Applying spatial distribution analysis techniques to classification of 3D medical images

• Published in: Artificial intelligence in medicine Vol. 33; no. 3; pp. 261 - 280
• Main Authors: Pokrajac, Dragoljub, Megalooikonomou, Vasileios, Lazarevic, Aleksandar, Kontos, Despina, Obradovic, Zoran
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier B.V 01.03.2005
• Subjects: 3 dimensional images; Algorithms; Alzheimer Disease - physiopathology; Alzheimer's Disease; Analysis; Attention Deficit Disorder with Hyperactivity - physiopathology; Classification; Diagnostic Imaging - classification; Diagnostic Imaging - statistics & numerical data; Humans; Image Processing, Computer-Assisted - classification; Image Processing, Computer-Assisted - methods; Image Processing, Computer-Assisted - statistics & numerical data; Imaging, Three-Dimensional - classification; Imaging, Three-Dimensional - statistics & numerical data; Language; Likelihood Functions; Magnetic Resonance Imaging; Medical images; Medicine; Normal Distribution; Probability; Probability distributions; Regions of interest; Similarity measures; Spatial data mining; Spatial distribution; Spatial data mining; Medical images; Regions of interest; Probability distributions; Similarity measures; Classification
• ISSN: 0933-3657; 1873-2860
• DOI: 10.1016/j.artmed.2004.07.001

The objective of this paper is to classify 3D medical images by analyzing spatial distributions to model and characterize the arrangement of the regions of interest (ROIs) in 3D space. Two methods are proposed for facilitating such classification. The first method uses measures of similarity, such as the Mahalanobis distance and the Kullback–Leibler (KL) divergence, to compute the difference between spatial probability distributions of ROIs in an image of a new subject and each of the considered classes represented by historical data (e.g., normal versus disease class). A new subject is predicted to belong to the class corresponding to the most similar dataset. The second method employs the maximum likelihood (ML) principle to predict the class that most likely produced the dataset of the new subject. The proposed methods have been experimentally evaluated on three datasets: synthetic data (mixtures of Gaussian distributions), realistic lesion-deficit data (generated by a simulator conforming to a clinical study), and functional MRI activation data obtained from a study designed to explore neuroanatomical correlates of semantic processing in Alzheimer's disease (AD). Performed experiments demonstrated that the approaches based on the KL divergence and the ML method provide superior accuracy compared to the Mahalanobis distance. The later technique could still be a method of choice when the distributions differ significantly, since it is faster and less complex. The obtained classification accuracy with errors smaller than 1% supports that useful diagnosis assistance could be achieved assuming sufficiently informative historic data and sufficient information on the new subject.