A5078930083- Colorectal Cancer Treatments and Studies
- Cancer, Hypoxia, and Metabolism
- Cancer-related Molecular Pathways
- Big Data and Business Intelligence
- Information Systems Theories and Implementation
- Protein Kinase Regulation and GTPase Signaling
- Retinal Diseases and Treatments
- Data-Driven Disease Surveillance
- Anomaly Detection Techniques and Applications
- Information Technology Governance and Strategy
- Melanoma and MAPK Pathways
- Retinal Imaging and Analysis
- COVID-19 epidemiological studies
- Service-Oriented Architecture and Web Services
- Collaboration in agile enterprises
- Cancer therapeutics and mechanisms
- Genomics and Phylogenetic Studies
- Machine Learning and Data Classification
- Complex Systems and Decision Making
- Mass Spectrometry Techniques and Applications
- Artificial Intelligence in Healthcare
- Machine Learning in Healthcare
- Glaucoma and retinal disorders
- Advanced Proteomics Techniques and Applications
- COVID-19 diagnosis using AI
Queen's University Belfast
2020-2024
Polytechnic Institute of Coimbra
2024
i3S - Instituto de Investigação e Inovação em Saúde, Universidade do Porto
2015-2018
Instituto de Biologia Molecular e Celular
2010-2018
Universidade do Porto
2014-2015
Institute for Biotechnology and Bioengineering
2010-2014
Institut de Biologie Moléculaire et Cellulaire
2011
PurposeThe current study aimed to identify metabolites associated with age-related macular degeneration (AMD) by performing the largest metabolome association analysis in AMD date, as well aiming determine effect of AMD-associated genetic variants on metabolite levels and investigate associations between identified activity complement system, one main disease pathways.DesignCase-control metabolomics data.ParticipantsFive European cohorts consisting 2267 patients 4266 control...
Geographic atrophy (GA) secondary to age-related macular degeneration is considered a single entity. This study aimed determine whether GA subgroups exist that can be defined by their genotype and phenotype.Retrospective analysis of cross-sectional data.Individuals (196 eyes 196 patients) 50 years age or older with from the EYE-RISK database.Participants were graded for presence each following fundus features on color photography: large soft drusen, reticular pseudodrusen (RPD), refractile...
The availability of large datasets is providing the impetus for driving many current artificial intelligent developments. However, specific challenges arise in developing solutions that exploit small datasets, mainly due to practical and cost-effective deployment issues, as well opacity deep learning models. To address this, Comprehensive Abstraction Classification Tool Uncovering Structures (CACTUS) presented a means improving secure analytics by effectively employing explainable...
Uncovering information from well-defined organizations for social network analysis is straightforward, but such analyses of networks have not led to valid predictions about their actions or stability. For dark networks, which comprise illicit drug gangs terrorists, uncovering compute a more difficult solve. The authors used new theory that based on the conservation assess and concluding are properly constrained should be invaluable bookkeeping (storing recovered neighborhood canvasses as...
The lifecycle of COVID-19 pandemic curves requires timely decisions to protect public health while minimizing the impact global economy. New models are necessary predict effect mobility suppression/reactivation at a scale. This research presents an approach understand such tensions by modeling air travel restrictions during new coronavirus outbreak. paper begins with updated review on in infectious disease progression, followed adoption complex networks based semi-supervised statistical...
Novel covalent inhibitors of KRASG12C have shown limited response rates in patients with KRASG12C-mutant (MT) colorectal cancer. Thus, novel inhibitor combination strategies that can achieve deep and durable responses are needed. Small-molecule AZ'1569 AZ'8037 were used. To identify candidate for AZ'1569, we performed RNA sequencing, siRNA, high-throughput drug screening. Top hits validated a panel KRASG12CMT cancer cells vivo. AZ'1569-resistant generated characterized. We found to was...
When changes at few amino acid sites are the target of selection, adaptive in protein sequences can be identified using maximum-likelihood methods based on models codon substitution (such as codeml). Although such have been employed numerous times a variety different organisms, time needed to collect data and prepare input files means that tens or hundreds coding regions usually analyzed. Nevertheless, recent availability flexible easy use computer applications relevant BDBM) infer...
In personalised health, small datasets with missing data are quite common. Current Machine Learning methods unable to process such in a meaningful way due the huge volume requirement. To address this problem, we propose new Small and iNcomplete Dataset Analyser (SaNDA) way. Due characteristics of these criticality domain, an explainable method is mandatory facilitate decision-making interpretation. Thus, SaNDA prioritises explainability over efficiency by design. We evaluated our proposal...
Business/IT alignment is an information systems research field with a long existence and high number of researchers represents central direction on the thinking about relation between business systems. It aims to achieve paradigm, one in which there degree visibility availability information, sociomateriality. Complex-networks constitute approach study emergent properties complex that strongly focuses relies models measures, through build system interdependence. Several contributions...
Today's companies face unprecedented challenges on information systems development and adaptation to the constantly changes surrounding environment. The majority of in use do not present a fast effective response for dynamic needs due their high level customization hard-coded logical rules. actual state technology its pervasiveness creates unique systems. Virtual organizations (VOs) architecture provides structure agility flexibility benefits while accommodating numerous commercial...
Traditional approaches to information system architecture development have a little concerns with technology interdependency. They are sustained in positivist vision of the world, where systems, can be constructed from set requirements, done by detached observer, who captures or elicits it. Organizations each more concerned about flow and knowledge availability as central element for innovation empowerment. This always relation There is clear construction technosocial whose relevance...
Organizations live in a world where interdependence, self-organization and emergence are factors for agility, adaptability flexibility plunged into networks. Software-based information systems go service oriented architecture direction the same goes to Infrastructures services become structures available Inspired empirical studies of networked such as Internet, social networks, biological researchers have recent years developed variety techniques models help us structurally understand or...
Continuous learning became a fundamental element in any organization as tool for managing development, sustainability and innovation from knowledge capabilities. In this paper, we start by describing the combination of two phenomenon', Shared Service Center (SSC) E-Learning that expand Then, produce conceptual model developing continuous organization. This conceptualizes integrates SSC organizational memory repository enhancer pull push communication process. model, mathematically deduct Δ...
The applicability of widely adopted machine learning (ML) methods to classification is circumscribed by the imperatives explicability and uncertainty, particularly evident in domains such as healthcare, behavioural sciences, finances, wherein accountability assumes priority. Recently, Small Incomplete Dataset Analyser (SaNDA) has been proposed enhance ability perform domains, developing a data abstraction protocol using ROC curve-based method. This paper focuses on column-wise...
Detailed and invasive clinical investigations are required to identify the causes of haematuria. Highly unbalanced patient population (predominantly male) a wide range potential make ability correctly classify patients patient-specific biomarkers major challenge. Studies have shown that it is possible improve diagnosis using multi-marker analysis, even in datasets, by applying advanced analytical methods. Here, we applied several machine learning algorithms from haematuria cohort (HaBio)...
Noise is a fundamental problem in learning theory with huge effects the application of Machine Learning (ML) methods, due to real world data tendency be noisy. Additionally, introduction malicious noise can make ML methods fail critically, as case adversarial attacks. Thus, finding and developing alternatives improve robustness ML. In this paper, we propose method deal noise: mitigating its effect through use abstractions. The goal reduce over model's performance loss information produced by...
Novel cancer biomarkers discoveries are driven by the application of omics technologies. The vast quantity highly dimensional data necessitates implementation feature selection. mathematical basis different selection methods varies considerably, which may influence subsequent inferences. In study, and classification were employed to identify six signature gene sets grade 2 3 astrocytoma samples from Rembrandt repository. Subsequently, impact these variables on further discovery biological...
In a commoditized world of information technology, organizations aim to achieve unique behavior, able empower its sustainability and development. Understanding that sociomaterial entanglement technology organizational users is practice research challenge in the systems community. this paper, we address trough modelling virtual organization emergent from enactment sociomateriality. This approach highlights nature enactment, contribute conceptualization cooperative environment. We develop...