Giorgio Dolci

Mr. Giorgio Dolci

orcid

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
17 September 2025 in Journal of Neural Engineering
Verified
Journal Article
Open access
PDF file
GDGiorgio Dolci
Giorgio Dolci
FCFederica Cruciani
Federica Cruciani
ARAbdur Rahaman
Abdur Rahaman
AAAnees Abrol
Anees Abrol
JCJiayu Chen
Jiayu Chen
VCVince Calhoun
Vince Calhoun
Objective: Alzheimer’s disease (AD) is the most prevalent form of dementia worldwide, encompassing a prodromal stage known as Mild Cognitive Impairment (MCI), where patients may either progress to AD or remain stable. The objective of the work was to capture structural and functional modulations of brain structure and function relying on multimodal MRI data and Single Nucleotide Polymorphisms, also in case of missing views, with the twofold goal of classifying AD patients versus healthy controls and detecting MCI converters. Approach: We propose a multimodal DL-based classification framework where a generative module employing Cycle Generative Adversarial Networks was introduced in the latent space for imputing missing data (a common issue of multimodal approaches). Explainable AI method was then used to extract input features’ relevance allowing for post-hoc validation and enhancing the interpretability of the learned representations. Main results: Experimental results on two tasks, AD detection and MCI conversion, showed that our framework reached competitive performance in the state-of-the-art with an accuracy of 0.926 ± 0.02 (CI [0.90, 0.95]) and 0.711 ± 0.01 (CI [0.70, 0.72]) in the two tasks, respectively. The interpretability analysis revealed gray matter modulations in cortical and subcortical brain areas typically associated with AD. Moreover, impairments in sensory-motor and visual resting state networks along the disease continuum, as well as genetic mutations defining biological processes linked to endocytosis, amyloid-beta, and cholesterol, were identified. Significance: Our integrative and interpretable DL approach shows promising performance for AD detection and MCI prediction while shedding light on important biological insights.

20 March 2025 in Network Neuroscience
Verified
Journal Article
Open access
GDGiorgio Dolci
Giorgio Dolci
CECharles A. Ellis
Charles A. Ellis
FCFederica Cruciani
Federica Cruciani
LBLorenza Brusini
Lorenza Brusini
AAAnees Abrol
Anees Abrol
FIFor The Alzheimer’s Disease Neuroimaging Initiative
For The Alzheimer’s Disease Neuroimaging Initiative
Amyloid-β (Aβ) plaques in conjunction with hyperphosphorylated tau proteins in the form of neurofibrillary tangles are the two neuropathological hallmarks of Alzheimer’s disease. It is well-known that the identification of individuals with Aβ positivity could enable early diagnosis. In this work, we aim at capturing the Aβ positivity status in an unbalanced cohort enclosing subjects at different disease stages, exploiting the underlying structural and connectivity disease-induced modulations as revealed by structural, functional, and diffusion MRI. Of note, due to the unbalanced cohort, the outcomes may be guided by those factors rather than amyloid accumulation. The partial views provided by each modality are integrated in the model, allowing to take full advantage of their complementarity in encoding the effects of the Aβ accumulation, leading to an accuracy of 0.762 ± 0.04. The specificity of the information brought by each modality is assessed by post hoc explainability analysis (guided backpropagation), highlighting the underlying structural and functional changes. Noteworthy, well-established biomarker key regions related to Aβ deposition could be identified by all modalities, including the hippocampus, thalamus, precuneus, and cingulate gyrus, witnessing in favor of the reliability of the method as well as its potential in shedding light on modality-specific possibly unknown Aβ deposition signatures.

1 November 2022
Verified
Conference Paper
Closed access
GDGiorgio Dolci
Giorgio Dolci
ARAbdur Rahaman
Abdur Rahaman
JCJiayu Chen
Jiayu Chen
KDKuaikuai Duan
Kuaikuai Duan
MRMd Mahfuzur Rahman
Md Mahfuzur Rahman
VCVince Calhoun
Vince Calhoun
Alzheimer's disease (AD) is a neurodegenerative process characterized by the accumulation of amyloid-beta plaques and neurofibrillary tangles and is the most common cause of dementia. Studies have been striving to analyze the disease using available physiological and behavioral data. Functional/structural neuroimaging and genomics are complementary modalities for exploring the mechanisms subserving the development of AD. In this paper, we present a deep multimodal generative data fusion framework for integrating these sources in a classification task involving AD patients and healthy controls from the ADNI database. Biological data fusion has the potential to improve the final prediction, but at the same time, it is particularly challenging due to the unavailability of all sources of input for the entire cohort of subjects. Our proposed model allows us to perform prediction even if individuals are missing certain modalities. Our method addresses the missing modalities problem via knowledge transfer from two generative adversarial networks. The model exhibits superior performance for predicting AD versus healthy control, even with missing modalities. This could have an important impact from the patient point of view since certain clinical tests may not be necessary or available to a given individual.

1 January 2020 in European Heart Journal - Cardiovascular Imaging
Verified
Journal Article
Open access
PDF file
FPF Prevedello
F Prevedello
GDGiorgio Dolci
Giorgio Dolci
RTR M Tenaglia
R M Tenaglia
ILI Lobascio
I Lobascio
CPC Paolini
C Paolini
Claudio BilatoClaudio Bilato
Claudio Bilato
Background: Nivolumab is a monoclonal antibody targeting programmed cell death protein 1 (PD-1), with a demonstrated clinical efficacy against metastatic cancer. Immunotherapy cardiovascular toxicity is rare, but recent scientific literature demonstrated possible and different cardiovascular side effects. Nivolumab was found to be associated with fatal acute myocarditis with diffuse myocardial necrosis and heart failure. To our knowledge no case of Nivolumab induced Tako-tsubo syndrome has ever been reported. Case presentation: A 76-year-old woman was admitted to the emergency department for dyspnoea with electrocardiographic evidence of extensive anterior ST segment elevation (Figure 1A); the patient was hemodynamically unstable with increasing shortness of breath and oxygen desaturation (arterial blood pressure 90/50 mmHg, heart rate 120 beats/min, O2 Sat. 87%). She was scheduled for urgent coronary angiography which eventually demonstrated no significant coronary artery disease. The ventriculography revealed dyskinesis of the left apex and peri-apical segments and basal hyperkinesis (Figure 1B). The final diagnosis was stress cardiomyopathy or Tako-tsubo syndrome. The echocardiogram performed in the acute setting revealed severe left ventricular impairment (ejection fraction, EF = 30%) confirming the typical apical ballooning pattern of the syndrome (Figure 1C). Left ventricular outflow tract obstruction was not found. At admission, the level of TroponinT was mildly elevated (745 ng/L) and NT-proBNP was 35 000 pg/ml. The patient was treated with optimal heart failure medical therapy with gradual and complete left ventricular function recovery (EF at discharge = 55%). Of note, the patient had a history of primary lung adenocarcinoma firstly treated in 2017 with carboplatin; due to cancer progression she had undergone Nivolumab therapy in October 2018 with a total of three cycles, the last one interrupted one month before the cardiovascular event due to ipokaliemia and mild renal impairment. At that time the patient was defined as having a good functional status. Apart from the oncologic treatment, patient’s history did not reveal any "acute" stressful event. Conclusions: To the best of our knowledge this is the first Takotsubo cardiomyopathy ever detected after Nivolumab therapy. Stress cardiomyopathy is a rare but possibly severe complication that could appear in the course of oncological treatment and, in this backdrop, its aetiology is not well understood. As a matter of fact, cardiac function monitoring and strict clinical follow-up for early detection of complications is mandatory in these patients. Favourable outcomes may be achieved with early diagnosis and prompt treatment introduction. Abstract P710 Figure. Diagnostic evaluation of Takotsubo Sdr

Verified
Journal Article
Open access
PDF file
LBLorenza Brusini
Lorenza Brusini
GDGiorgio Dolci
Giorgio Dolci
LPLorenzo Pini
Lorenzo Pini
FCFederica Cruciani
Federica Cruciani
FPFabrizio Pizzagalli
Fabrizio Pizzagalli
IGIlaria Boscolo Galazzo
Ilaria Boscolo Galazzo
Bridging the gap between cortical morphometric remodeling and gene expression can help to clarify the effects of the selective brain accumulation of Amyloid-β (Aβ) and tau proteins occurring in the Alzheimer’s disease (AD). To this aim, we derived morphometric similarity (MS) networks from 126 Aβ- and tau-positive (Aβ+/tau+) and 172 Aβ−/tau− subjects, and we investigated the association between group-wise regional MS differences and transcriptional correlates thanks to an imaging transcriptomics approach grounded in the Allen Human Brain Atlas (AHBA). The expressed gene with the highest correlation with MS alterations was BCHE, a gene related to Aβ homeostasis. In addition, notably, among the most promising results derived from the enrichment analysis, we found the immune response to be a biological process and astrocytes, microglia, and oligodendrocyte precursors for the cell types. In summary, by relating cortical MS and AHBA-derived transcriptomics, we were able to retrieve findings suggesting the biological mechanisms underlying the Aβ- and tau- induced cortical MS alterations in the AD continuum.

1 January 2026 in IEEE Transactions on Medical Imaging
Verified
Journal Article
Open access
PPPavel Popov
Pavel Popov
CMCristian Morasso
Cristian Morasso
GDGiorgio Dolci
Giorgio Dolci
IGIlaria Boscolo Galazzo
Ilaria Boscolo Galazzo
GPGodfrey Pearlson
Godfrey Pearlson
SPSergey Plis
Sergey Plis
Deep learning (DL) approaches find increasingly more applications in medical imaging analysis, where they excel at predictive tasks such as diagnosis classification, but often offer little insight on the underlying mechanisms governing complex systems, a key goal of scientific inquiry. One way to bridge this gap is to design DL architectures for modeling the underlying system reflected in imaging data, which can give insights into the system through the analysis of fitted models. In this spirit, we present a novel approach for deep extraction of causally informed features via restricted architecture (DECIFRA), designed for multivariate time series modeling. DECIFRA introduces an interpretable bottleneck in its computation flow, where dynamically generated transition matrices exclusively model the directed influences between channels over time. These matrices function as both the model's internal computational mechanism and its primary, interpretable output that gives insights into the cross-channel interactions. To validate our method, we apply it to the challenging task of learning dynamic brain connectivity from functional magnetic resonance imaging (fMRI) data. We demonstrate that DECIFRA learns stable, reproducible connectivity patterns that are discriminative for group-level analysis. Moreover, these learned dynamics can be fine-tuned for specific classification tasks, improving predictive performance while retaining interpretability. Our findings show that DECIFRA captures functionally grounded brain connectivity patterns, with certain dynamics associated with age or schizophrenia classification tasks and others consistently preserved across tasks.