FDNA

FDNA develops AI-powered image analysis and decision support tools intended to expedite the diagnosis of developmental delays, genetic disorders, and rare diseases. The company serves a broad range of users including parents, pediatric specialists, healthcare providers, hospitals, and pharmaceutical companies, and its technology is used in over 10,000 medical centers, clinics, and hospitals across more than 130 countries. FDNA's work is supported by over 100 peer-reviewed publications and articles.

FDNA's platform supports the full patient journey, from early identification of potential conditions at home or in clinical settings through to care management and therapy planning. The company also operates a data clearinghouse that enables interoperability between healthcare providers, patients, pharmaceutical companies, and contract research organizations (CROs).

Products and Tools

• Face2Gene: An informational tool for licensed healthcare professionals that provides AI-powered image analysis and decision support for diagnosing developmental and genetic disorders. It does not provide a diagnosis and is not intended to be the sole basis for clinical decision-making; all outputs must be validated through clinical evaluation and appropriate testing.
• Digital Front Door for Healthcare Providers: Enables healthcare providers to receive and triage patients flagged with potential developmental, neurological, or genetic conditions.
• Symptom Assessment Tool: Guides parents through an assessment of their child's condition, behaviors, symptoms, and physical features, and provides content and suggestions for specialist referrals using a statistical matching algorithm between detected symptoms and potential health conditions.
• Data Clearinghouse: Provides streamlined interoperability between patients, healthcare providers, pharmaceutical companies, and CROs, including therapy services.

Multi-Modal Algorithm Framework

• Deep Gestalt®: An algorithm that maps images to potential syndromes.
• Gestalt Matcher™: An algorithm designed to discover new or rare syndromes and to support patient matching.
• Feature Extraction Algorithm: Searches for facial characteristics using images and maps these characteristics to Human Phenotype Ontology (HPO) terms.
• Heat Maps: Provides clinicians with visual references explaining why the Deep Gestalt algorithm matched certain results.
• Focus Assessment: Guides users through a post-image analysis assessment when the AI produces two or more high-confidence results, intended to reduce ambiguity between syndromes with similar dysmorphic signals.
• Feature Match™: An algorithm utilizing HPO terms and syndrome atrophy models, mapped to numerical vectors and syndrome-to-HPO probability scores.
• Clinical Notes Analyzer: A natural language processing (NLP) algorithm that extracts HPO terms, biomarkers, and symptoms from health records and clinical notes, supporting both text and voice input.
• D-Score™: An algorithm that determines the degree of face dysmorphology present in an image.
• A-Score™: An algorithm that seeks to identify facial characteristics in autistic syndromic and non-syndromic children, distinguishing between those with a diagnosed genetic condition involving specific facial dysmorphology and those without a described genetic condition.

Data and Research

• FDNA develops a comprehensive database of phenotypic information crowdsourced from clinicians, patients, laboratories, and researchers.
• The platform integrates multiple input modalities including image analysis, symptom assessment, NLP, video analysis, and voice analysis to detect health signals and support validation of health conditions.
• The technology and its underlying algorithms are intended to help expedite diagnosis and reduce the cost of care for developmental and genetic disorders.

FDNA's tools are designed to connect concerned parents and patients with appropriate healthcare providers and specialists as early as possible, while also supporting pharmaceutical companies and CROs through structured data interoperability for research and therapy development purposes.