The I3LUNG project aims to achieve the highest performance in personalized medicine
through Artificial Intelligence/Machine Learning (AI/ML) modelled on multimodal patients'
data, together with implementing an AI/ML model in a real-life setting. A set of
patient-centered ML tools designed and validated for the project, which make use of the
novel virtual patient AVATAR entity for predicting progression and outcome. To maximize
its impact, the use of Trustworthy explanaible AI methodology will integrate the AI's
inherent performances with the input of human intuition to construct a responsible AI
application able to fully implement truly individualized treatment decisions in NSCLC
interpretable and trustworthy for clinicians. The final objective is the establishment of
a Worldwide Data Sharing and Elaboration Platform (DSEP). The DSEP will provide guiding
tools for patients, providing information to generate awareness on treatments. Lastly, it
gives access to researchers and the general scientific community to the most up-to-date
data sources on NSCLC.
Within the I3LUNG project, an ad-hoc IPDAS for NSCLC patients will be developed. Patient
decision aids are tools that might be used by patients either before or within a
consultation with physicians. Patient decision aids explicitly represent the decision to
be made and provide patients with user-friendly information about each treatment option
by focusing on harms and benefits. This tool could allow patients to explain and clarify
the high complexity of the information provided by the AI/ML approach. These decisional
support systems have been demonstrated to be effective in empowering patients, improving
their knowledge, promoting their active participation in clinical decision-making about
treatments, and improving overall patient satisfaction with care while decreasing
decisional conflict and decisional regret (26-30).
Finally, within the I3LUNG project it will be assessed whether using the IPDAS during the
clinical consultation would foster the quality of the shared decision-making as well as
the quality of the doctor-patient communication. Alongside the evaluation of the impact
of the IPDAS, it will be also evaluated whether the inclusion of the AI/ML predictive
models in clinical practice will be added value in supporting oncologists' clinical
decision-making and decreasing cognitive fatigue and decisional conflict.
I3LUNG adopts a two-pronged approach to develop a medical device through the creation and
validation of retrospective and prospective AI-based models to predict immunotherapy
efficacy for NSCLC patients using the integration of multisource data (real word and
multi-omics data) through a retrospective - setting up a transnational platform of
available data from 2000 patients - and a prospective - multi-omics prospective data
collection in 200 NSCLS patients - study phase.
The retrospective part of the I3LUNG project includes the analysis of a multicentric
retrospective cohort of more than 2,000 patients. This cohort will be used to perform a
preliminary knowledge extraction phase and to build a retrospective predictive model for
IO (R-Model), that will be used in the prospective study phase to create a first version
of the PDSS tool, an AI-based tool to provide an easy and ready-to-use access to
predictive models, increasing care appropriateness, reducing the negative impacts of
prolonged and toxic treatments on wellbeing and healthcare costs. Also, CT and PET scans
will be collected and a first radiomic signature will be created to feed the R-Model.
The prospective part of the project includes the collection and the analysis of
multi-OMICs data from a multicentric prospective cohort of about 200 patients. This
cohort will be used to validate the results obtained from the retrospective model through
the creation of a new model (P-Model), which will be used to create the final PDSS tool.
