Approximately 30% of specialist referrals in Italy are considered inappropriate, a figure that rises to 70% in some university hospital settings, according to data revealed by the AOU Senese, as reported by Il Sole 24 Ore. This systemic issue stems from a fragmented specialist network, leaving primary care physicians navigating a complex web of portals and evolving directives. The result is wasted resources, delayed diagnoses, and increased patient frustration. Now, a new Italian startup, Geen.ai, is aiming to address this challenge with an artificial intelligence-powered platform designed to optimize healthcare pathways and ensure patients receive the right care, at the right time.
Geen.ai isn’t simply another digital health tool; it’s an infrastructure built on knowledge graphs, artificial intelligence, and a deep understanding of the Italian healthcare landscape. Founded in 2024 by Giulia Marchese and Lucia Vernino, the company seeks to connect symptoms to the most appropriate specialist, factoring in clinical guidelines, regional healthcare services, and individual patient needs. The core of their solution lies in a sophisticated system that moves beyond simple keyword matching, reasoning through complex medical scenarios to suggest the most effective course of action.
The genesis of Geen.ai is rooted in both professional expertise and personal experience. CEO Giulia Marchese, a data scientist with a doctorate earned in Germany and Mexico, spent over a decade consulting for UN agencies on issues like gender data gaps and access to healthcare for vulnerable populations. CMO Lucia Vernino brings extensive experience in marketing and communications from multinational corporations, with a focus on sustainability and gender equality. Vernino’s personal journey with multiple sclerosis – and the difficulties she faced accessing truly multidisciplinary care, extending beyond neurology to include specialized areas like pelvic floor rehabilitation – proved to be a pivotal moment. “Her experience became the zero case use,” Marchese explained, highlighting the need for a system that could connect a patient’s full range of symptoms to the appropriate specialists and support services.
Navigating a Complex System with a Knowledge Graph
Geen.ai’s technology is built around a “knowledge graph” – a vast network of interconnected data points encompassing symptoms, medical specialties, subspecialties, national guidelines, regional healthcare plans (PDTA), and social services. This graph, powered by GraphRAG – an architecture combining knowledge graphs, large language models, and verifiable clinical rules – allows the system to “reason” through a patient’s case, considering factors like gender, age, location, and medical history. The platform doesn’t just suggest a specialist; it maps a pathway through the often-opaque Italian healthcare system.
“In Italy, guidelines, PDTA, directives, and fee schedules exist as PDFs scattered across thousands of websites,” Marchese explained. “We reconstruct this information into a single, coherent, and constantly updated database.” When integrated into a healthcare organization, Geen.ai incorporates internal directives, providing a comprehensive view even for staff members who may not be fully aware of all the nuances of local protocols. The ultimate goal is to improve the “appropriateness of first referral,” increasing the likelihood that a patient is directed to the correct specialist from the outset.
Addressing the Gender Data Gap in Healthcare
A critical component of Geen.ai’s approach is its commitment to addressing the “gender data gap” – the historical underrepresentation of women in clinical studies. This bias can lead to misdiagnosis and ineffective treatment for women, as medical knowledge is often based on the “standard” male patient. Marchese emphasizes that their knowledge base is deliberately constructed to equitably represent both male and female health, and they are actively working to include data on non-binary and intersex individuals, groups historically excluded from medical research. “A technology cannot be efficient if it is not equitable,” she stated. “Efficiency and equity must always be at the same level.” Gender is not an afterthought but an integral step in the system’s reasoning process.
From FemTech to a Broad Healthcare Infrastructure
Geen.ai initially launched as a FemTech startup, focusing on women’s sexual and reproductive health. However, the founders quickly realized the broader applicability of their technology. “We realized that a very vertical solution was difficult to sell to healthcare organizations. And the graph didn’t have natural boundaries of specialty,” Marchese explained. The interconnected nature of symptoms, subspecialties, and social services transcends specific medical fields. Today, Geen.ai operates on a B2B and B2G model, selling to healthcare providers, insurance companies, municipalities, and private facilities, integrating with welfare portals, hospital systems, and insurance apps.
The company has established key technology partnerships, including with Microsoft, through Frontech – an accelerator backed by CDP Venture Capital, Cariplo Factory, and Gellify. This connection provides a potential pathway to reach Microsoft’s extensive network of Italian clients. A partnership with NTT Data is also underway. Geen.ai’s first client is the Comune di Messina (Municipality of Messina), a collaboration that highlighted the challenges of implementing AI in the public sector, requiring the municipality to first establish internal guidelines for its use. “The sales process is almost a political job,” Marchese noted, “involving workshops and mediating between the speed of technology and the pace of institutional governance.”
Looking internationally, Geen.ai is currently piloting a proof-of-concept project with the Isle of Man, a self-governing territory with a population of approximately 85,000, offering an ideal testing ground for scalability before expanding into the broader European market. The startup is also preparing for a pre-seed funding round, seeking a lead investor with experience in health tech who can contribute not only capital but also strategic guidance on product development.
Counting What Matters: Visibility in Healthcare
Marchese often uses the phrase “What isn’t counted doesn’t exist,” a concept drawn from her work on femicide in Mexico and access to healthcare for vulnerable populations. This principle underscores Geen.ai’s core mission: to produce visible what the healthcare system often overlooks. By creating a comprehensive map of healthcare pathways, the platform aims to reduce the estimated 30% of inappropriate specialist referrals, minimizing diagnostic errors, wasted resources, and the prolonged suffering of patients searching for the right care. Geen.ai’s approach, powered by a knowledge graph, a commitment to gender-inclusive medicine, and the belief that efficiency and equity are inextricably linked, represents a significant step towards a more accessible and effective healthcare system.
Disclaimer: This article provides information for general knowledge and informational purposes only, and does not constitute medical advice. It is essential to consult with a qualified healthcare professional for any health concerns or before making any decisions related to your health or treatment.
The future of healthcare navigation in Italy – and potentially beyond – is being shaped by innovations like Geen.ai. The company’s next key milestone is securing its pre-seed funding and expanding its pilot programs. What are your thoughts on the role of AI in improving healthcare access and efficiency? Share your comments below.
