Dirbtinis intelektas klinikiniuose tyrimuose: žmogaus centriškumo vaidmuo teisiniame reguliavime
Nustatymai
Item type:Publikacija, research article[2024][S1][S004][17]; ; Schiuma, GiovanniEuropean journal of innovation management. Bingley : Emerald, 2022, 27, 1, 2024, vol. 27, no. 1, p. 153-169Purpose – This paper aims to model organizational resilience structure. Based on the central insights of the scientific literature, organizational resilience is modelled as the result of an organizational capacity to bounce- back and bounce-forward. Design/methodology/approach – The paper is based on a quantitative empirical study to support the structural perspective of organizational resilience and investigate the relationships among the dimensions to test the above hypothesis by applying confirmatory factor analysis (CFA) and structural equation modelling (SEM) methods. Findings – The results confirmed three models that could be adopted to assess organizational resilience. The first model endorsed endogenous positive interrelationship among all three dimensions. The second model indicated that bounce-back dimensions, i.e. network and leadership and culture, have endogenous effects. Only the leadership and culture dimension is positively associated with a bounce-forward, i.e. change-ready and learning dimension. And the third model demonstrated that the network dimension is linked to leadership and culture, which is linked to the change ready and learning dimension. Originality/value – This study attempts to provide empirical evidence identifying the links between the bounce-back and bounce-forward stages of organizational resilience. These results contribute to the development of organizational resilience theory, confirming the conceptual statements that resilience is the ability to return to the routine and to adapt to the changing environment by overcoming dynamic events, stressing the idea of the importance of enhanced learning capacity, which allows for growth by constantly learning from oneself by gaining unique experiences.
48Scopus© citavimai 46 Item type:Publikacija, research article[2025][S1][M001,S001][8]; ; Peccatori, FedroAI in Precision Oncology., 2025, p. 8-15Abstract Background: Historically, pregnant patients have not been included in clinical research, as the risk of damaging the healthy development of the fetus was considered too high. This practice is changing, and the applications of artificial intelligence (AI) to pregnancy research offer a wide range of solutions that are expected to help closing the knowledge gap regarding how best to treat pregnant patients with cancer. The AI tools currently available offer support in screening, diagnosing, treatment planning, and patient counselling. These AI tools, however, were not developed by clinicians or medical researchers; thus, close collaboration between clinicians and the developers of AI tools is urgently needed. Methods: Many of the questions that emerge when AI is used to better approach a pregnant patient with cancer are not technological. Instead, these questions concern value conflicts that cannot be solved by technological tools alone and where wider public debate and agreement are required. How should AI tools consider the developing fetus and its pregnant mother? Should both the pregnant person and the developing fetus be regarded as patients or research participants? What kind of relationship between the two should be incorporated into AI algorithms? This article provides an in-depth interdisciplinary analysis of the above questions and concludes with recommendations for the immediate future. Results: Key recommendations include (1) adhering to existing biomedical ethics principles of respect for patient autonomy, including the relational context; the balance of maternal and fetal beneficence; protection of the vulnerable; and reasonable resource allocation in the given circumstances; (2) when recruiting pregnant patients to research studies, focusing on building pregnant patients’ trust in clinical research and on enhancing pregnant patients’ knowledge so that they feel able to understand and adhere to clinical research requirements; (3) various AI tools can help health care professionals and researchers to plan clinical studies and to create patient- and clinician-directed educational and decision-making tools, while also making them more accessible. Conclusions: Wider cross-disciplinary debate is still needed in order to establish how AI systems and tools for cancer treatment during pregnancy and pregnancy research in general should regard pregnant patients and their developing fetuses, especially where moral status questions are concerned.
11