Revolutionizing healthcare with AI that's autonomous, personalized, and built to scale.
Unifying pathology, radiology, genomics, and multimodal data with AI.
Hover to discover
Our AI seamlessly integrates imaging data from CT, MRI, and X-ray with histopathological analysis, creating a unified diagnostic framework that captures the complete clinical picture for each patient.
SeleneX is advancing the future of healthcare through autonomous AI – built on deep integration of multi-modal patient data, machine learning, statistical reasoning, and scalable models for next-generation diagnosis, personalized interventions, and long-term care.
SeleneX integrates real-time AI, generative inference, and multi-modal patient data into intelligent systems that support diagnosis, prediction, and personalized care at scale.
Of ovarian cancer cases are detected at late stages, drastically reducing survival rates.
Late-stage survival rates are dramatically lower than early detection.
Current screening methods fail to capture multimodal signals of early-stage HGSOC.
Single-modality approaches miss critical biomarkers, increasing risk of missed diagnoses and patient mortality.
Generic protocols ignore individual patient trajectories, leading to delayed interventions and preventable deaths.
Scarcity and demographic gaps in available datasets limit AI fairness and accuracy.
Moving from reactive standard care to predictive, continuous monitoring.
Platform Overview
An innovative AI-driven platform leveraging cutting-edge multimodal data fusion to detect cancer at its earliest, most treatable stages.
20%+
Detection Improvement
100%
Privacy-First Design
CT / MRI
Genomics
EHRs
Symptoms
Imaging, omics, EHRs, and patient-reported data
Connecting diverse data sources intelligently
AI Agent Active
Orchestrating progressive data analysis...
Intelligent orchestration of diagnostic workflows
EHR & Anamnesis
Patient history
Biomarkers
Blood & Urine
Imaging
CT / MRI / US
Advanced
If needed
Starting non-invasive, advancing only when necessary — reducing patient burden and healthcare costs
Federated learning keeps data secure and compliant
Low Risk
Review
Why?
Transparent decisions clinicians can trust
Simulating...
Virtual patient models for treatment simulation
Core Technologies
Advanced capabilities powering the next generation of early cancer detection.
Unified patient view through cross-modal data fusion. Integrating imaging, genomics, EHRs, and patient-reported symptoms into a single intelligent system.
Proprietary knowledge graph connecting diverse data sources intelligently.
Privacy-preserving AI that learns without centralizing sensitive patient data.
Transparent reasoning with attention heatmaps and confidence scoring.
Privacy-compliant data generation for rare cancer subtypes using diffusion models.
Patient-specific virtual models for therapy forecasting and in silico testing.
The largest curated ovarian cancer data library available today. Combining ultrasound datasets, biomarker tables, omics matrices, and clinical schemas into one unified resource.
Learns patterns from actual clinical records
Generates realistic patient profiles
Perfect for fairness testing
No patient identity leakage
99.2%
Fidelity Score
0.01%
Privacy Risk
<2min
Generation Time
SeleneX transforms fragmented medical data into a unified, predictive patient view through a 4-step multimodal engine.
Transforming diverse patient data into secure, computable vectors. We ensure 100% data sovereignty with on-premise encryption before any analysis begins.
The Proof
Rigorous validation across diverse populations and clinical settings ensures reliability when it matters most.
Global Expertise
15+ PhDs
AI, oncology, radiology, pathology & bioinformatics
Data Modalities
Multimodal5+
Countries
Global7
Diagnostic Pathway
Patient Scan
CT_Thorax
ID: #8921
Processing
Oncologist Review
In Review
Priority: High
Confidence: 98%
Symptom Analysis
Week 34Anomaly Detected
Unusual spike in reported fatigue.
Distribute signing authority across regions securely.
Data Sources
Research DatasetsMultimodal Clinical Registry
Expanding clinical-imaging pairs
Knowledge Graph (PKG)
Proprietary Ovarian Pharmacogenomics
Data Pipeline
Training on validated public datasets.
Global Impact
A vast, living network of intelligence. Every case analyzed refines the global model — transforming individual data points into shared diagnostic intelligence.
A continuously expanding foundation of synthetic patient trajectories used for research, model development, and validation. Hundreds of thousands of generated trajectories across modalities.
Reducing diagnostic noise by contextualizing anomalies across multimodal patient data — minimizing unnecessary interventions.
Designed to surface risk indicators earlier in clinical workflows, where time-sensitive decisions matter most.
A team of 15+ PhDs and clinical, technical, and commercial experts from the U.S., Canada, Armenia, Portugal, Spain, Germany, the Netherlands, Brazil, and Mexico — spanning oncology, bioinformatics, AI, radiology, and regulatory science. United by a mission to build trustworthy, high-impact AI in healthcare.
Built to support clinicians with explainable, transparent AI — augmenting decision-making rather than replacing it.
Models developed and evaluated across diverse datasets, with ongoing validation across populations and clinical settings. Built with validation pathways aligned to clinical and regulatory standards.
Designed to reason across imaging, biomarkers, clinical history, and emerging data modalities.
Seamless integration with hospital IT infrastructure — empowering diagnostics and care pathways without disrupting existing workflows.
Optimized for responsive, real-time assistance in clinical settings. Supports low-latency risk scoring, scan interpretation, or triage, depending on local setup.
Designed to meet the highest data protection standards: HIPAA (US), GDPR (EU), and privacy-preserving architectures for hospital-grade deployments — including on-premise and hybrid options.
Enables decentralized collaboration: hospitals can contribute to global model improvement without sharing sensitive patient data.
SeleneX includes a built-in Synthetic Data Generation Engine to safely simulate diverse patient populations across imaging, biomarkers, clinical variables, and omics. This unlocks safer model prototyping, auditing, and federated learning — even in data-scarce or privacy-sensitive environments.
SeleneX is designed to integrate real-world, multi-institutional datasets — including public cohorts (e.g., TCGA, MIMIC-IV) and curated hospital data — into a continually evolving, privacy-safe catalog of cancer knowledge.
SeleneX was born from a shared conviction: that high-quality, personalized healthcare should not be a privilege of access, geography, or institution — and that AI, when designed with care and clinical depth, can radically expand the reach and equity of diagnostics and treatment.
Founded in 2025, SeleneX brings together a multidisciplinary, globally distributed team united by a clear mission to transform early cancer detection through multimodal AI.
To build the AI systems that will power intelligent, personalized, and autonomous care pathways — grounded in real-world clinical data, rigorous science, and human trust.
Founding Team
Together, they bring a rare combination of AI systems engineering, global health policy, and deep regulatory awareness.
CEO & Chief AI Architect
10+ years in AI, econometrics, and data science, with international leadership across the Netherlands, UK, Italy, Canada, and the U.S.
Directs SeleneX's systemic AI architecture, ethical governance, and global clinical partnerships.
First-authored publications in machine learning and AI, with contributions to scientific writing and academic research that bridge theory and clinical application.
CTO & Co-Founder
9+ years in advanced AI/ML system architecture, large-scale agent deployment, and robust cybersecurity practices.
Lead architect of SeleneX's multi-modal fusion engines, autonomous AI agents, and production-grade RAG systems.
Active contributor to open-source AI infrastructure, with a focus on building scalable, secure systems for healthcare applications.
Tatev and Vahe Aslanyan founded SeleneX with a singular conviction: that decoding biology requires not just data, but a fundamental rethinking of how intelligence scales. Coming from distinct worlds—econometrics and high-performance engineering—they united to build the infrastructure for the next century of healthcare.
From machine learning engineers to clinical researchers, our diversity is our strength. Operating across 3 continents to solve healthcare's hardest problems.
Hubs in San Francisco, London, and Amsterdam driving 24/7 innovation cycles across three distinct timezones.
35%
AI & Machine Learning
20%
Computational Biology
MD+
Clinical Strategy
Nationalities
15+
PhDs & Researchers
Help us decode the future of human health. We are looking for extraordinary talent to join our global hubs.
AI/ML engineers, data scientists, and econometricians
Oncologists, radiologists, surgeons, and infectious disease specialists
Bioinformaticians and omics experts
Clinical researchers, regulators, and commercialization leads
Academic professors and institutional collaborators
"Science knows no borders—and neither does our mission."
SeleneX is supported by a Scientific & Clinical Expert Council of over 15 senior researchers, physicians, and professors—each with 15–20+ years of experience in key areas of medicine and health technology.
Their guidance anchors our work in real clinical needs, safety, and scientific rigor.
This council provides clinical depth, validation oversight, and alignment with evolving global regulatory standards—ensuring our technologies remain clinically relevant and future-proof.
Ukraine, Canada, Armenia
Head of Radiology, Armenia
Armenia, the Netherlands
Portugal, U.S.
U.S., Canada
The Problem
"AI's most profound application in the long-term will be in healthcare — especially oncology — as AI combined with gene sequencing and CRISPR accelerates medical breakthroughs."
Cathie Wood, ARK Invest CEO • All-In Summit 2025
71% of ovarian-cancer cases are diagnosed at a late stage (Stage III–IV).
Target Ovarian Cancer • Key facts and figures
"5-year survival drops from ~92% (localized) to ~32% (distant). Stage is destiny."
American Cancer Society • SEER 2015–2021
AI support increased breast-cancer detection by 17.6% (6.7 vs 5.7 per 1,000) without worsening recall rates.
Eisemann et al. • Nature Medicine (2025)
"Ovarian cancer is usually found late: 55% diagnosed after metastasis; only 20% caught while localized."
US NCI SEER • Cancer Stat Facts
Moving diagnosis from late to early stage can improve 5-year survival from 32% to 92% — nearly tripling patient outcomes through AI-enabled early detection.