Mario Mascalchi

YNICL Journal 2021 Journal Article

Artificial intelligence for classification of temporal lobe epilepsy with ROI-level MRI data: A worldwide ENIGMA-Epilepsy study

• Ezequiel Gleichgerrcht
• Brent C. Munsell
• Saud Alhusaini
• Marina K.M. Alvim
• Núria Bargalló
• Benjamin Bender
• Andrea Bernasconi
• Neda Bernasconi

Artificial intelligence has recently gained popularity across different medical fields to aid in the detection of diseases based on pathology samples or medical imaging findings. Brain magnetic resonance imaging (MRI) is a key assessment tool for patients with temporal lobe epilepsy (TLE). The role of machine learning and artificial intelligence to increase detection of brain abnormalities in TLE remains inconclusive. We used support vector machine (SV) and deep learning (DL) models based on region of interest (ROI-based) structural (n = 336) and diffusion (n = 863) brain MRI data from patients with TLE with ("lesional") and without ("non-lesional") radiographic features suggestive of underlying hippocampal sclerosis from the multinational (multi-center) ENIGMA-Epilepsy consortium. Our data showed that models to identify TLE performed better or similar (68-75%) compared to models to lateralize the side of TLE (56-73%, except structural-based) based on diffusion data with the opposite pattern seen for structural data (67-75% to diagnose vs. 83% to lateralize). In other aspects, structural and diffusion-based models showed similar classification accuracies. Our classification models for patients with hippocampal sclerosis were more accurate (68-76%) than models that stratified non-lesional patients (53-62%). Overall, SV and DL models performed similarly with several instances in which SV mildly outperformed DL. We discuss the relative performance of these models with ROI-level data and the implications for future applications of machine learning and artificial intelligence in epilepsy care.

YNICL Journal 2019 Journal Article

Fractal dimension of cerebral white matter: A consistent feature for prediction of the cognitive performance in patients with small vessel disease and mild cognitive impairment

• Leonardo Pantoni
• Chiara Marzi
• Anna Poggesi
• Antonio Giorgio
• Nicola De Stefano
• Mario Mascalchi
• Domenico Inzitari
• Emilia Salvadori

Patients with cerebral small vessel disease (SVD) frequently show decline in cognitive performance. However, neuroimaging in SVD patients discloses a wide range of brain lesions and alterations so that it is often difficult to understand which of these changes are the most relevant for cognitive decline. It has also become evident that visually-rated alterations do not fully explain the neuroimaging correlates of cognitive decline in SVD. Fractal dimension (FD), a unitless feature of structural complexity that can be computed from high-resolution T1-weighted images, has been recently applied to the neuroimaging evaluation of the human brain. Indeed, white matter (WM) and cortical gray matter (GM) exhibit an inherent structural complexity that can be measured through the FD. In our study, we included 64 patients (mean age ± standard deviation, 74. 6 ± 6. 9, education 7. 9 ± 4. 2 years, 53% males) with SVD and mild cognitive impairment (MCI), and a control group of 24 healthy subjects (mean age ± standard deviation, 72. 3 ± 4. 4 years, 50% males). With the aim of assessing whether the FD values of cerebral WM (WM FD) and cortical GM (GM FD) could be valuable structural predictors of cognitive performance in patients with SVD and MCI, we employed a machine learning strategy based on LASSO (least absolute shrinkage and selection operator) regression applied on a set of standard and advanced neuroimaging features in a nested cross-validation (CV) loop. This approach was aimed at 1) choosing the best predictive models, able to reliably predict the individual neuropsychological scores sensitive to attention and executive dysfunctions (prominent features of subcortical vascular cognitive impairment) and 2) identifying a features ranking according to their importance in the model through the assessment of the out-of-sample error. For each neuropsychological test, using 1000 repetitions of LASSO regression and 5000 random permutations, we found that the statistically significant models were those for the Montreal Cognitive Assessment scores (p-value =. 039), Symbol Digit Modalities Test scores (p-value =. 039), and Trail Making Test Part A scores (p-value =. 025). Significant prediction of these scores was obtained using different sets of neuroimaging features in which the WM FD was the most frequently selected feature. In conclusion, we showed that a machine learning approach could be useful in SVD research field using standard and advanced neuroimaging features. Our study results raise the possibility that FD may represent a consistent feature in predicting cognitive decline in SVD that can complement standard imaging.

YNICL Journal 2019 Journal Article

Relevance of brain lesion location for cognition in vascular mild cognitive impairment

• Antonio Giorgio
• Ilaria Di Donato
• Alessandro De Leucio
• Jian Zhang
• Emilia Salvadori
• Anna Poggesi
• Stefano Diciotti
• Mirco Cosottini

BACKGROUND: Vascular mild cognitive impairment (VMCI) is a potentially transitional state between normal aging and vascular dementia. The presence of macroscopic white matter lesions (WML) of moderate or severe extension on brain MRI is the hallmark of the VMCI. OBJECTIVE: To assess the clinical relevance of the frequency of WML in patients with VMCI independently of total lesion volume (LV). METHODS: In this multicenter study, we included 110 patients with VMCI (age: 74.3 ± 6.6 years; sex: 60 women). Cognitive assessment was performed with the VMCI-Tuscany Neuropsychological Battery, which allowed to identify four VMCI groups: amnestic single (n = 9) and multi-domain (n = 76), non-amnestic single- (n = 10) and multi-domain (n = 15). Distribution and frequency of WML on MRI FLAIR images were evaluated with lesion probability map (LPM). Voxelwise statistics was performed with nonparametric permutation tests, controlling for age, sex, slice thickness, center, magnetic field strength, total LV and head size (p < .01, family-wise error-corrected for multiple comparisons across space). RESULTS: LPM of the WML had a fairly symmetric and widespread distribution across brain. A higher frequency of WML along association tracts of the WM such as inferior longitudinal fascicle, inferior fronto-occipital fascicle and superior longitudinal fascicle, was correlated with worst cognitive scores at the Trail Making Test Part A and Copy of the Rey-Osterrieth Complex Figure. The non-amnestic groups showed a higher frequency of WML in the anterior cingulum and superior longitudinal fascicle close to the frontal gyrus. CONCLUSIONS: Our study showed that in patients with VMCI, independently of total LV, the higher frequency of lesions along association tracts of the WM, which mediate intrahemispheric long-range connectivity, is related with psychomotor speed and constructional praxis. Moreover, a prevalence of lesions in the frontal WM seems to characterize VMCI patients with involvement of non-amnestic domains.

JBHI Journal 2016 Journal Article

Prediction of Impaired Performance in Trail Making Test in MCI Patients With Small Vessel Disease Using DTI Data

• Stefano Ciulli
• Luca Citi
• Emilia Salvadori
• Raffaella Valenti
• Anna Poggesi
• Domenico Inzitari
• Mario Mascalchi
• Nicola Toschi

Mild cognitive impairment (MCI) is a common condition in patients with diffuse hyperintensities of cerebral white matter (WM) in T2-weighted magnetic resonance images and cerebral small vessel disease (SVD). In MCI due to SVD, the most prominent feature of cognitive impairment lies in degradation of executive functions, i. e. , of processes that supervise the organization and execution of complex behavior. The trail making test is a widely employed test sensitive to cognitive processing speed and executive functioning. MCI due to SVD has been hypothesized to be the effect of WM damage, and diffusion tensor imaging (DTI) is a well-established technique for in vivo characterization of WM. We propose a machine learning scheme tailored to 1) predicting the impairment in executive functions in patients with MCI and SVD, and 2) examining the brain substrates of this impairment. We employed data from 40 MCI patients with SVD and created feature vectors by averaging mean diffusivity (MD) and fractional anisotropy maps within 50 WM regions of interest. We trained support vector machines (SVMs) with polynomial as well as radial basis function kernels using different DTI-derived features while simultaneously optimizing parameters in leave-one-out nested cross validation. The best performance was obtained using MD features only and linear kernel SVMs, which were able to distinguish an impaired performance with high sensitivity (72. 7%-89. 5%), specificity (71. 4%-83. 3%), and accuracy (77. 5%-80. 0%). While brain substrates of executive functions are still debated, feature ranking confirm that MD in several WM regions, not limited to the frontal lobes, are truly predictive of executive functions.

YNIMG Journal 2008 Journal Article

Brain white matter damage in SCA1 and SCA2. An in vivo study using voxel-based morphometry, histogram analysis of mean diffusivity and tract-based spatial statistics

• Riccardo Della Nave
• Andrea Ginestroni
• Carlo Tessa
• Elena Salvatore
• Domenico De Grandis
• Rosaria Plasmati
• Fabrizio Salvi
• Giuseppe De Michele

Background and purpose Neurodegeneration in spinocerebellar ataxia type 1(SCA1) and 2(SCA2) is associated with white matter(WM) damage. Voxel-Based Morphometry(VBM), histogram analysis of mean diffusivity(MD) and Tract-Based Spatial Statistics(TBSS) enable an in vivo quantitative analysis of WM volume and structure. We assessed with these 3 techniques the whole brain WM damage in SCA1 and SCA2. Patients and methods Ten patients with SCA1, 10 patients with SCA2 and 10 controls underwent MRI with acquisition of T1-weighted and diffusion tensor images. The results were correlated with severity of clinical deficit. Results VBM showed atrophy of the brainstem and cerebellar WM without significant differences between SCA1 and SCA2. Focal atrophy of the cerebral subcortical WM was also present. Histogram analysis revealed increased MD in the brainstem and cerebellum in patients with SCA1 and SCA2 which in SCA2 was more pronounced and combined with mild increase of the MD in the cerebral hemispheres in SCA2. In SCA1 and SCA2 TBSS revealed decreased fractional anisotropy(FA) in the inferior, middle and superior cerebellar peduncles, pontine transverse fibres, medial and lateral lemnisci, spinothalamic tracts, corticospinal tracts and corpus callosum. The extent of tract changes was greater in SCA2 patients who also showed decreased FA in the short intracerebellar tracts. In both diseases VBM, histogram and TBSS results correlated with clinical severity. Conclusions Brain WM damage featuring a pontocerebeellar atrophy is similar in SCA1 and SCA2 but more pronounced in SCA2. In both diseases it correlates with severity of the clinical deficit.

YNIMG Journal 2008 Journal Article

Brain white matter tracts degeneration in Friedreich ataxia. An in vivo MRI study using tract-based spatial statistics and voxel-based morphometry

• Riccardo Della Nave
• Andrea Ginestroni
• Carlo Tessa
• Elena Salvatore
• Ilaria Bartolomei
• Fabrizio Salvi
• Maria Teresa Dotti
• Giuseppe De Michele

Background and purpose: Neuropathological examination in Friedreich ataxia (FRDA) reveals neuronal loss in the gray matter (GM) nuclei and degeneration of the white matter (WM) tracts in the spinal cord, brainstem and cerebellum, while the cerebral hemispheres are substantially spared. Tract-based spatial statistics (TBSS) enables an unbiased whole-brain quantitative analysis of the fractional anisotropy (FA) and mean diffusivity (MD) of the brain WM tracts in vivo. Patients and methods: We assessed with TBSS 14 patients with genetically confirmed FRDA and 14 age- and sex-matched healthy controls who were also examined with voxel-based morphometry (VBM) to assess regional atrophy of the GM and WM. Results: TBSS revealed decreased FA in the inferior and superior cerebellar peduncles and the corticospinal tracts in the medullary pyramis, in WM tracts of the right cerebellar hemisphere and in the right occipito-frontal and inferior longitudinal fasciculi. Increased MD was observed in the superior cerebellar peduncles, deep cerebellar WM, posterior limbs of the internal capsule and retrolenticular area, bilaterally, and in the WM underlying the left central sulcus. Decreased FA in the left superior cerebellar peduncle correlated with clinical severity. VBM showed small symmetric areas of loss of bulk of the peridentate WM which also correlated with clinical severity. Conclusions: TBSS enables in vivo demonstration of degeneration of the brainstem and cerebellar WM tracts which neuropathological examination indicates to be specifically affected in FRDA. TBSS complements VBM and might be a more sensitive tool to detect WM structural changes in degenerative diseases of the CNS.

YNIMG Journal 2007 Journal Article

Self-paced frequency of a simple motor task and brain activation

• Stefano Diciotti
• Cinzia Gavazzi
• Riccardo Della Nave
• Enrico Boni
• Andrea Ginestroni
• Lorenzo Paoli
• Paolo Cecchi
• Nicola De Stefano

Application of fMRI to clinical neurology implies the selection of a simple task and control of the task performance. The capability to objectively monitor variables related to task execution is, therefore, important and could improve accuracy of clinical fMRI studies. We assessed the influence of different self-paced frequencies of a simple motor task on brain activation in healthy subjects. A device was developed to measure the force exerted by a subject in pressing an air-filled rubber bulb with the last four fingers of the dominant hand. The task frequency was determined by analysis of the force signal. Nine healthy subjects performed twice the task with self-paced slow (0. 35±0. 09 Hz), intermediate (0. 58±0. 21 Hz) or fast (0. 98±0. 32 Hz) frequency. The device revealed impaired task execution in 1 subject. The coefficient of variation of frequency was 8. 7% for slow, 12. 2% for intermediate and 15. 8% for fast paced task. No significant differences were found comparing the activation maps obtained at slow, intermediate and fast frequencies in the contralateral sensorimotor cortex and ipsilateral cerebellum. Cluster reproducibility was good for location (standard deviation≤7. 3 mm), but poor for signal intensity (coefficient of variation 0–176. 8%) and extent (coefficient of variation 1. 9–140. 6%). In conclusion, self-paced frequency variations of a simple motor task in the 0. 2–2 Hz range are not a relevant source of the variability of the fMRI results in healthy subjects. Use of the device for evaluation of the neurologically impaired patients might broaden the clinical applications of fMRI.

YNIMG Journal 2004 Journal Article

ADC mapping of neurodegeneration in the brainstem and cerebellum of patients with progressive ataxias

• Riccardo Della Nave
• Silvia Foresti
• Carlo Tessa
• Marco Moretti
• Andrea Ginestroni
• Cinzia Gavazzi
• Laura Guerrini
• Fabrizio Salvi

Analysis of the apparent diffusion coefficient (ADC) maps derived from diffusion-weighted MR imaging is emerging as a reproducible, sensitive, and quantitative tool to evaluate brain damage in diseases of the white and gray matter. To explore the potentials of ADC maps analysis in degenerative ataxias, we examined 28 patients and 26 age-matched controls with T1, T2, and diffusion (b values 0–1000 along the three main body axes)-weighted MR images. Twenty-four patients had inherited genetically proven diseases including spinocerebellar ataxia type 1 (SCA1) (n = 9), spinocerebellar ataxia type 2 (SCA2) (n = 8), and Friedreich's ataxia (FA) (n = 7), whereas four patients had sporadic adult onset pure cerebellar ataxia (three idiopathic, one gluten intolerance). Area and linear measurements of the CNS structures contained in the posterior cranial fossa (PCF) preliminary enabled classification of the patients in the three morphological categories reflecting the gross pathology findings, namely olivopontocerebellar atrophy (OPCA) (n = 10: six SCA2 and four SCA1), spinal atrophy (SA) (n = 7: all FA), and cortical cerebellar atrophy (CCA) (n = 4: three idiopathic and one gluten intolerance). Seven patients with SCA1 (n = 5) or SCA2 (n = 2) had morphologic changes reminiscent of OPCA, but their values were still in the lower normal range and were classified as undefined. Mean diffusivity (D̄) maps of the entire brain were generated and D̄ was measured with regions of interest (ROI) in the medulla, pons, middle cerebellar peduncles, and the peridentate white matter. Moreover, after exclusion of the skull with manual segmentation and of the CSF with application of a threshold value, histograms were obtained for D̄ of the brainstem and cerebellum and for D̄ of the cerebral hemispheres. As compared to controls, a (P < 0. 001) increase of D̄ was observed in the medulla, middle cerebellar peduncles, and peridentate white matter in OPCA and undefined patients groups who had also significantly increased values of the 25th and 50th percentiles in the brainstem and cerebellum D̄ histogram. In CCA (P = 0. 01), an increase of the 25th and 50th percentile of the D̄ value was observed in the brainstem and cerebellum histograms. The SA group showed (P < 0. 001) an increased D̄ in the medulla only. A correlation between clinical severity as assessed with the Inherited Ataxias Clinical Rating Scale (IACRS) and the 50th percentile of the D̄ value in the brainstem and cerebellum histogram (r = 0. 69) was observed in patients with SCA1 or SCA2. Diffusion MR imaging reveals variable patterns of increase of D̄ in the brainstem, cerebellum, and cerebral hemispheres in degenerative ataxias that match the known distribution of the neuropathological changes.

YNIMG Journal 2003 Journal Article

A functional magnetic resonance imaging study of patients with secondary progressive multiple sclerosis

• Maria A Rocca
• Cinzia Gavazzi
• Domenico M Mezzapesa
• Andrea Falini
• Bruno Colombo
• Mario Mascalchi
• Giuseppe Scotti
• Giancarlo Comi

Although several functional magnetic resonance imaging (fMRI) studies have shown adaptive cortical changes in patients with early multiple sclerosis (MS), the presence of brain plasticity and its role in limiting the functional consequences of brain tissue damage in patients with secondary progressive (SP) MS have not been fully investigated yet. In this study, we assessed the movement-associated brain pattern of cortical activations in patients with SPMS and investigated whether the extent of cortical brain activations is correlated with the extent of brain structural changes. From 13 right-handed SPMS patients and 15 sex- and age-matched healthy volunteers, we obtained: (a) brain dual-echo scans; (b) brain mean diffusivity and fractional anisotropy maps of the normal-appearing white (NAWM) and gray matter (NAGM); (c) fMRI during the performance of simple motor tasks [flexion–extension of the last four fingers of the right hand (task 1) and flexion–extension of the right foot (task 2)]. Compared to healthy volunteers, during task 1 performance, SPMS patients showed more significant activations of the ipsilateral inferior frontal gyrus, middle frontal gyrus, bilaterally, and contralateral intraparietal sulcus. During task 2 performance, SPMS patients had more significant activations of the contralateral primary sensorimotor cortex and thalamus and of the ipsilateral upper bank of sylvian fessure. For both tasks, strong correlations (r values ranging from −0. 83 to 0. 88) were found between relative activations of cortical areas of the motor network and the severity of structural changes of the NAWM and NAGM. This study demonstrates that cortical plasticity does occur in patients with SPMS and that it might have a role in limiting the clinical impact of MS-related damage. It also suggests that, in these patients, functional abilities are sustained by increased recruitment of highly specialized cortical areas.

YNIMG Journal 2001 Journal Article

An fMRI study of the role of noise in image processing by the human perceptive system

• Michele Piana
• Massimo Riani
• Michela Canfora
• Cinzia Gavazzi
• Marco Moretti
• Gianni Pellicano
• Mario Mascalchi