Frequently asked questions

EVAL Research Institute (EVALRI) is an independent California-based research organization. It develops structured clinical AI, population research, longitudinal assessment methods and evidence-based tools for medicine and public health.

Ad-Infer is an evolving causal inferential system developed by Maurice M. Ohayon beginning in 1987. Maurice M. Ohayon personally holds all rights in Ad-Infer and its associated intellectual property. EVAL Research Institute holds no ownership rights in Ad-Infer. It connects clinical domain knowledge, Type-2 causal reasoning, fuzzy evaluation, neural and deep-learning mechanisms, Bayesian belief updating and integrated language-model capabilities.

The LLM is an internal functional component of Ad-Infer. The inference engine directs its questioning and interpretation using current clinical hypotheses, causal relationships and the belief state. It supports standardized wording, clarification, translation, WHY explanations and semantic understanding. Returned information is re-evaluated by EVAL-KBS before it contributes to clinical inference.

Feedback includes which questions were asked or skipped, understood or misunderstood, accepted or refused, together with answers and clarifications that support, qualify or challenge a hypothesis. Ad-Infer can use this information to strengthen a positive diagnosis, revise its belief state, ask additional questions or redirect the differential diagnosis.

The Ad-Infer inference engine retains diagnostic authority. The integrated LLM contributes language and semantic functions; its output is assessed through the system’s structured causal reasoning, fuzzy evaluation and belief-updating processes.

EVAL-KBS provides knowledge processing, causal reasoning, fuzzy evaluation and learning within Ad-Infer. Sleep-EVAL is the domain knowledge base for sleep disorders used by Ad-Infer. Explore the knowledge bases.

Since 1987, its development has extended computational foundations, knowledge representation, adaptive interviewing, population research, longitudinal assessment and web-based applications. Current LLM capabilities are integrated into the same inferential architecture. See the architecture and development history.

Population-based, longitudinal and condition-specific research in sleep medicine, psychiatry, neurology, epidemiology, gastroenterology and related fields. Projects examine prevalence, incidence, persistence, remission, treatment, function and multidomain clinical burden.

Structured phenotyping, clinical trial prescreening and recruitment support, longitudinal assessment, patient-subgroup characterization, screening-tool development, real-world evidence programs and customized clinical AI applications. Explore products and services.

Ad-Infer conducts adaptive, structured interviews. Questions follow the current information and clinical hypotheses. Defined criteria, consistency checks and inferential rules support comparability across participants, sites, languages and assessment waves.

Research assessments and screening tools support research and clinical workflows. They do not replace examination, diagnosis or treatment by a qualified health professional.

Access is limited to authorized personnel, research information is used for the purposes described in the applicable study materials, and research findings are reported in aggregated form. See the Privacy Policy and study-specific consent information.

The Publications page includes selected recent and foundational research, with links to the original journal article, DOI or PubMed record. Historical descriptions refer to the configurations used in those studies.

Use the Contact page or email contact@evalri.com.