Research & Validation
Research Evidence & Product Validation
ProXplor™ develops health and medical intelligence through a continuing process of research, technology development, structured validation and product translation.
Our technology lineage spans multiple generations of research and engineering, with intellectual property, institutional collaboration, historical evaluation and regulated product development forming part of that foundation.
ProXplor maintains a clear distinction between different forms of evidence.
Historical knowledge is not validation. Historical validation is not automatically current-product performance. Regulatory authorization applies only to the specific product, intended use and jurisdiction for which it was granted.
This distinction is fundamental to how ProXplor approaches scientific and product claims.
Research, Technology & Intellectual Property Foundation
ProXplor’s current technology system builds on years of research, product development and successive generations of health-assessment technology.
The broader technology lineage has generated more than 70 registered intellectual property assets across related technology entities, covering software, product technology, engineering and proprietary health-intelligence development.
These intellectual property assets form part of the cumulative technology foundation supporting the continued development of:
ProXplor Research Framework™
ProXplor Nexus™
InferPheno™
InferFunc™
Phenotype Series™
HNS Series™
Intellectual property reflects the depth and continuity of technology development.
It is not, by itself, evidence of scientific or clinical validity.
Validation is evaluated separately.
Our Evidence Approach
Different forms of evidence answer different questions.
Knowledge Sources
Scientific literature, medical knowledge, historical medical observations and other structured sources may inform research questions, feature discovery and model development.
They provide a basis for investigation.
They do not, by themselves, validate a technology or product.
Technology Validation
Technology validation examines whether an analytical method or model performs consistently and appropriately for its defined task.
Depending on the technology, this may include reliability, repeatability, agreement, analytical performance and external evaluation.
Product Validation
Product validation concerns a specific product configuration, generation, intended use and operating environment.
Evidence from an earlier generation cannot automatically be treated as evidence for a later product.
Regulatory Evidence
Regulatory authorization confirms that a specific product has satisfied the applicable requirements for a defined intended use and jurisdiction.
It does not automatically extend to other products, purposes or markets.
Real-World & Longitudinal Evidence
Real-world deployment and repeated assessments may provide additional evidence regarding usability, consistency, population performance and changes observed over time.
These evidence categories are complementary, but they are not interchangeable.
InferFunc™ Research & Validation
Functional-State Inference Engine
InferFunc™ represents the functional-state intelligence branch of the ProXplor technology system.
Its development builds on multiple generations of functional-assessment technology, professional product engineering, institutional evaluation and real-world application.
Current InferFunc™ capabilities support structured assessment across 11 major physiological systems and 224 organ and functional units, including functional-state and repair-state assessment at applicable levels.
InferFunc™ also supports a separate TCM-informed meridian-state assessment framework covering the 12 primary meridians and 8 extraordinary meridians.
The physiological and TCM-informed frameworks are maintained as distinct analytical systems and should not be interpreted as equivalent scientific models.
Historical InferFunc™ Technology Evaluation
Earlier generations of the underlying functional-assessment technology were evaluated through research and institutional programs in China.
Historical publicly documented results include comparative and technical evaluations conducted with healthcare and research institutions.
Reported historical findings included approximately 83% overall clinical agreement in early comparative evaluations involving the Health Management Center of PLA General Hospital and China-Japan Friendship Hospital.
Historical evaluation associated with Jiangsu Provincial Hospital of Chinese Medicine reported more than 80% comparative agreement at the applicable evaluation level.
A historical technical evaluation conducted with the Institute of Aviation Medicine, Air Force reported 96.88% reliability and 87% validity under the methodology used in that evaluation.
Historical clinical validation programs completed before 2019 also reported a validated sample of 344 cases.
These figures relate to earlier generations of the underlying technology and historical evaluation methodologies.
They should not be interpreted as direct performance claims for current InferFunc™ or current HNS Series™ products.
Current-generation product claims must be supported by evidence applicable to the specific technology version, product configuration and intended use.
Research & Evaluation Institutions
Research and technology development within the broader ProXplor technology lineage has involved collaboration with academic, research and healthcare institutions in China.
Historical academic and research collaboration has included institutions such as:
Tsinghua University
Zhejiang University
Chinese Academy of Sciences
Southeast University
Nanjing University of Chinese Medicine
Historical evaluation and application programs have involved institutions including:
Health Management Center of PLA General Hospital
Institute of Aviation Medicine, Air Force
Jiangsu Provincial Hospital of Chinese Medicine
Jiangsu Provincial Hospital of Integrated Chinese and Western Medicine
The nature, scope and responsible entity of each collaboration may differ.
Academic collaboration, technical evaluation, clinical validation and product deployment are therefore not presented as equivalent forms of evidence.
InferPheno™ Research & Validation
Digital Phenotype Inference Engine
InferPheno™ is a distinct technology from InferFunc™ and requires its own evidence appropriate to digital phenotype–based health-state analysis.
Its research pathway combines historical medical knowledge, structured visual phenotype research, standardized annotation, artificial-intelligence modeling and contemporary validation methods.
InferPheno™ does not rely on the historical validation evidence of InferFunc™ as evidence of its own performance.
From Historical Observation to Structured Phenotype Intelligence
Historical Chinese medical literature contains extensive systematic observations of visible human characteristics, including facial appearance, color, morphology, tongue characteristics and other externally observable features associated with descriptions of human health and functional state.
ProXplor treats this body of historical medical knowledge as a research and knowledge source, not as modern validation evidence.
The research process therefore does not convert historical medical descriptions directly into AI conclusions.
Instead, relevant observations are progressively translated into forms that can be studied using contemporary methods.
Historical Medical Knowledge
Relevant descriptions and recurring observations are systematically reviewed and organized from historical medical literature.
These sources contribute to feature discovery and hypothesis generation.
Structured Visual Phenotype Research
Historically described visual characteristics are translated into clearly defined and structured phenotype features that can be observed and annotated under standardized conditions.
This provides a structured framework for visual phenotype feature definition and annotation.
The images involved are standardized phenotype images rather than conventional radiological or diagnostic medical imaging.
Contemporary Interpretation & Mapping
Where scientifically appropriate, phenotype observations may be examined in relation to contemporary biomedical knowledge, physiological concepts or independently measurable health variables.
Such mapping is used to develop research hypotheses and analytical relationships.
Mapping does not imply equivalence.
A traditional medical description is not automatically treated as a modern biomedical diagnosis, biomarker or disease state.
AI Modeling
Structured phenotype features and appropriately defined health information can then support model development, pattern analysis and digital phenotype inference.
Contemporary Validation
The resulting technology must be evaluated using modern research and validation methods appropriate to its intended use.
Historical medical knowledge therefore contributes to the origin of research questions and feature structures, while contemporary evidence determines whether the resulting technology performs adequately.
InferPheno™ Validation Dimensions
Relevant validation dimensions for InferPheno™ include:
Annotation Consistency
Evaluating agreement in the definition and annotation of structured phenotype features.
Model Agreement
Evaluating model outputs against appropriately defined expert or reference annotations.
Repeatability
Assessing the consistency of relevant outputs under standardized capture and repeat-assessment conditions.
Population & Device Performance
Evaluating performance across different populations, devices and operating environments.
External Validation
Testing the technology using data or populations not used in model development.
Relationship to Independent Health Information
Examining associations between defined phenotype features and independently obtained health or physiological information where scientifically appropriate.
Longitudinal Evaluation
Studying whether relevant phenotype outputs change in meaningful and reproducible ways across repeated assessments over time.
The appropriate validation method depends on the specific output being evaluated and the intended use of the applicable product.
ProXplor does not treat all InferPheno™ outputs as though they require or represent the same type of evidence.
Product & Regulatory Evidence
Research technologies, commercial wellness products and regulated medical products operate under different evidence requirements.
Non-Clinical Products
Products designed for proactive health, general wellness and health optimization require validation appropriate to their actual claims and intended use.
They are not presented as diagnostic technologies solely because they use health-related information or artificial intelligence.
Clinical Products
Clinical products require separate validation, quality-management and regulatory processes appropriate to their defined medical intended use.
Specific products within the historical HNS clinical technology lineage have obtained Class II medical device registration in China.
Such authorization applies only to the specific registered product, intended use and regulatory scope identified in the applicable documentation.
It does not automatically authorize other HNS products, AIHNS™, InferFunc™, InferPheno™ or products offered in other jurisdictions for the same clinical use.
Ongoing Validation
Validation is not treated as a one-time event.
As technologies, models and product generations evolve, the evidence supporting them must also evolve.
ProXplor continues to develop validation appropriate to each technology and intended use, including work relating to:
reliability
repeatability
analytical performance
expert agreement
external validation
population performance
longitudinal reassessment
product-specific clinical validation where required
New evidence may support, refine or challenge existing models and assumptions.
That process is part of responsible scientific development.
Evidence Proportional to the Claim
ProXplor follows a simple principle:
Evidence should remain proportional to the claim.
Historical medical knowledge can inform research.
Historical technical validation can demonstrate technology lineage.
Current product validation can support current-product claims.
Regulatory authorization can establish specific legally authorized uses.
None of these should be substituted for another.
By maintaining these distinctions, ProXplor aims to develop health and medical intelligence that is not only technically capable, but also scientifically disciplined, transparent about its evidence and responsible in real-world application.
Explore ProXplor Research Framework™
Explore Inference Engines
Explore HNS Series™
About ProXplor