GB/T 27420-2018Conformity assessment—Application guide for evaluation and expression of uncertainty in biological sample measurement (English PDF)

3.1 Measurand

The amount to be measured.

3.2 Metrological traceability

Through the uninterrupted calibration chain specified in the document, the measurement results are linked to the reference object, and each calibration in the calibration chain will Introduce measurement uncertainty.

3.3 Measurement principle

Used as a basis for measurement.

3.4 Measurement method

A general description of the logical arrangement given for the operations used in the measurement process.

3.5 Measurement procedure

Based on one or more measurement principles and a given measurement method, based on the measurement model and the calculations required to obtain the measurement results, A detailed description of the amount.

3.6 Reference measurement procedure

When calibrating or setting a standard substance, the measurement procedure used to provide the measurement results is suitable for evaluation by the same amount of other The measurement accuracy of the measured magnitude obtained by the measurement procedure.

3.7 Primary reference measurement procedure; primary reference procedure

A reference measurement procedure used to obtain measurements that are not related to the same type of measurement standard.

3.8 Regular measurement procedure

The measurement procedure used to test samples, whose measurements usually require metrology traceability.

3.9 Measuring system

A set of one or more measuring instruments that are assembled and adapted to give a measured amount of information within a specified interval, usually including He is equipped with devices such as reagents and power supplies.

3.10 Matrix effect

Sample characteristics that affect the measurement and measurable values independent of the presence of the analyte.

3.11 Reference material; RM

Substances with specific properties that are sufficiently uniform and stable, whose properties have proven to be suitable for the intended use in measurement or nominal property inspection.

3.12 Certified reference material; CRM

A document issued by an authority that provides one or more characteristics of uncertainty and traceability obtained using an effective procedure The standard substance of the quantity.

3.13 Calibration

The first step in a set of operations under specified conditions is to determine the relationship between the magnitude provided by the measurement standard and the corresponding indication.

Then, this information is used to determine the relationship between the measurement results obtained by the indication value, where the measurement value provided by the measurement standard and the corresponding indication value have inaccurate measurement. Degree.

3.14 Measurement accuracy; accuracy

The degree to which the measured value is measured and its true value.

3.15 Measurement trueness; trueness

Infinitely repeated measures the degree of agreement between the average of the magnitudes obtained and a reference magnitude.

3.16 Measurement precision; precision

Under the specified conditions, the measured value or the degree of agreement between the measured values is repeated for the same or similar object to be measured.

3.17 Measurement repeatability; repeatability

Measurement precision under a set of repetitive measurement conditions.

3.18 Repeatability condition of measurement; repeatability condition

Same measurement procedure, same operator, same measurement system, same operating conditions and same location, and the same or similar in a short time A set of measurement conditions that are like repeated measurements by the measured object.

Note: In chemistry, the term "intra-sequence precision measurement conditions" is sometimes used to mean "repetitive measurement conditions."

3.19 Measurement reproducibility; reproducibility

Measurement precision under reproducible measurement conditions.

3.20 Reproducibility condition of measurement; reproducibility condition

A set of measurement conditions for different locations, different operators, different measurement systems, and repeated measurements of the same or similar objects.

3.21 Intermediate measurement precision; intermediate precision

Measurement precision under precision measurement conditions during a set of periods.

3.22 Intermediate precision condition of measurement; intermediate precision condition

Except for the same measurement procedure, the same location, and a set of measurements that repeat measurements on the same or similar subject over a longer period of time In addition to the quantitative conditions, other conditions involving changes may also be included.

3.23 Within-laboratory reproducibility

Measurement precision under reproducible measurement conditions in the same laboratory.

3.24 Within-laboratory reproducibility condition

Reproducible measurement conditions in the same laboratory.

Note 1. Indoor reproducibility conditions are the measurement conditions for the greatest variation in precision in the same laboratory.

Note 2. For some laboratories, there may not be multiple measurement systems, and the precision may indicate the maximum possible imprecision measurement conditions.

Note 3. Indoor reproducibility conditions and period precision conditions are not strictly defined. When applied, the description of conditions should include changed and unchanging conditions and How much has actually changed.

3.25 Measurement error; error

The measured magnitude is subtracted from the reference magnitude.

3.26 Measurement bias; bias

An estimate of the system measurement error.

3.27 Measurement uncertainty; uncertainty

Based on the information obtained, a non-negative parameter that imparts dispersibility to the measured value is characterized.

Note 1. Measurement uncertainty includes components caused by system influences, such as components related to corrections and measurement criteria, and defined uncertainties. Sometimes the estimated system impact is not corrected, but treated as an uncertainty component.

Note 2. This parameter can be such as the standard deviation (or its specific multiple) called the standard measurement uncertainty, or the interval half width including the probability.

Note 3. Measurement uncertainty generally consists of several components. Some of these components can be based on a statistical distribution of a series of measurements, according to the class A of measurement uncertainty. The assessment was assessed and characterized by experimental standard deviation. Other components can be estimated based on the probability distribution of experience or other information. The Class B assessment of certainty is assessed and also characterized by standard deviation.

Note 4. Generally, for a given set of information, the measurement uncertainty corresponds to the amount of magnitude assigned to the measurement. A change in this value will result in a corresponding inaccuracy The change in the degree.

3.28 Standard uncertainty; standard measurement uncertainty

Measurement uncertainty expressed in standard deviation.

3.29 Type A evaluation of measurement uncertainty; Type A evaluation

The measurement uncertainty component is evaluated by the statistical analysis method for the measured value under the specified measurement conditions.

Note: The specified measurement conditions refer to repetitive measurement conditions, period precision measurement conditions, or reproducibility measurement conditions.

3.30 Type B evaluation of measurement uncertainty; Type B evaluation

The measurement uncertainty component is evaluated by a method different from the measurement uncertainty class A assessment.

Example: The rating is based on the following information.

---The amount of authority issued by the authority;

--- The amount of certified reference material;

--- Calibration certificate;

--- Instrument drift;

--- The accuracy level of the measured measuring instrument;

--- Limit values inferred based on personnel experience, etc.

3.31 Combined standard uncertainty; combined standard measurement uncertainty

The standard measurement uncertainty of the output obtained from the standard measurement uncertainty for each input in a measurement model.

Note: In the case of input quantities related in the measurement model, the covariance must be considered when calculating the synthetic standard uncertainty.

3.32 Expanded uncertainty; expanded measurement uncertainty

The composite standard uncertainty is the product of a numerical factor greater than one.

Note 1. This factor depends on the type of probability distribution of the output in the measurement model and the selected inclusion probability.

Note 2. The term "factor" in this definition refers to the inclusion factor.

3.33 Coverage interval

An interval containing a measured set of magnitudes determined based on the available information, the measured value falling within the interval with a certain probability.

Note 1. The inclusion interval is not necessarily centered on the selected measured value.

Note 2. The inclusion interval should not be called a "confidence interval" to avoid confusion with statistical concepts.

Note 3. The inclusion interval can be derived from the extended measurement uncertainty.

3.34 Coverage probability

The probability of including a measured set of magnitudes within a specified inclusion interval.

Note 1. To avoid confusion with statistical concepts, inclusion probabilities should not be called confidence levels.

Note 2. The inclusion of probability in GUM is also called "level of confidence".

Note 3. The inclusion probability replaces the "confidence level" that has been used.

3.35 Coverage factor

To obtain extended uncertainty, a number greater than one that is multiplied by the synthetic standard uncertainty.

Note: The inclusion factor is usually indicated by the symbol k.

3.36 Probability distribution

Give a (random variable) function that takes a random variable to take any given value or the probability of a given set.

3.37 Distribution function

For each threshold, a function is given for the probability that the random variable X is less than or equal to xi. GX(xi)=Pr(X <=xi)

3.38 Probability density function

The derivative of the distribution function, if the derivative exists, then gX(xi)=dGX(xi)/dxi, where GX(xi) is the distribution function and X is the random variable.

3.39 Normal distribution

The probability distribution of the continuous random variable X.

3.40 t distribution

The probability distribution of the continuous random variable X.

3.41 Distribution propagation

A method of determining the probability distribution of the output by applying a probability distribution of the input quantities associated with the output.

5.1 Metrological traceability of measurement results

5.1.1 Tracing to international or national standards is the basis for the comparison of measurement results.

5.1.2 The amount of some biological parameters is not yet available in metrologically recognized international or national standards. According to GB/T 21415, according to the measurement knot The metrological traceability of fruit is divided into the following five types of measurement procedures for biological samples.

a) Type I, with a primary reference measurement procedure and a primary calibrator, can be traced to the SI unit measurement procedure;

b) Type II, with internationally agreed reference measurement procedures (non-primary) and internationally agreed calibrators, cannot be traced to SI units Measuring procedure;

c) Type III, with internationally agreed reference measurement procedures (non-primary), no internationally agreed calibrators, not traceable to SI units Measuring procedure;

d) Type IV, with internationally agreed calibrators (non-primary), no internationally agreed reference measurement procedures, cannot be traced to SI units Measuring procedure;

e) Type V, no internationally agreed reference measurement procedures, no internationally agreed calibrators, and cannot be traced to SI units in measurement program.

5.1.3 In the absence of calibrators and reference measurement procedures, the results of the alignment can be compared to the reference material of the protocol or The mean of the protocol method.

5.2 Uncertainty of measurement results

5.2.1 The purpose of the measurement is to determine the value to be measured. The measurement result is usually only an approximation or an estimate of the measured value. Complete measurement junction If you need to attach a statement of uncertainty with the results.

5.2.2 There are many potential factors affecting the measurement results, including but not limited to the following factors.

a) the definition of the measurement is incomplete;

b) the reproduction of the definition of the measurement is not perfect;

c) insufficient representation of the sample being measured;

d) insufficient understanding of the impact of the measurement on environmental conditions or imperfect measurement or control of environmental conditions;

e) personnel reading error;

f) limits on instrument resolution or identification thresholds;

g) the measurement standards and the magnitude of the reference materials are inaccurate;

h) the values of constants and other parameters obtained from the outside and used in the data reduction algorithm are inaccurate;

i) approximations and assumptions in measurement methods and procedures;

j) repeating the variability of the measurement process under seemingly identical conditions;

k) Factors that have not been recognized or recognized.

5.2.3 The factors affecting the measurement results are not necessarily independent of each other. For example, some factors in 5.2.2a)~i) may contribute to j).

6.1 Overview of assessment methods

6.1.1 The exact value of the contribution that causes the measurement result to produce an error, although unknown or unknown, but with the random effects and systems that cause the error The uncertainty associated with the impact is measurable. Measurement uncertainty can be assessed in a variety of ways, and the assessment methods described in this document include. Bottom-up method (hereinafter referred to as bottom-up method) (6.1.2), top-down method (hereinafter referred to as Top-down method) (6.1.3) and Monte Carlo method (hereinafter referred to as MCM method) (6.1.4).

6.1.2 The bottom-up method focuses on how the input affects the results. The evaluation steps include.

a) The definition is measured. The accuracy of the measurement depends on the level of detail defined or specified by the measurement and is measured for the required accuracy. The definition should be complete enough to guarantee the uniqueness of its value.

b) mathematically express the relationship between the input quantity Xi related to the measured Y and the measured Y. Y=f(X1,,XN), function f shall contain each quantity, including all corrections and corrections for components that have a significant influence on the uncertainty of the measurement results. factor.

c) Determine the estimated value xi of the input quantity Xi, either based on a statistical analysis of a series of observations or by other methods.

d) Evaluate the standard uncertainty u(xi) for each input estimate xi. […]