Title: Validation of Predictive Models: Acceptable Prediction Zone Method
1Validation of Predictive Models Acceptable
Prediction Zone Method
- Thomas P. Oscar, Ph.D.
- USDA, Agricultural Research Service
- Microbial Food Safety Research Unit
- University of Maryland Eastern Shore
- Princess Anne, MD
2Background Information
3Terminology
- Performance evaluation
- Process of comparing observed and predicted
values. - Validation
- A potential outcome of performance evaluation.
- Requires establishment of criteria.
4Criteria
- Test Data
- Interpolation
- Extrapolation
- Performance
- Bias
- Accuracy
- Systematic Bias
5Predictive Modeling
Secondary Models
Tertiary Model
No Model
Observed No
Predicted No
Observed N(t)
l Model
Observed l
Predicted l
Primary Model
Primary Model
mmax Model
Observed mmax
Predicted mmax
Predicted N(t)
Predicted N(t)
Nmax Model
Observed Nmax
Predicted Nmax
6Performance Evaluation
Stage 1
Goodness-of-fit Primary/Secondary Models
Verification Tertiary Models
Stage 2
Interpolation All Models
Stage 3
Extrapolation All Models
7Test Data CriteriaInterpolation
- Independent data.
- Within the response surface.
- Uniform coverage.
- Collected with same methods.
Incomplete and biased evaluation Model data (10
to 40C) versus Test data (25 to 40C)
8Test Data CriteriaExtrapolation
- Independent data.
- Outside the response surface.
- Only one variable differs.
- Collected with same methods.
Confounded comparison Strain A in broth
versus Strain B in food
9Acceptable Prediction Zone MethodDescription
10Relative Error (RE)
RE for ? (predicted - observed)/predicted
RE for N(t), No, ?max and Nmax (observed -
predicted)/predicted RE lt 0 are fail-safe RE
gt 0 are fail-dangerous
11Performance Factor RE REIN/RETOTAL
12 Performance Criteria
- Acceptable Predictions
- -0.30 lt RE lt 0.15 for mmax
- -0.60 lt RE lt 0.30 for l
- -0.80 lt RE lt 0.40 for N(t), No, Nmax
- Acceptable Performance
- RE gt 70
13Acceptable Prediction Zone MethodDemonstration
14Model Development Design
- Salmonella Typhimurium
- No 4.8 log CFU/g
- Sterile cooked chicken
- 10, 12, 14, 16, 20, 24, 28, 32, 36, 38, 40C
- Viable counts
- BHI agar
- 12 per growth curve
15Performance Evaluation DesignSecondary Models
(Interpolation)
- Salmonella Typhimurium
- No 4.8 log CFU/g
- Sterile cooked chicken
- 11, 13, 15, 18, 22, 26, 30, 34, 37, 39C
- Viable counts
- BHI agar
- 12 per growth curve
16Primary ModelLogistic with Delay
N No if t ? ? N Nmax/(1(Nmax/No)-
1exp-?max (t-?)) if t gt ?
17Primary Model PerformanceGoodness-of-fit
18Secondary Model for No
No mean No
19No Model Performance
20Secondary Model for lHyperbola with Shape Factor
? 41.47/(T - 7.325)1.44
21l Model Performance
22Secondary Model for mmaxModified Square Root
?max 0.01885 if T ? 11.43 ?max
0.01885 0.004325(T 11.43)1.306 if T gt
11.43
23mmax Model Performance
24Secondary Model for NmaxAsymptote Model
Nmax exp(2.348((T 9.64)(T 40.74))/((T
9.606)(T 40.76)))
25Nmax Model Performance
26Predictive Modeling
Secondary Models
Tertiary Model
No Model
Observed No
Predicted No
Observed N(t)
l Model
Observed l
Predicted l
Primary Model
Primary Model
mmax Model
Observed mmax
Predicted mmax
Predicted N(t)
Predicted N(t)
Nmax Model
Observed Nmax
Predicted Nmax
27Tertiary Model PerformanceVerification
RE 90.7
28Comparison of Models
Model REIN REOUT RETOTAL
Primary 121 8 129
Tertiary 117 12 129
Total 238 20 258
Fishers exact test P 0.48, not significant.
29Performance Evaluation DesignTertiary Model
(Interpolation)
- Salmonella Typhimurium
- No 4.8 log CFU/g
- Sterile cooked chicken
- 11, 13, 15, 18, 22, 26, 30, 34, 37, 39C
- Viable counts
- BHI agar
- 4 per growth curve
30Tertiary Model Performance Interpolation
31Tertiary Model Performance Interpolation
RE 97.5
32Should the validated tertiary model be used to
predict chicken safety?
- Evaluation for extrapolation to
- other initial densities (No) ?
- other strains
- other chicken products
33Performance Evaluation DesignTertiary Model
(Extrapolation)
- Salmonella Typhimurium
- No 0.8 log CFU/g
- Sterile cooked chicken
- 10, 12, 14, 16, 20, 24, 28, 32, 36, 40C
- Viable counts
- BHI agar
- 4 per growth curve
34Tertiary Model Extrapolation to low No
35Tertiary Model PerformanceExtrapolation to low No
RE 2.5
36Conclusions
- Criteria are important for evaluating performance
of models. - Consensus on validation would improve the quality
and use of predictive models in the food industry.