Tatev Aslanyan is the Co-Founder and CEO of LunarTech, which holds a European Innovation Council Seal of Excellence and is recognised as a Top Data Science Bootcamp of 2025, serving 50,000+ learners across 144 countries. She is also Co-Founder and CEO of SeleneX, a multimodal generative-AI HealthTech platform building a clinical superintelligence system for ovarian cancer — from early detection through treatment personalisation and longitudinal follow-up care. Tatev holds an M.Sc. in Econometrics & Operations Research from Erasmus University Rotterdam and a B.Sc. from Tilburg University, with 10+ years of global AI engineering and data science experience and first-authored ACM publications. She is the author of 10+ AI handbooks on freeCodeCamp, ranked among the Top 2% Global AI Influencers on LinkedIn, and has spoken at Web Summit, Inclusion Summit, ACM SIGAPP, and multiple international universities. She has also been featured in Forbes, Yahoo, Entrepreneur, Benzinga, CEO Weekly, and Business Insider.

SeleneX: From Detection to Care - A Multimodal Generative-AI Superintelligence Platform for Ovarian Cancer   Ovarian cancer claims over 200,000 lives every year, and approximately 70% of cases are diagnosed at Stage III or IV — where 5-year survival collapses to ~32%, even though detection at Stage I yields survival rates exceeding 90%. This is not a biology problem. It is a systems problem: no validated population-level screening pathway exists for ovarian cancer — unlike breast cancer (mammography) or cervical cancer (Pap smear) — and approximately 7 in 10 symptomatic women are initially dismissed or misdiagnosed at first clinical contact. The technology to close this gap exists. The clinical architecture to deliver it does not — yet.   This session presents SeleneX, a multimodal generative-AI platform engineered as a clinical superintelligence system for ovarian cancer — spanning the entire patient trajectory from early screening and diagnostic confirmation through treatment protocol automation, personalisation, and intelligent longitudinal follow-up care. SeleneX is designed to encode the reasoning of world-class gynaecological oncologists into every patient interaction, regardless of geography, institution, or the experience level of the attending clinician.    The platform is built on an evidence-layered architecture that mirrors how experienced clinicians build diagnostic confidence over time. Beginning with structured clinical intake — symptoms, physician observations, and electronic health records — the system progressively integrates laboratory biomarkers (CA-125, HE4), imaging analysed via computer vision (ultrasound, CT, MRI), genomic and molecular profiles, and longitudinal monitoring signals. The analytical stack combines foundation machine learning, deep learning, computer vision, large language models, and retrieval-augmented generation — applied simultaneously across structured tabular data, unstructured clinical text, imaging, and genomic modalities.    Two principles define the design. First, full explainability: every risk assessment, protocol recommendation, and monitoring alert is paired with a natural-language clinical reasoning summary — identifying which factors drove the conclusion, with what confidence, and what the recommended next step is. Outputs are formatted separately for the clinician (documentation-ready, MDT-review-suitable) and for the patient (plain language, downloadable). Second, automated adherence to international clinical protocols (NCCN, NICE, ESMO) through auditable decision-tree architectures, with personalisation driven by patient-specific molecular profiles, comorbidities, treatment histories, and continuously updated monitoring signals.    SeleneX has been co-developed from inception with an international consortium of 17+ gynaecological oncologists, radiologists, and clinical specialists with 15–20+ years of experience from institutions across North America, Latin America, Europe, and Africa — active contributors to the clinical logic, diagnostic weights, and protocol structures at the core of the system. The internal engineering team consists of 12+ engineers spanning foundation ML, deep learning, computer vision, econometrics, RAG systems, and production-grade clinical AI deployment. The platform is grounded in a curated, harmonised dataset of 200,000+ ovarian cancer records processed through a three-tier quality pipeline, augmented by a proprietary synthetic data generation engine that produces biologically realistic, privacy-preserving training data for the early-stage scenarios most critical to detection performance. LunarTech, SeleneX's parent technology company, holds the European Innovation Council Seal of Excellence.    This talk offers a frank look at how multimodal generative AI is moving from research demonstrations to deployable clinical infrastructure — the architectural decisions that make AI agents trustworthy in oncology settings, the trade-offs between automation and clinician control, and the data, regulatory, and translational realities of building deep-tech in healthcare at the frontier of one of the world's most lethal cancers — with the potential to shift population-level diagnosis toward earlier, more survivable stages of disease while personalising treatment and follow-up for every individual patient.