Predicting Conversion from Mild Cognitive Impairment to Alzheimer’s Disease Using <i>K</i>-Means Clustering on MRI Data
Alzheimer’s disease (<i>AD</i>) is a neurodegenerative disorder that leads to the loss of cognitive functions due to the deterioration of brain tissue. Current diagnostic methods are often invasive or costly, limiting their widespread use. Developing non-invasive and cost-effective scree...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-02-01
|
| Series: | Information |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2078-2489/15/2/96 |
| _version_ | 1797297941075984384 |
|---|---|
| author | Miranda Bellezza Azzurra di Palma Andrea Frosini |
| author_facet | Miranda Bellezza Azzurra di Palma Andrea Frosini |
| author_sort | Miranda Bellezza |
| collection | DOAJ |
| description | Alzheimer’s disease (<i>AD</i>) is a neurodegenerative disorder that leads to the loss of cognitive functions due to the deterioration of brain tissue. Current diagnostic methods are often invasive or costly, limiting their widespread use. Developing non-invasive and cost-effective screening methods is crucial, especially for identifying patients with mild cognitive impairment (<i>MCI</i>) at the risk of developing Alzheimer’s disease. This study employs a Machine Learning (ML) approach, specifically <i>K</i>-means clustering, on a subset of pixels common to all magnetic resonance imaging (MRI) images to rapidly classify subjects with <i>AD</i> and those with normal Normal Cognitive (<i>NC</i>). In particular, we benefited from defining significant pixels, a narrow subset of points (in the range of 1.5% to 6% of the total) common to all MRI images and related to more intense degeneration of white or gray matter. We performed <i>K</i>-means clustering, with k = 2, on the significant pixels of <i>AD</i> and <i>NC</i> MRI images to separate subjects belonging to the two classes and detect the class centroids. Subsequently, we classified subjects with <i>MCI</i> using only the significant pixels. This approach enables quick classification of subjects with <i>AD</i> and <i>NC</i>, and more importantly, it predicts <i>MCI</i>-to-<i>AD</i> conversion with high accuracy and low computational cost, making it a rapid and effective diagnostic tool for real-time assessments. |
| first_indexed | 2024-03-07T22:27:37Z |
| format | Article |
| id | doaj.art-64abbd56fab3469d8a9e6d1e8211abf8 |
| institution | Directory Open Access Journal |
| issn | 2078-2489 |
| language | English |
| last_indexed | 2024-03-07T22:27:37Z |
| publishDate | 2024-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Information |
| spelling | doaj.art-64abbd56fab3469d8a9e6d1e8211abf82024-02-23T15:21:06ZengMDPI AGInformation2078-24892024-02-011529610.3390/info15020096Predicting Conversion from Mild Cognitive Impairment to Alzheimer’s Disease Using <i>K</i>-Means Clustering on MRI DataMiranda Bellezza0Azzurra di Palma1Andrea Frosini2Department of Mathematics and Informatics, University of Florence, 50134 Florence, ItalyDepartment of Mathematics and Informatics, University of Florence, 50134 Florence, ItalyDepartment of Mathematics and Informatics, University of Florence, 50134 Florence, ItalyAlzheimer’s disease (<i>AD</i>) is a neurodegenerative disorder that leads to the loss of cognitive functions due to the deterioration of brain tissue. Current diagnostic methods are often invasive or costly, limiting their widespread use. Developing non-invasive and cost-effective screening methods is crucial, especially for identifying patients with mild cognitive impairment (<i>MCI</i>) at the risk of developing Alzheimer’s disease. This study employs a Machine Learning (ML) approach, specifically <i>K</i>-means clustering, on a subset of pixels common to all magnetic resonance imaging (MRI) images to rapidly classify subjects with <i>AD</i> and those with normal Normal Cognitive (<i>NC</i>). In particular, we benefited from defining significant pixels, a narrow subset of points (in the range of 1.5% to 6% of the total) common to all MRI images and related to more intense degeneration of white or gray matter. We performed <i>K</i>-means clustering, with k = 2, on the significant pixels of <i>AD</i> and <i>NC</i> MRI images to separate subjects belonging to the two classes and detect the class centroids. Subsequently, we classified subjects with <i>MCI</i> using only the significant pixels. This approach enables quick classification of subjects with <i>AD</i> and <i>NC</i>, and more importantly, it predicts <i>MCI</i>-to-<i>AD</i> conversion with high accuracy and low computational cost, making it a rapid and effective diagnostic tool for real-time assessments.https://www.mdpi.com/2078-2489/15/2/96Alzheimer’s disease<i>K</i>-means clusteringpermutation testMRI image |
| spellingShingle | Miranda Bellezza Azzurra di Palma Andrea Frosini Predicting Conversion from Mild Cognitive Impairment to Alzheimer’s Disease Using <i>K</i>-Means Clustering on MRI Data Information Alzheimer’s disease <i>K</i>-means clustering permutation test MRI image |
| title | Predicting Conversion from Mild Cognitive Impairment to Alzheimer’s Disease Using <i>K</i>-Means Clustering on MRI Data |
| title_full | Predicting Conversion from Mild Cognitive Impairment to Alzheimer’s Disease Using <i>K</i>-Means Clustering on MRI Data |
| title_fullStr | Predicting Conversion from Mild Cognitive Impairment to Alzheimer’s Disease Using <i>K</i>-Means Clustering on MRI Data |
| title_full_unstemmed | Predicting Conversion from Mild Cognitive Impairment to Alzheimer’s Disease Using <i>K</i>-Means Clustering on MRI Data |
| title_short | Predicting Conversion from Mild Cognitive Impairment to Alzheimer’s Disease Using <i>K</i>-Means Clustering on MRI Data |
| title_sort | predicting conversion from mild cognitive impairment to alzheimer s disease using i k i means clustering on mri data |
| topic | Alzheimer’s disease <i>K</i>-means clustering permutation test MRI image |
| url | https://www.mdpi.com/2078-2489/15/2/96 |
| work_keys_str_mv | AT mirandabellezza predictingconversionfrommildcognitiveimpairmenttoalzheimersdiseaseusingikimeansclusteringonmridata AT azzurradipalma predictingconversionfrommildcognitiveimpairmenttoalzheimersdiseaseusingikimeansclusteringonmridata AT andreafrosini predictingconversionfrommildcognitiveimpairmenttoalzheimersdiseaseusingikimeansclusteringonmridata |