Publications

Read the blog post 

Read the blog post

Diagnostic technologies are critical for improving health outcomes and extending quality of life. Accurate and
timely diagnosis plays a critical role in preventing disease, guiding treatment, and improving clinical outcomes.
Over the past century, diagnostic technologies have advanced to enable earlier disease detection and more precise health monitoring, allowing more timely and efective intervention. Recently, smart diagnostics driven by artifcial intelligence (AI), portable and mobile systems integrated with the Internet of Things (IoT), and noninvasive yet highly accurate detection methods have promised a bright future, although signifcant technological challenges remain. […] (Read more online)

Key Points

Question  Considering their risk compared with current smokers, when should former smokers begin lung cancer screening?

Findings  In this cohort study of 86 035 former or current heavy smokers, former heavy smokers could be screened between 3 and 17 years later than current heavy smokers, depending on the length of smoking cessation, in order to more closely reflect their risk levels. Whereas the US Preventive Services Task Force currently recommends a unified starting age of 50 years, the derived risk-adapted screening start ages for former heavy smokers ranged between 53 and 67 years of age.

Meaning  This study’s findings suggest that using differentiated, risk-adapted starting ages of former smokers could allow for enhanced lung cancer screening strategies.

This policy brief summarises recommendations from the third “Understanding (Risk Factors & Determinants)” cluster meeting to strengthen cancer prevention, early detection and equity access in the context of the EU Cancer Mission. It calls for support on interoperable, FAIR-by-design cancer data ecosystems that connect project-level resources with national cancer data nodes and emerging European hubs such as EUCAIM, UNCAN and CANDLE, ensuring sustainable reuse of legacy and new datasets. The report urges EU policymakers to prioritise integrated risk-stratification tools and minimally invasive biomarkers, backed by trustworthy AI, federated analytics and shared quality standards to accelerate translation into screening, surveillance, and personalised care. It further recommends that EU-level action on cancer inequalities adopt a broader perspective, including prevention, diagnostics, treatment and survivorship, while systematically embedding patient and citizen voices and clear governance frameworks for digital health and AI. Finally, the brief proposes that the European Commission and Member States should use these suggestions to guide future legislation, funding programmes, and mission-oriented initiatives, ensuring that scientific advances are matched by equitable access and long-term infrastructure support.

This document summarises the discussions and conclusions as presented by the participating projects. It does not introduce new content. This document reports on the third annual meeting of the Horizon Europe “Understanding (Risk Factors & Determinants)” cluster, hosted by the MELCAYA project in Barcelona the 15th of October 2025. It brought together the five cluster projects (GENIAL, LUCIA, ELMUMY, DISCERN and MELCAYA) focused on how risk factors and health determinants drive cancer development and progression. The first part of the meeting reviewed collaborative work on FAIR data management convergence on healthcare data standards and plans for long-term interoperability with emerging EU infrastructures such as UNCAN and EUCAIM. Scientific sessions showcased cross-project results on genetic susceptibility, exposome, omics and functional models, as risk stratification and early diagnosis tools, including AI-based models, smart sensors and proteomic signatures.

Further sessions addressed health policy implementation, inequalities and citizen engagement, highlighting gaps in rare cancers, the value of patient-generated evidence and methods such as social labs and Delphi surveys to co-create recommendations. The meeting concluded with presentations from UNCAN-Connect, CANDLE and EUCAIM projects, outlining how new federated European cancer data hubs will provide sustainable, secure environments for reusing cluster data and models. Overall, the cluster is progressing towards integrated, policy-relevant evidence and interoperable infrastructures that support the EU Missions on Cancer.

Screening heavy smokers by low-dose computed tomography (LDCT) can reduce lung cancer (LC) mortality, but defining the population that benefits most, a prerequisite for cost-effective screening, is challenging. In order to contribute to a more nuanced risk stratification of high-risk target populations, we developed and validated a blood-based protein marker model for LC. A two-stage design was implemented in this study, and the derivation set comprised 18 868 participants from the UK Biobank, which included 200 incident LC cases identified at 6 years of follow-up. The independent validation set included 101 LC cases identified at 6 years of follow-up. A total of 2025 protein markers measured by proximity extension assays available for both datasets were used for analysis. A risk prediction algorithm by least absolute shrinkage and selection operator regression with bootstrap method was developed in the derivation set and then externally evaluated in the independent validation set. The risk discriminatory performance of the protein marker model was compared with the established PLCO_m2012 model, USPSTF 2020 guidelines and trial criteria used in different LDCT trials. The protein marker model comprising of four protein biomarkers—CEACAM5, CXCL17, MMP12, and WFDC2—outperformed the PLCO_m2012 model, and the areas under the receiver operating curve (AUCs) for the protein marker model in the derivation and validation sets were 0.814 [95% confidence interval (95% CI), 0.785–0.843] and 0.814 (95% CI, 0.756–0.873), respectively. The addition of the protein marker model to the PLCO_m2012 model increased the AUCs up to 0.056 and 0.057 and yielded up to 16 and 12 percentage points higher sensitivities to identify future LC cases compared to the LDCT trial criteria, in the derivation and validation sets, respectively. The protein marker model improves the selection of high LC risk individuals for LDCT screening and thereby enhances screening efficacy.

Read the Blogpost

Conventional electromagnetic communication systems face limitations in dense environments, including high energy consumption, signal attenuation, and interference. To overcome these challenges, we present a bioinspired molecular communication (MC) platform using spatiotemporally allied single-walled carbon nanotube (SWCNT) sensors for volatile organic compound (VOC)-based signal transmission. Inspired by nature’s chemical signaling, this system employs hierarchical functionalized SWCNT sensor arrays to detect and interpret data-specified VOC pulses with high precision, mimicking pheromone-based communication. The system employs hydrophobic and biodegradable polymer-functionalized SWCNTs on nanoporous cellulose paper for enhanced VOC selectivity and response dynamics, enabling spatial and temporal signal encoding for robust multibit data transmission. Integrated machine learning (ML) algorithms facilitate signal decoding, pattern recognition, and environmental adaptation, ensuring reliable communication under varying conditions. The hierarchical sensor architecture and selective VOC interactions enable applications in gas detection, environmental monitoring, industrial safety, and real-time communication in inaccessible areas. Chromatographic detection of VOC mixtures within the layered sensor network further expands data transmission capacity, offering a scalable, energy-efficient alternative to conventional methods. This study advances bioinspired molecular communication, integrating nanomaterials with spatiotemporal sensing for next-generation, low-power, high-fidelity communication.