Today, we are announcing SeleneX, a clinical AI platform designed to bring the reasoning of top oncologists directly in front of every patient, from the very first signal through diagnosis, treatment, and long-term follow-up. SeleneX is built to address one of the most fundamental failures in modern oncology: not the inability to treat disease, but the inability to detect it early and manage it continuously. It is designed to detect ovarian cancer earlier, support clinicians in making more informed and consistent decisions, personalize treatment pathways, and reduce the risk of recurrence by ensuring that no clinically relevant signal is overlooked.

Ovarian cancer remains one of the deadliest and most silent gynecological diseases, not because it is untreatable, but because it is most often identified too late. There is no effective global screening system, and early-stage detection remains exceptionally difficult due to vague, non-specific symptoms that are frequently dismissed or fragmented across clinical encounters. Decisions are often made on isolated data points — a single scan, a lab result, or a one-time consultation — rather than on a continuous and evolving understanding of the patient. As a result, the majority of patients are diagnosed at advanced stages, where survival outcomes drop dramatically compared to early-stage detection. This gap between early and late detection is not marginal; it defines the difference between intervention and survival.

SeleneX is designed to fundamentally shift this paradigm. It is not a diagnostic tool, nor a chatbot, nor an additional superficial layer on top of existing healthcare systems. It is a deeply engineered clinical intelligence platform that introduces a continuous, multimodal understanding of the patient. It begins with what does not exist today: early, non-invasive risk detection based on subtle and often overlooked signals. From there, it extends across the entire trajectory of care, integrating clinical records, imaging data, laboratory results, biomarkers, pathology, genetic information, and patient history into a unified system that evolves over time.

Rather than evaluating a patient once, SeleneX enables a model where the patient is continuously understood. It connects fragmented data into a coherent clinical picture, allowing for earlier identification of risk, more informed treatment decisions aligned with both clinical protocols and individual patient context, and ongoing monitoring that detects progression or recurrence at an earlier stage. This approach transforms care from reactive to predictive, and from episodic to continuous.

The underlying system is built on a combination of foundational statistics, machine learning, deep learning, and generative AI, working together to process and reason across multimodal data. This is not an API-driven abstraction or a lightweight software layer; it is a production-grade intelligence system designed for real-world deployment in clinical environments. The architecture is intended to bridge the gap between data and biological reality, enabling a level of clinical reasoning that is both scalable and grounded in evidence.

SeleneX connects the full cancer journey into a single, integrated system, spanning early detection, diagnosis, treatment, follow-up, and recurrence risk management. By doing so, it enables earlier detection of disease, increases the consistency and quality of clinical decision-making, provides continuous visibility into patient progression, and reduces the likelihood of missed or delayed signals. The goal is not to replace clinicians, but to augment them with a system that supports earlier, better, and more reliable decisions at every stage of care.

The platform is built by a globally distributed team operating at the intersection of artificial intelligence, clinical medicine, regulatory science, and large-scale system engineering. It is not structured as a research initiative, but as an engineering-driven system supported by active clinical and scientific integration. The team brings extensive experience across AI, machine learning, multimodal systems, natural language processing, generative AI, bioinformatics, and production infrastructure, and is complemented by an active network of clinicians, researchers, and domain experts across oncology, radiology, gynecology, clinical trials, and healthcare systems. This integration ensures that the system is not only technically robust, but clinically grounded and aligned with real-world constraints and requirements.

SeleneX operates across multiple regions, with collaborators and partners spanning Europe, North America, and beyond, enabling exposure to diverse healthcare systems, regulatory environments, and patient populations. This global perspective is critical in building a system that is both scalable and adaptable across different clinical contexts, while maintaining a strong focus on clinical validity, safety, and long-term impact.

Importantly, SeleneX is not a research concept. Core system components have already been prototyped, a multimodal cancer data infrastructure has been developed, and clinical collaborations and validation pathways are actively being established. From its inception, SeleneX has been designed as a deployable clinical system, with a clear path toward real-world integration and impact.

At its core, SeleneX exists to address a simple but critical reality: ovarian cancer is deadly because it is detected too late. By bringing the reasoning of top oncologists to every patient, continuously and at scale, SeleneX aims to shift detection earlier, improve decision-making, and support patients throughout the full trajectory of their care. The objective is not incremental improvement, but a fundamental rethinking of how cancer is detected, understood, and managed.

SeleneX is being developed as a global clinical system, and we are actively expanding our network of collaborators, clinicians, researchers, and institutional partners. We welcome engagement from those working across oncology, artificial intelligence, and healthcare systems who are interested in contributing to, validating, or deploying this infrastructure. All collaborations are approached with a clear emphasis on scientific integrity, clinical rigor, and long-term impact, with the shared goal of building a system that meaningfully improves outcomes in cancer care.