Glossar Archiv » QFINITY

A C D E F G H I L P Q V

AI Governance

AI Governance is the overall set of organizational structures, policies, processes and rulesets that ensure the responsible and effective use of AI within an organization. It covers roles and responsibilities, decision-making functions and bodies, a use-case register with risk classification,…

AI Model

The AI model is the AI itself within the AI system – the trained component that derives outputs from inputs. The EU AI Act defines the AI system (Art. 3 No. 1) and – for general-purpose models only – the…

AI System

An AI system is a machine-based system that is designed to operate with varying levels of autonomy, that may exhibit adaptiveness after deployment, and that, for explicit or implicit objectives, infers from the input it receives how to generate outputs…

ALCOA

ALCOA is a common acronym summarizing the criteria of data integrity (the ALCOA principles). It stands for: Attributable (i.e., identification of the source, the data originator – person, device, or system) Legible (and understandable) Contemporaneous (i.e., recorded at the time)…

Audit

An Audit is a systematic, independent and documented process for obtaining audit evidence and evaluating it objectively to determine the extent to which agreed criteria are fulfilled.…

Audit Trail

An audit trail is the secure, computer-generated, time-stamped record of relevant creations, changes, and deletions of GxP-relevant data – who, what, when, and why; which events it captures is defined on a risk basis (EU GMP Annex 11 §9; 21…

Coding

Coding is the generation of program code to create software modules. The coding is subject to programming guidelines, which define the quality requirements for code generation and thus ensure the traceability of the program code. The correctness of the generated…

Computer

A Computer is a functional unit that can perform substantial computations, including numerous arithmetic operations and logic operations without human intervention.…

Computer Software Assurance (CSA)

Computer Software Assurance is the FDA’s risk-based approach for establishing and maintaining confidence that software is fit for its Intended Use – in the guidance’s own wording, the Intended Use of the software as part of production or the quality…

Computer System

Computer systems are a combination of hardware, software, and firmware that work together to perform specific tasks. They consist of various components such as processors, memory, input and output devices, and operating system software. These components work together to process,…

Computerized System

The term computerized system covers a broad range of systems including, but not limited to, automated laboratory equipment, laboratory information management, clinical trials data management, vigilance systems, process control and process analytics, manufacturing resource planning, automated manufacturing equipment, manufacturing execution,…

Computerized System Validation

Computerized system validation describes achieving and maintaining compliance with applicable GxP and Medical Device regulations and establishing documented evidence that the system is fit for intended use by: the adoption of principles, approaches, and life cycle activities within the validation…

Context of Use

Context of Use (CoU) is the specific role and scope of an AI model used to address a Question of Interest – the FDA’s articulation of the Intended Use at the process level. The CoU spans forward: it defines the…

Critical Thinking

Critical thinking is a systematic, rational, and disciplined process of evaluating information from a variety of perspectives to yield a balanced and well-reasoned answer. Critical thinking promotes informed decision-making and good judgment about where and how to apply and scale…

CSV vs. CSA

CSV (Computerized System Validation) and CSA (Computer Software Assurance) are not an either-or: CSA does not replace CSV. The final FDA guidance (2025) merely supersedes Section 6 of the FDA guidance “General Principles of Software Validation” and takes over the…

Data

Data generally denotes facts, values (numerical or otherwise) or recordable findings, such as those obtained by measurements or observation.…

Data Governance

Data governance encompasses the organizational and technical structures that keep data available, with integrity – changeable under control and traceable – and ALCOA++-compliant across the entire data lifecycle. This includes roles and responsibilities, decision-making functions and committees, and controls –…

Data Integrity

Data integrity is the degree to which a collection of data is complete, consistent and accurate throughout the data lifecycle. The collected data should meet the ALCOA++ criteria. Ensuring data integrity requires appropriate quality and risk management systems and compliance…

Data Lifecycle

The data lifecycle covers all phases data passes through – from creation to archiving or deletion. Data risks are assessed and managed in a way that reflects their potential impact on Product quality, Patient safety and/or the reliability of the…

Electronic Record

Electronic record describes any combination of text, graphics, data, audio, images or other informational representations in digital format generated, modified, maintained, archived, accessed or distributed by a computer system (cf. 21 CFR Part 11).…

EU GMP Annex 11

EU GMP Annex 11 (“Computerised Systems”) is the European GMP annex for computerized systems: introduced in 1992, current in its 2011 version, part of EudraLex Vol. 4. Its principle divides the world of evidence in two: the application is validated,…

Evidence

Evidence is the documented proof – the records – that regulated activities were carried out as specified and that compliance is demonstrated. EU GMP Chapter 4 (Documentation) governs how these records are generated, controlled and retained.…

Fit for Purpose

Fit for purpose denotes the suitability of a system or model for its intended purpose – its context of use, or the intended use derived from the process. Distinct from verification: verification checks whether the model or AI system meets…

GAMP 5

GAMP 5 – “A Risk-Based Approach to Compliant GxP Computerized Systems” – is ISPE’s industry guide to the validation of computerized systems: published in 2008, and available since 2022 in the Second Edition, which strengthens critical thinking and scaled evidence.…

Human Oversight

Human Oversight is human supervision of AI-driven and computerized decisions: accountability rests with people, never with the system. This includes Human in the Loop (participate).…

Instructions vs. Records

Two types of documents carry every GxP quality management system. Instructions (specifications, procedures, plans) are reviewed and approved before release – approved, signed and dated; signed alongside the instructions are the evaluating reports that carry the release decision. Records evidence…

Integration

Integration is assembling hardware and/or software components into a complete system. Integration tests check the function and interaction of the individual components.…

Intended Use

Intended Use is the purpose a system is meant to serve. That purpose comes from the business process, not from the technical system – which only has a function (fit for purpose). Intended Use defines what the overall system is…

Least Burdensome

Least Burdensome is the FDA principle of the least required effort: generate no more effort and evidence than controlling the identified risk demands. It explicitly does not mean testing or documenting as little as possible – high risk still requires…

Lifecycle

The lifecycle describes the entire lifespan of a computerized system or AI model: concept, specification, verification and validation, operation, and decommissioning. Each specification step has its corresponding verification – classically in the V-model, and equally in agile and iterative models.…

Predicate Rules

Predicate Rules sind die eigentlichen Anforderungen der FDA – etwa 21 CFR 211 für Arzneimittel oder Part 820 (seit 2026 QMSR) für Medizinprodukte -, die festlegen, welche Aufzeichnungen zu führen und welche Unterschriften zu leisten sind. 21 CFR Part 11…

Process

A process is a set of interrelated or interacting activities which transform inputs into outputs.…

Process Validation

Process validation is the documented evidence that the process, operated within established parameters, can perform effectively and reproducibly to produce a medicinal product meeting its predetermined specifications and quality attributes (EU GMP Annex 15). Initial validation establishes this capability; the…

QMSR

The Quality Management System Regulation (QMSR) is the U.S. regulation of the quality management system for medical devices – 21 CFR Part 820, effective February 2, 2026. It replaces the previous Quality System Regulation by directly incorporating the requirements of…

Qualification

Qualification is the act of demonstration for the stable, stationary object – plant, equipment, and piping in GMP manufacturing, and likewise the IT infrastructure (frame of reference: EU GMP Annex 15; Annex 11, principle: validate the application, qualify the IT…

Quality

The term quality is derived from the Latin word “qualitas”, which means characteristic, feature, property or condition. In other words, the term quality defines the degree to which a set of inherent characteristics of an object meets requirements. Inherent means…

Quality Management

Quality Management (QM) refers to all organizational measures that relate to the structure of the organization and the efficient execution of processes at various levels of the company. With consistent implementation, quality management enables the continuous improvement of corporate culture,…

Validation of AI

Validation of AI is the common shorthand for the lifecycle management of AI (sub-)systems in the GxP environment. To be technically precise, “AI” itself is not validated: the overall computerized system is validated against its Intended Use; the AI model…

Verification

Verification is demonstration against a specification – the right form of evidence for everything that moves and changes: software, cloud, AI, and models. An AI model is always verified – not qualified; validation happens at the level of the overall…