Author name cluster

Adriano Chiò

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

5 papers

1 author row

JBHI Journal 2024 Journal Article

DYNAMITE: Integrating Archetypal Analysis and Process Mining for Interpretable Disease Progression Modelling

Isotta Trescato
Erica Tavazzi
Martina Vettoretti
Roberto Gatta
Rosario Vasta
Adriano Chiò
Barbara Di Camillo

DYNAMITE, an acronym for DYNamic Archetypal analysis for MIning disease TrajEctories, is a new methodology developed specifically to model disease progression by exploiting information available in longitudinal clinical datasets. First, archetypal analysis is applied to data organised in matrix form, with the aim of finding extreme and representative disease states (archetypes) linked to the original data through convex coefficients. Then, each original observation is associated with a single archetype based on their similarity; finally, an event log is created encoding the progression of disease states for each patient in terms of archetype states. In the last stage of the procedure, archetypal analysis is coupled with process mining, which allows the event log archetypes to be visualised graphically as sequences of disease states, allowing the clinical trajectories of patients to be extracted and examined. As a proof of concept, we applied the proposed method to data from a cohort of amyotrophic lateral sclerosis patients whose progression was monitored using the 12-item ALSFRS-R questionnaire. Without any a priori knowledge, DYNAMITE identified six archetypes clearly describing different types and severity of impairment and provided reliable clinical trajectories consistent with the prognosis of amyotrophic lateral sclerosis patients. DYNAMITE offers high interpretability at every stage of the analysis, which makes it particularly suitable for use in healthcare where explainability is paramount, and enables analysis of clinical trajectories at both individual and population levels.

Details DOI

AIIM Journal 2023 Journal Article

Artificial intelligence and statistical methods for stratification and prediction of progression in amyotrophic lateral sclerosis: A systematic review

Erica Tavazzi
Enrico Longato
Martina Vettoretti
Helena Aidos
Isotta Trescato
Chiara Roversi
Andreia S. Martins
Eduardo N. Castanho

Background: Amyotrophic Lateral Sclerosis (ALS) is a fatal neurodegenerative disorder characterised by the progressive loss of motor neurons in the brain and spinal cord. The fact that ALS’s disease course is highly heterogeneous, and its determinants not fully known, combined with ALS’s relatively low prevalence, renders the successful application of artificial intelligence (AI) techniques particularly arduous. Objective: This systematic review aims at identifying areas of agreement and unanswered questions regarding two notable applications of AI in ALS, namely the automatic, data-driven stratification of patients according to their phenotype, and the prediction of ALS progression. Differently from previous works, this review is focused on the methodological landscape of AI in ALS. Methods: We conducted a systematic search of the Scopus and PubMed databases, looking for studies on data-driven stratification methods based on unsupervised techniques resulting in (A) automatic group discovery or (B) a transformation of the feature space allowing patient subgroups to be identified; and for studies on internally or externally validated methods for the prediction of ALS progression. We described the selected studies according to the following characteristics, when applicable: variables used, methodology, splitting criteria and number of groups, prediction outcomes, validation schemes, and metrics. Results: Of the starting 1604 unique reports (2837 combined hits between Scopus and PubMed), 239 were selected for thorough screening, leading to the inclusion of 15 studies on patient stratification, 28 on prediction of ALS progression, and 6 on both stratification and prediction. In terms of variables used, most stratification and prediction studies included demographics and features derived from the ALSFRS or ALSFRS-R scores, which were also the main prediction targets. The most represented stratification methods were K-means, and hierarchical and expectation-maximisation clustering; while random forests, logistic regression, the Cox proportional hazard model, and various flavours of deep learning were the most widely used prediction methods. Predictive model validation was, albeit unexpectedly, quite rarely performed in absolute terms (leading to the exclusion of 78 eligible studies), with the overwhelming majority of included studies resorting to internal validation only. Conclusion: This systematic review highlighted a general agreement in terms of input variable selection for both stratification and prediction of ALS progression, and in terms of prediction targets. A striking lack of validated models emerged, as well as a general difficulty in reproducing many published studies, mainly due to the absence of the corresponding parameter lists. While deep learning seems promising for prediction applications, its superiority with respect to traditional methods has not been established; there is, instead, ample room for its application in the subfield of patient stratification. Finally, an open question remains on the role of new environmental and behavioural variables collected via novel, real-time sensors.

Details DOI

YNICL Journal 2023 Journal Article

C9orf72 ALS mutation carriers show extensive cortical and subcortical damage compared to matched wild-type ALS patients

Anna Nigri
Manera Umberto
Mario Stanziano
Stefania Ferraro
Davide Fedeli
Jean Paul Medina Carrion
Sara Palermo
Laura Lequio

OBJECTIVE: C9orf72 mutation carriers with different neurological phenotypes show cortical and subcortical atrophy in multiple different brain regions, even in pre-symptomatic phases. Despite there is a substantial amount of knowledge, small sample sizes, clinical heterogeneity, as well as different choices of image analysis may hide anatomical abnormalities that are unique to amyotrophic lateral sclerosis (ALS) patients with this genotype or that are indicative of the C9orf72-specific trait overlain in fronto-temporal dementia patients. METHODS: Brain structural and resting state functional magnetic imaging was obtained in 24 C9orf72 positive (ALSC9+) ALS patients paired for burden disease with 24 C9orf72 negative (ALSC9-) ALS patients. A comprehensive structural evaluation of cortical thickness and subcortical volumes between ALSC9+ and ALSC9- patients was performed while a region of interest (ROI)-ROI analysis of functional connectivity was implemented to assess functional alterations among abnormal cortical and subcortical regions. Results were corrected for multiple comparisons. RESULTS: Compared to ALSC9- patients, ALSC9+ patients exhibited extensive disease-specific patterns of thalamo-cortico-striatal atrophy, supported by functional alterations of the identified abnormal regions. Cortical thinning was most pronounced in posterior areas and extended to frontal regions. Bilateral atrophy of the mediodorsal and pulvinar nuclei was observed, emphasizing a focal rather than global thalamus atrophy. Volume loss in a large portion of bilateral caudate and left putamen was reported. The marked reduction of functional connectivity observed between the left posterior thalamus and almost all the atrophic cortical regions support the central role of the thalamus in the pathogenic mechanism of C9orf72-mediated disease. CONCLUSIONS: These findings constitute a coherent and robust picture of ALS patients with C9orf72-mediated disease, unveiling a specific structural and functional characterization of thalamo-cortico-striatal circuit alteration. Our study introduces new evidence in the characterization of the pathogenic mechanisms of C9orf72 mutation.

Details DOI

YNICL Journal 2020 Journal Article

Lifetime sport practice and brain metabolism in Amyotrophic Lateral Sclerosis

Antonio Canosa
Fabrizio D'Ovidio
Andrea Calvo
Cristina Moglia
Umberto Manera
Maria Claudia Torrieri
Rosario Vasta
Angelina Cistaro

OBJECTIVE: F-FDG-PET. METHODS: F-FDG-PET. Exposure to sports was expressed as MET (Metabolic Equivalent of Task). We considered only regular practice (at least 2 h/week, for at least three months). We compared brain metabolism between two groups: subjects who did not report regular sport practice during life (N-group) and patients who did (Y-group). The resulting significant clusters were used in each group as seed regions in an interregional correlation analysis (IRCA) to evaluate the impact of lifetime sport practice on brain networks typically involved by the neurodegenerative process of ALS. Each group was compared to healthy controls (HC, n = 40). RESULTS: We found a significant, relative cerebellar hypermetabolism in the N-group compared to the Y-group. The metabolism of such cerebellar cluster resulted correlated to more significant and widespread metabolic changes in areas known to be affected by ALS (i.e. frontotemporal regions and corticospinal tracts) in the N-group as compared to the Y-group, despite the same level of disability as expressed by the ALS FRS-R. Such findings resulted independent of age, sex, site of onset (bulbar/spinal), presence/absence of C9ORF72 expansion, cognitive status and physical activity related to hobbies and occupations. When compared to HC, the N-group showed more widespread metabolic changes than the Y-group in cortical regions known to be relatively hypometabolic in ALS patients as compared to HC. CONCLUSIONS: We hypothesize that patients of the N-group might cope better with the neurodegenerative process, since they show more widespread metabolic changes as compared to the Y-group, despite the same level of disability. Nevertheless, further studies are necessary to corroborate this hypothesis.

Details DOI

YNICL Journal 2017 Journal Article

Multimodal structural MRI in the diagnosis of motor neuron diseases

Pilar M. Ferraro
Federica Agosta
Nilo Riva
Massimiliano Copetti
Edoardo Gioele Spinelli
Yuri Falzone
Gianni Sorarù
Giancarlo Comi

This prospective study developed an MRI-based method for identification of individual motor neuron disease (MND) patients and test its accuracy at the individual patient level in an independent sample compared with mimic disorders. 123 patients with amyotrophic lateral sclerosis (ALS), 44 patients with predominantly upper motor neuron disease (PUMN), 20 patients with ALS-mimic disorders, and 78 healthy controls were studied. The diagnostic accuracy of precentral cortical thickness and diffusion tensor (DT) MRI metrics of corticospinal and motor callosal tracts were assessed in a training cohort and externally proved in a validation cohort using a random forest analysis. In the training set, precentral cortical thickness showed 0. 86 and 0. 89 accuracy in differentiating ALS and PUMN patients from controls, while DT MRI distinguished the two groups from controls with 0. 78 and 0. 92 accuracy. In ALS vs controls, the combination of cortical thickness and DT MRI metrics (combined model) improved the classification pattern (0. 91 accuracy). In the validation cohort, the best accuracy was reached by DT MRI (0. 87 and 0. 95 accuracy in ALS and PUMN vs mimic disorders). The combined model distinguished ALS and PUMN patients from mimic syndromes with 0. 87 and 0. 94 accuracy. A multimodal MRI approach that incorporates motor cortical and white matter alterations yields statistically significant improvement in accuracy over using each modality separately in the individual MND patient classification. DT MRI represents the most powerful tool to distinguish MND from mimic disorders.

Details DOI