Defining multivariate normative rules for healthy aging using neuroimaging and machine learning: an application to Alzheimer's disease.

BACKGROUND: Neuroimaging techniques combined with computational neuroanatomy have been playing a role in the investigation of healthy aging and Alzheimer's disease (AD). The definition of normative rules for brain features is a crucial step to establish typical and atypical aging trajectories. OBJECTIVE: To introduce an unsupervised pattern recognition method; to define multivariate normative rules of neuroanatomical measures; and to propose a brain abnormality index. METHODS: This study was based on a machine learning approach (one class classification or novelty detection) to neuroanatomical measures (brain regions, volume, and cortical thickness) extracted from the Alzheimer's Disease Neuroimaging Initiative (ADNI)'s database. We applied a ν-One-Class Support Vector Machine (ν-OC-SVM) trained with data from healthy subjects to build an abnormality index, which was compared with subjects diagnosed with mild cognitive impairment and AD. RESULTS: The method was able to classify AD subjects as outliers with an accuracy of 84.3% at a false alarm rate of 32.5%. The proposed brain abnormality index was found to be significantly associated with group diagnosis, clinical data, biomarkers, and future conversion to AD. CONCLUSION: These results suggest that one-class classification may be a promising approach to help in the detection of disease conditions. Our findings support a framework considering the continuum of brain abnormalities from healthy aging to AD, which is correlated with cognitive impairment and biomarkers measurements.

Medida da espessura cortical com imagens de ressonância magnética: técnica e aplicações na doença de Alzheimer e na polimicrogiria / Measurement of cortical thickness using magnetic resonance imaging: technique and applications in Alzheimer's disease and polymicrogyria

As técnicas de morfometria baseadas em imagens ressonância magnética desde o início do século XXI tem sido uma importante ferramenta para estudar doenças neurológicas. Um dos benefícios desta técnica é prover medidas de parâmetros que manualmente são difíceis de estimar e com um grande viés de operador. Este trabalho teve como objetivo descrever as alterações corticais em duas doenças, uma malformação do desenvolvimento cortical, e outra neurodegenerativa utilizando as técnicas de morfometria por estudo de superfície cortical. As imagens de ressonância magnética foram obtidas por sequências volumétricas ponderadas em T1 e analisadas através de um algoritmo específico, implementado no software FreeSurfer (Fischl et al., Harvard University - Estados Unidos). De um grupo de pacientes com malformação do desenvolvimento cortical foram analisados 3 pacientes com polimicrogiria (PMG), comparados com um grupo de 14 pessoas saudáveis e de idade similar. Foram também analisados 14 pacientes com alteração degenerativa (Doença de Alzheimer - DA) com manifestação recente, idade avançada e alto nível de escolaridade comparados com 20 idosos saudáveis. Este dado foi posteriormente utilizado para separar os dois grupos através de uma técnica de classificação multivariada, implementada pelo autor. A análise de cada um dos indivíduos com PMG comparados ao grupo controle apresentou aumento de espessura cortical nas áreas com polimicrogiria e redução também estatisticamente significante de espessura cortical em áreas sem polimicrogiria. No estudo do grupo de pacientes com DA observou-se redução de espessura cortical nas regiões do córtex para hipocampal, entorrinal, córtex límbico e córtex temporal superior. Além disso, a volumetria de estruturas subcorticais obtidas apresentou poder de classificação de 84,6% na comparação de pacientes com DA e controles saudáveis...

The techniques of brain morphometry based on magnetic resonance images since the beginning of the twenty-first century has been an important tool for studying neurological diseases. One of the benefits of this technique is to provide measures of parameters that are difficult to estimate manually and subject to operator bias. This study aimed to describe the changes in the cortex in two diseases, a malformation of cortical development and a neurodegenerative one, using surface based morphometry techniques. From a group of patients with malformation of cortical development were analyzed 3 with polymicrogyria (PMG), compared to a group of 14 healthy age matched subjects. The images from a T1 weighted volume were analyzed using a specific algorithm, implemented in the software FreeSurfer (Fischl et al., Harvard University - United States). We analyzed 14 patients with degenerative changes (Alzheimers disease - AD) with recent onset, advanced age and high level of education compared to 20 healthy age matched elderly. The images of this study were also obtained from a T1 weighted volume and were analyzed with the same software package. This analyzed data was used to separate the two groups through a multivariate classifier, implemented by the author. The analysis of each individual with PMG compared to the control group showed a statistically significant increase in cortical thickness in the areas with polymicrogyria and also a statistically significant reduction of cortical thickness in some areas without polymicrogyria. In the study group of AD patients the result was a cortical thinning statistically significant in hippocampal, entorhinal, limbic and superior temporal cortex. Moreover, the volume of subcortical structures has provided a classification power of 84.6% discriminating AD patients from healthy controls...