A5056607122- Advanced Graph Neural Networks
- Bioinformatics and Genomic Networks
- Biomedical Text Mining and Ontologies
- Complex Network Analysis Techniques
- Graph Theory and Algorithms
- Long-Term Effects of COVID-19
- Gene expression and cancer classification
- COVID-19 Clinical Research Studies
- Computational Drug Discovery Methods
- Human Mobility and Location-Based Analysis
- Cancer, Stress, Anesthesia, and Immune Response
- Medical Imaging and Pathology Studies
- Tuberculosis Research and Epidemiology
- Legionella and Acanthamoeba research
- Breast Cancer Treatment Studies
- Imbalanced Data Classification Techniques
- Machine Learning in Healthcare
- Intensive Care Unit Cognitive Disorders
- Web Data Mining and Analysis
- Parasitic infections in humans and animals
- Emergency and Acute Care Studies
- Drug-Induced Hepatotoxicity and Protection
- Misinformation and Its Impacts
- Multiple Sclerosis Research Studies
- Radiomics and Machine Learning in Medical Imaging
University of Milan
2021-2024
Istituti di Ricovero e Cura a Carattere Scientifico
2024
Direction Générale Déléguée aux Ressources
2024
Politecnico di Milano
2020
Università degli Studi del Piemonte Orientale “Amedeo Avogadro”
2013
Stratification of patients with post-acute sequelae SARS-CoV-2 infection (PASC, or long COVID) would allow precision clinical management strategies. However, COVID is incompletely understood and characterised by a wide range manifestations that are difficult to analyse computationally. Additionally, the generalisability machine learning classification COVID-19 outcomes has rarely been tested.
Integrated, up-to-date data about SARS-CoV-2 and COVID-19 is crucial for the ongoing response to pandemic by biomedical research community. While rich biological knowledge exists related viruses (SARS-CoV, MERS-CoV), integrating this difficult time-consuming, since much of it in siloed databases or textual format. Furthermore, required community vary drastically different tasks; optimal a machine learning task, example, from used populate browsable user interface clinicians. To address these...
Between January and October of 2020, the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) virus has infected more than 34 million persons in a worldwide pandemic leading to over one deaths (data from Johns Hopkins University). Since begun spread, emergency departments were busy with COVID-19 patients for whom quick decision regarding in- or outpatient care was required. The can cause characteristic abnormalities chest radiographs (CXR), but, due low sensitivity CXR, additional...
Bridging the gap between genetic variations, environmental determinants, and phenotypic outcomes is critical for supporting clinical diagnosis understanding mechanisms of diseases. It requires integrating open data at a global scale. The Monarch Initiative advances these goals by developing ontologies, semantic models, knowledge graphs translational research. App an integrated platform combining about genes, phenotypes, diseases across species. Monarch's APIs enable access to carefully...
Non-steroidal anti-inflammatory drugs (NSAIDs) are commonly used to reduce pain, fever, and inflammation but have been associated with complications in community-acquired pneumonia. Observations shortly after the start of COVID-19 pandemic 2020 suggested that ibuprofen was an increased risk adverse events patients, subsequent observational studies failed demonstrate one case showed reduced NSAID use.
Abstract Graph representation learning methods opened new avenues for addressing complex, real-world problems represented by graphs. However, many graphs used in these applications comprise millions of nodes and billions edges are beyond the capabilities current software implementations. We present GRAPE (Graph Representation Learning, Prediction Evaluation), a resource graph processing embedding that is able to scale with big using specialized smart data structures, algorithms, fast...
Abstract Accurate stratification of patients with post-acute sequelae SARS-CoV-2 infection (PASC, or long COVID) would allow precision clinical management strategies. However, the natural history COVID is incompletely understood and characterized by an extremely wide range manifestations that are difficult to analyze computationally. In addition, generalizability machine learning classification COVID-19 outcomes has rarely been tested. We present a method for computationally modeling PASC...
Abstract Background Non-steroidal anti-inflammatory drugs (NSAIDs) are commonly used to reduce pain, fever, and inflammation but have been associated with complications in community-acquired pneumonia. Observations shortly after the start of COVID-19 pandemic 2020 suggested that ibuprofen was an increased risk adverse events patients, subsequent observational studies failed demonstrate one case showed reduced NSAID use. Methods A 38-center retrospective cohort study performed leveraged...
Missing data imputation has been a hot topic in the past decade, and many state-of-the-art works have presented to propose novel, interesting solutions that applied variety of fields. In successful results achieved by deep learning techniques opened way their application for solving difficult problems where human skill is not able provide reliable solution. Not surprisingly, some learners, mainly exploiting encoder-decoder architectures, also designed task missing imputation. However, most...
Integrated, up-to-date data about SARS-CoV-2 and coronavirus disease 2019 (COVID-19) is crucial for the ongoing response to COVID-19 pandemic by biomedical research community. While rich biological knowledge exists related viruses (SARS-CoV, MERS-CoV), integrating this difficult time consuming, since much of it in siloed databases or textual format. Furthermore, required community varies drastically different tasks - optimal a machine learning task, example, from used populate browsable user...
Abstract Motivation Graph representation learning is a family of related approaches that learn low-dimensional vector representations nodes and other graph elements called embeddings. Embeddings approximate characteristics the can be used for variety machine-learning tasks such as novel edge prediction. For many biomedical applications, partial knowledge exists about positive edges represent relationships between pairs entities, but little to no available negative explicit lack relationship...
Abstract Background The addition of pertuzumab (P) to trastuzumab (H) and standard chemotherapy (CT) as neoadjuvant treatment (NaT) for patients with HER2 + breast cancer (BC), has shown increase the pathological complete response (pCR) rate, without main safety concerns. aim NeoPowER trial is evaluate efficacy P H CT in a real–world population. Methods We retrospectively reviewed medical records stage II–III, BC treated NaT: who received (neopower group) 5 Emilia Romagna institutions were...
Abstract Graph Representation Learning methods opened new possibilities for addressing complex,real-world problems represented by graphs. However, many graphs used in these applicationscomprise millions of nodes and billions edges are beyond the capabilities current methodsand software implementations. We present GRAPE, a resource graph processing andrepresentation learning that is able to scale with big using specialized smart datastructures, algorithms, fast parallel implementation. When...
Abstract Inhibiting protein kinases (PKs) that cause cancers has been an important topic in cancer therapy for years. So far, almost 8% of >530 PKs have targeted by FDA-approved medications, and around 150 kinase inhibitors (PKIs) tested clinical trials. We present approach based on natural language processing machine learning to investigate the relations between cancers, predicting whose inhibition would be efficacious treat a certain cancer. Our represents as semantically meaningful...
The Novara 118 emergency medical system (EMS) dispatch center manages calls coming from a region that spreads out over 1,400 km2 and includes 88 towns population of 385,000 people; inhabitant density is 275 inhabitants/km2.
We report the results of a yearlong effort at Laboratory for Web Algorithmics and Inria to port WebGraph framework [4] from Java Rust. For two decades has been instrumental in analysis distribution large graphs research community TheWebConf, but intrinsic limitations Virtual Machine had become bottleneck very use cases, such as Software Heritage Merkle graph [2] with its half trillion arcs. As part this clean-slate implementation Rust, we developed few ancillary projects bringing Rust...
Integrated, up-to-date data about SARS-CoV-2 and coronavirus disease 2019 (COVID-19) is crucial for the ongoing response to COVID-19 pandemic by biomedical research community. While rich biological knowledge exists related viruses (SARS-CoV, MERS-CoV), integrating this difficult time consuming, since much of it in siloed databases or textual format. Furthermore, required community varies drastically different tasks - optimal a machine learning task, example, from used populate browsable user...
Abstract Graph representation learning is a family of related approaches that learn low-dimensional vector representations nodes and other graph elements called embeddings. Embeddings approximate characteristics the can be used for variety machine-learning tasks such as novel edge prediction. For many biomedical applications, partial knowledge exists about positive edges represent relationships between pairs entities, but little to no available negative explicit lack relationship two nodes....
ABSTRACT Inhibiting protein kinases (PKs) that cause cancers has been an important topic in cancer therapy for years. So far, almost 8% of more than 530 PKs have targeted by FDA-approved medications and around 150 kinase inhibitors (PKIs) tested clinical trials. We present approach based on natural language processing machine learning to the relations between cancers, predicting whose inhibition would be efficacious treat a certain cancer. Our represents as semantically meaningful...
This paper presents an algorithm for detecting attributed high-degree node isomorphism. High-degree isomorphic nodes seldom happen by chance and often represent duplicated entities or data processing errors. By definition, are topologically indistinguishable can be problematic in graph ML tasks. The employs a parallel, "degree-bounded" approach that fingerprints each node's local properties through hash, which constrains the search to within hash-defined buckets, thus minimising number of...