[Bi-modality image classification based on independent component analysis].

Area of Science:

• Neuroimaging
• Machine Learning
• Medical Diagnostics

Background:

• Alzheimer's disease (AD) and mild cognitive impairment (MCI) pose significant diagnostic challenges.
• Accurate and early diagnosis is crucial for effective patient management and therapeutic interventions.
• Current diagnostic methods often rely on a combination of clinical assessments and neuroimaging, but improved accuracy is needed.

Purpose of the Study:

• To develop and evaluate a novel classification method for diagnosing Alzheimer's disease (AD) and mild cognitive impairment (MCI).
• To enhance diagnostic accuracy by integrating features from structural magnetic resonance imaging (sMRI) and positron emission tomography (PET).
• To compare the performance of a bi-modality approach against single-modality methods.

Main Methods:

• Applied infomax independent component analysis (ICA) to extract features from sMRI and PET data independently.
• Combined the extracted single-modality features using a weighted combination method.
• Utilized the combined features for classification of AD and MCI against healthy controls (HC).

Main Results:

• The bi-modality classification method achieved 93.75% accuracy for AD vs. HC (100% sensitivity, 87.64% specificity).
• For MCI vs. HC, the method reached 89.35% accuracy (81.85% sensitivity, 99.36% specificity).
• The integrated bi-modality approach demonstrated superior classification accuracy compared to using sMRI or PET data alone.

Conclusions:

• The proposed infomax ICA-based bi-modality feature combination method effectively improves the accuracy of AD and MCI diagnosis.
• Integrating sMRI and PET data offers significant advantages over single-modality neuroimaging for classifying neurodegenerative conditions.
• This approach holds promise for more precise and reliable early detection of Alzheimer's disease and mild cognitive impairment.