MRI texture-based radiomics analysis for the identification of altered functional networks in alcoholic patients and animal models

Alcohol use disorder (AUD) is a complex condition representing a leading risk factor for death, disease and disability. Its high prevalence and severe health consequences make necessary a better understanding of the brain network alterations to improve diagnosis and treatment. The purpose of this study was to evaluate the potential of resting-state fMRI 3D texture features as a novel source of biomarkers to identify AUD brain network alterations following a radiomics approach. A longitudinal study was conducted in Marchigian Sardinian alcohol-preferring msP rats (N = 36) who underwent resting-state functional and structural MRI before and after 30 days of alcohol or water consumption. A cross-sectional human study was also conducted among 33 healthy controls and 35 AUD patients. The preprocessed functional data corresponding to control and alcohol conditions were used to perform a probabilistic independent component analysis, identifying seven independent components as resting-state networks. Forty-three radiomic features extracted from each network were compared using a Wilcoxon signed-rank test with Holm correction to identify the network most affected by alcohol consumption. Features extracted from this network were then used in the machine learning process, evaluating two feature selection methods and six predictive models within a nested cross-validation structure. The classification was evaluated by computing the area under the ROC curve. Images were quantized using different numbers of gray-levels to test their influence on the results. The influence of ageing, data preprocessing, and brain iron accumulation were also analyzed. The methodology was validated using structural scans. The striatal network in alcohol-exposed msP rats presented the most significant number of altered features. The radiomics approach supported this result achieving good classification performance in animals (AUC = 0.915 ± 0.100, with 12 features) and humans (AUC = 0.724 ± 0.117, with 9 features) using a random forest model. Using the structural scans, high accuracy was achieved with a multilayer perceptron in both species (animals: AUC > 0.95 with 2 features, humans: AUC > 0.82 with 18 features). The best results were obtained using a feature selection method based on the p-value. The proposed radiomics approach is able to identify AUD patients and alcohol-exposed rats with good accuracy, employing a subset of 3D features extracted from fMRI. Furthermore, it can help identify relevant networks in drug addiction.

MAOA-VNTR Genotype Effects on Ventral Striatum-Hippocampus Network in Alzheimer's Disease: Analysis Using Structural Covariance Network and Correlation with Neurobehavior Performance

Functional polymorphisms in the promoter region of the monoamine oxidase A (MAOA) gene are associated with brain MAOA activity and transcriptional efficiency in patients with Alzheimer's disease (AD). This study investigated structural covariance networks mediated by MAOA-variable number tandem repeat (VNTR) genotypes in patients with AD, and assessed whether this effect was associated with sex. A total of 193 patients with AD were classified into four genotype groups based on MAOA transcriptional efficiency (female low [L], low-high + high activity groups [LH + H]; male L, male H groups). Structural covariance networks were constructed focusing on triple-network and striatal networks. Covariance strength was analyzed in the four groups, and the genotype and sex main effects and their interactions were analyzed. Significant peak cluster volumes were correlated with neurobehavioral scores to establish the clinical significance. MAOA genotypes mediated the structural covariance strength on the dorsolateral prefrontal cortex (dLPFC)-caudate axis in both sexes, but a higher covariance strength was shown in the female L group and male H group. The independent effect of male sex was related to higher covariance strength in the frontal medial superior region in the dLPFC, dorsal caudate (DC), and ventral superior striatum (VSs) seeds. In contrast, female sex had higher covariance strength in the frontal opercular areas anchored by the dLPFC, DC, and VSs seeds. Topographies showing higher covariance strength with sex interactions were found in the male H group and female L group in the dLPFC supplementary motor axis, DC-SMA, and DC-precentral axis. In our patients with AD, MAOA-VNTR polymorphisms and sex had independent and interactive effects on structural covariance networks, of which the dLPFC-, VSs-, and DC-anchored networks represented major endophenotypes that determined cognitive outcomes. The sex-genotype interaction model suggested that male high activity and female low activity may modulate brain morphometric connectivity and determine cognitive scores.

Catechol-O-Methyltransferase Val158Met Polymorphism on Striatum Structural Covariance Networks in Alzheimer's Disease

The catechol-O-methyltransferase enzyme metabolizes dopamine in the prefrontal axis, and its genetic polymorphism (rs4680; Val158Met) is a known determinant of dopamine signaling. In this study, we investigated the possible structural covariance networks that may be modulated by this functional polymorphism in patients with Alzheimer’s disease. Structural covariance networks were constructed by 3D T1 magnetic resonance imaging. The patients were divided into two groups: Met-carriers (n = 91) and Val-homozygotes (n = 101). Seed-based analysis was performed focusing on triple-network models and six striatal networks. Neurobehavioral scores served as the major outcome factors. The role of seed or peak cluster volumes, or a covariance strength showing Met-carriers > Val-homozygotes were tested for the effect on dopamine. Clinically, the Met-carriers had higher mental manipulation and hallucination scores than the Val-homozygotes. The volume-score correlations suggested the significance of the putaminal seed in the Met-carriers and caudate seed in the Val-homozygotes. Only the dorsal-rostral and dorsal-caudal putamen interconnected peak clusters showed covariance strength interactions (Met-carriers > Val-homozygotes), and the peak clusters also correlated with the neurobehavioral scores. Although the triple-network model is important for a diagnosis of Alzheimer’s disease, our results validated the role of the dorsal-putaminal-anchored network by the catechol-O-methyltransferase Val158Met polymorphism in predicting the severity of cognitive and behavior in subjects with Alzheimer’s disease.