Alvarado V, Tamez-Peña J. Improved multimodal biomarkers for Alzheimer's disease and mild cognitive impairment diagnosis - Data from ADNI. 2013. Papel presentado en Proceedings of SPIE - The International Society for Optical Engineering, . https://doi.org/10.1117/12.2008100