Michelle Cavariani Catta-Preta

Dr. Michelle Cavariani Catta-Preta

orcid

Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

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

Representative Publications

View All Publications
20 January 2025 in Future Internet
Verified
Journal Article
Open access
PDF file
JPJan Ferrer i Picó
Jan Ferrer i Picó
Michelle Cavariani Catta-PretaMichelle Cavariani Catta-Preta
Michelle Cavariani Catta-Preta
Alex Trejo OmeñacaAlex Trejo Omeñaca
Alex Trejo Omeñaca
MVMarc Vidal
Marc Vidal
JFJosep Monguet Fierro
Josep Monguet Fierro
Contemporary society faces unprecedented challenges—from rapid technological evolution to climate change and demographic tensions—compelling organisations to anticipate the future for informed decision-making. This case study aimed to design a digital system for end-users called the Time Machine, which enables a generative artificial intelligence (GAI) system to produce prospective future scenarios based on the input information automatically, proposing hypotheses and prioritising trends to streamline and make the formulation of future scenarios more accessible. The system’s design, development, and testing progressed through three versions of prompts for the OpenAI GPT-4 LLM, with six trials conducted involving 222 participants. This iterative approach allowed for gradual adjustment of instructions given to the machine and encouraged refinement. Results from the six trials demonstrated that the Time Machine is an effective tool for generating future scenarios that promote debate and stimulate new ideas in multidisciplinary teams. Our trials proved that GAI-generated scenarios could foster discussions on +70% of generated scenarios with appropriate prompting, and more than half included new ideas. In conclusion, large language models (LLMs) of GAI, with suitable prompt engineering and architecture, have the potential to generate useful future scenarios for organisations, transforming future intelligence into a more accessible and operational resource. However, critical use of these scenarios is essential.

11 November 2025 in Applied Sciences
Verified
Journal Article
Open access
PDF file
Michelle Cavariani Catta-PretaMichelle Cavariani Catta-Preta
Michelle Cavariani Catta-Preta
Alex Trejo OmeñacaAlex Trejo Omeñaca
Alex Trejo Omeñaca
JPJan Ferrer i Picó
Jan Ferrer i Picó
JFJosep Monguet Fierro
Josep Monguet Fierro
Conventional innovation management methodologies (IMMs) often struggle to respond to the complexity, uncertainty, and cognitive diversity that characterise contemporary innovation projects. This study introduces Innovation Flow (IF), a human-centred and adaptive framework grounded in Flow Theory and enhanced by Generative Artificial Intelligence (GenAI). At its core, IF operationalises Personalised Innovation Techniques (PInnTs)—adaptive variations of established methods tailored to project genetics and team profiles, generated dynamically through a GenAI-based system. Unlike traditional IMMs that rely on static toolkits and expert facilitation, Innovation Flow (IF) introduces a dynamic, GenAI-enhanced system capable of tailoring techniques in real time to each project’s characteristics and team profile. This adaptive model achieved a 60% reduction in ideation and prototyping time while maintaining high creative performance and autonomy. IF thus bridges the gap between human-centred design and AI augmentation, providing a scalable, personalised, and more inclusive pathway for managing innovation. Using a mixed-methods design that combines grounded theory with quasi-experimental validation, the framework was tested in 28 innovation projects across healthcare, manufacturing, and education. Findings show that personalisation improves application fidelity, engagement, and resilience, with 87% of cases achieving high efficacy. GenAI integration accelerated ideation and prototyping by more than 60%, reduced dependence on expert facilitators, and broadened participation by lowering the expertise barrier. Qualitative analyses emphasised the continuing centrality of human agency, as the most effective teams critically adapted rather than passively adopted AI outputs. The research establishes IF as a scalable methodology that augments, rather than replaces, human creativity, accelerating innovation cycles while reinforcing motivation and autonomy.

28 May 2025 in Computers
Verified
Journal Article
Open access
PDF file
Alex Trejo OmeñacaAlex Trejo Omeñaca
Alex Trejo Omeñaca
ELEsteve Llargués
Esteve Llargués
JSJonny Sloan
Jonny Sloan
Michelle Cavariani Catta-PretaMichelle Cavariani Catta-Preta
Michelle Cavariani Catta-Preta
JPJan Ferrer i Picó
Jan Ferrer i Picó
BGBeatriu Bayes Genis
Beatriu Bayes Genis
Clinical documentation, particularly the hospital discharge report (HDR), is essential for ensuring continuity of care, yet its preparation is time-consuming and places a considerable clinical and administrative burden on healthcare professionals. Recent advancements in Generative Artificial Intelligence (GenAI) and the use of prompt engineering in large language models (LLMs) offer opportunities to automate parts of this process, improving efficiency and documentation quality while reducing administrative workload. This study aims to design a digital system based on LLMs capable of automatically generating HDRs using information from clinical course notes and emergency care reports. The system was developed through iterative cycles, integrating various instruction flows and evaluating five different LLMs combined with prompt engineering strategies and agent-based architectures. Throughout the development, more than 60 discharge reports were generated and assessed, leading to continuous system refinement. In the production phase, 40 pneumology discharge reports were produced, receiving positive feedback from physicians, with an average score of 2.9 out of 4, indicating the system’s usefulness, with only minor edits needed in most cases. The ongoing expansion of the system to additional services and its integration within a hospital electronic system highlights the potential of LLMs, when combined with effective prompt engineering and agent-based architectures, to generate high-quality medical content and provide meaningful support to healthcare professionals. Hospital discharge reports (HDRs) are pivotal for continuity of care but consume substantial clinician time. Generative AI systems based on large language models (LLMs) could streamline this process, provided they deliver accurate, multilingual, and workflow-compatible outputs. We pursued a three-stage, design-science approach. Proof-of-concept: five state-of-the-art LLMs were benchmarked with multi-agent prompting to produce sample HDRs and define the optimal agent structure. Prototype: 60 HDRs spanning six specialties were generated and compared with clinician originals using ROUGE with average scores compatible with specialized news summarizing models in Spanish and Catalan (lower scores). A qualitative audit of 27 HDR pairs showed recurrent divergences in medication dose (56%) and social context (52%). Pilot deployment: The AI-HDR service was embedded in the hospital’s electronic health record. In the pilot, 47 HDRs were autogenerated in real-world settings and reviewed by attending physicians. Missing information and factual errors were flagged in 53% and 47% of drafts, respectively, while written assessments diminished the importance of these errors. An LLM-driven, agent-orchestrated pipeline can safely draft real-world HDRs, cutting administrative overhead while achieving clinician-acceptable quality, not without errors that require human supervision. Future work should refine specialty-specific prompts to curb omissions, add temporal consistency checks to prevent outdated data propagation, and validate time savings and clinical impact in multi-center trials.

22 January 2026 in Toxins
Verified
Journal Article
Open access
PDF file
Helena Bascuñana-AmbrósHelena Bascuñana-Ambrós
Helena Bascuñana-Ambrós
JFJacobo Formigo-Couceiro
Jacobo Formigo-Couceiro
JCJosé Maria Climent-Barberá
José Maria Climent-Barberá
LGLuis Guirao
Luis Guirao
Michelle Cavariani Catta-PretaMichelle Cavariani Catta-Preta
Michelle Cavariani Catta-Preta
JFJosep Monguet Fierro
Josep Monguet Fierro
Background: Spasticity, which occurs with certain neurological conditions, substantially affects quality of life, function, and social participation. Despite widespread botulinum toxin use, variability persists in patient information, access to specialized rehabilitation, and follow-up models. Methods: This two-phase Delphi-Dialogue Patients–Professionals study (DDPP), promoted by SERMEF, integrated perspectives from 77 patients and 141 rehabilitation professionals. Phase 1 used parallel surveys to assess satisfaction, perceived effectiveness of botulinum toxin, communication preferences, and rehabilitation follow-up. Phase 2 applied Real-Time Delphi with 38 experts to generate consensus recommendations to improve spasticity management. Results: Patients and professionals agreed on botulinum toxin benefits, the importance of ongoing rehabilitation, and the value of hybrid (in-person/remote) follow-up. Key gaps concerned access to Physical Medicine and Rehabilitation services, clarity and timing of information, and shared goal setting. Experts translated these misalignments into 10 prioritized recommendations, with highest feasibility for actions standardizing access pathways, optimizing botulinum toxin use, reinforcing structured education, and consolidating hybrid rehabilitation models. Conclusions: The DDPP approach offers a reproducible framework to align stakeholder perspectives by converting quantified divergence into consensus-based innovation priorities. Implementing the recommendations—particularly those strengthening communication, education, and hybrid pathways regarding botulinum toxin treatment—may support more accessible, personalized, patient-centered spasticity care.

12 June 2025 in Healthcare
Verified
Journal Article
Open access
PDF file
ABAlba Brugués Brugués
Alba Brugués Brugués
JMJessica Morillas
Jessica Morillas
EVEnric Mateo Viladomat
Enric Mateo Viladomat
GSGlòria Jodar Solà
Glòria Jodar Solà
Michelle Cavariani Catta-PretaMichelle Cavariani Catta-Preta
Michelle Cavariani Catta-Preta
JFJosep Monguet Fierro
Josep Monguet Fierro
Background: European health systems are shifting toward more proactive, person-centered models, thereby highlighting the need to strengthen nurses’ clinical leadership in primary care. Nurse demand management (NDM) has emerged as an innovative practice which allows nurses to autonomously and comprehensively respond to a population’s health needs. However, knowledge on its implementation varies widely, often being intuitive, partly due to the absence of standardized evaluation tools. The xGID instrument aims to measure the degree of NDM adoption in primary care teams (PCTs), activating collective intelligence mechanisms to foster shared diagnosis, organizational reflection, and the generation of targeted recommendations. Methods: We designed and implemented xGID in 47 PCTs in Catalonia, involving 1474 healthcare professionals. Data were collected through structured surveys assessing key dimensions of NDM adoption, including professional autonomy, teamwork, continuity, and accessibility. Results: Overall adoption of NDM was high, with a mean score of 7.6 out of 10. Notable differences emerged between professional groups and practice areas. Nurses tended to be more critical of teamwork, longitudinal care, and accessibility, reflecting the central yet high-pressure role they play in NDM. High-scoring dimensions included professional autonomy and the capacity to act across multiple domains, whereas weaker areas pointed to systemic organizational challenges. Conclusions: The preliminary findings indicate that a standardized tool for NDM evaluation is a cornerstone for identifying contextual barriers and guiding the transformation of care models. Its participatory and strategic approach offers novel pathways to embed data-driven decision-making into daily clinical practice, consolidating NDM as a key pillar of future primary care.

12 August 2026 in Healthcare
Verified
Journal Article
Open access
PDF file
LFlourdes gil Fraguas
lourdes gil Fraguas
SHSoraya Hijazi-Vega
Soraya Hijazi-Vega
Michelle Cavariani Catta-PretaMichelle Cavariani Catta-Preta
Michelle Cavariani Catta-Preta
Alex Trejo OmeñacaAlex Trejo Omeñaca
Alex Trejo Omeñaca
JPJan Ferrer i Picó
Jan Ferrer i Picó
Helena Bascuñana-AmbrósHelena Bascuñana-Ambrós
Helena Bascuñana-Ambrós
Background/Objectives: Pelvic floor dysfunctions (PFDs) are highly prevalent chronic conditions associated with substantial functional and psychosocial burden. Although multidisciplinary pelvic floor rehabilitation is recommended as first-line management, its effectiveness depends not only on intervention efficacy but also on care delivery, communication, patient engagement, and long-term adherence. This study applied the Delphi-Dialogue Patients–Professionals (DDPP) framework to quantify patient–professional alignment and translate identified patient–professional perceptual differences into candidate service-improvement strategies. Methods: A two-phase observational study incorporating parallel stakeholder surveys and a modified Real-Time Delphi feasibility-assessment process was conducted under the auspices of the Spanish Society of Physical Medicine and Rehabilitation (SERMEF). In Phase 1, parallel online questionnaires were completed by 225 patients with PFD and 164 rehabilitation professionals. Shared items were rated on a six-point Likert scale, and patient–professional divergence was quantified using an operational composite index integrating standardized differences in mean ratings and high-agreement responses. In Phase 2, the implementation feasibility of 10 improvement recommendations was evaluated by a Delphi panel comprising 35 PM&R physicians and two representatives of the Spanish Incontinence Association (ASIA). Results: Patient–professional agreement was domain-specific. The greatest divergences concerned digital competence, timing of patient education, bladder and bowel diary use, and adherence to prescribed rehabilitation programmes. Professionals attributed greater value to bladder and bowel diaries and selected group-based interventions, whereas patients reported higher digital competence, more timely information delivery, and better adherence. Comparatively similar response patterns were observed regarding hybrid rehabilitation models, educational resources, app-supported follow-up, and overall satisfaction with rehabilitation care. The Delphi process identified structured home exercise programmes, adherence-monitoring strategies, and improved visibility of rehabilitation pathways as the recommendations with the highest perceived implementation feasibility. Conclusions: The findings suggest that important challenges in pelvic floor rehabilitation may arise from implementation-related factors in addition to the effectiveness of available interventions. The DDPP framework provides a structured, stakeholder-informed methodology for translating patient–professional divergence into candidate organizational recommendations with comparatively higher perceived implementation feasibility.