IA-AMTAS has eight quality indicators, which can be used to assess the accuracy of the overall result.
- The first quality indicator refers to accuracy. This looks at the predicted average absolute difference between automated and manually tested thresholds.
- The second quality indicator relates to the masker alert rate. The masker alert rate is the proportion of thresholds for which masking noise was presented to the non-test ear at a level which may have been too low or too high. This is then divided by the number of measured thresholds. This tells us if there were any cases of under-masking or over-masking.
- The third quality indicator is time per trial. Time per trial informs us of the elapsed time from the onset of the stimulus to the subject response averaged across all responses.
- The fourth quality indicator is false alarm rate. The false alarm rate is the number of trials in which the subject reported the presence of a stimulus and no stimulus was presented. This is then divided by the total number of catch trials (trials in which there was no stimulus presented). This gives us information on whether the patient was guessing or not.
- The fifth quality indicator is the average test-retest difference. The average test-retest difference is the difference between the 1000 Hz test and the retest thresholds measured.
- The sixth quality indicator is the quality check fail rate. The quality check fail rate is the number of occurrences of quality check fails. This is when the subject fails to respond to stimuli presented at super threshold levels.
- The last two quality indicators relate to air-bone gaps. The seventh quality indicator looks at air-bone gaps which are greater than 35 dB, and the eight-quality indicator looks at the number of air-bone gaps below 10 dB.
Together these eight quality indicators will then provide us with an overall value of the quality of the automated audiometry. This is then presented in the software as a traffic light system. If the quality is indicated as red, then this suggests that the quality of the results is poor. Amber tells us the quality of the results is fair, and green tells us the quality of the results is good. In addition to obtaining an automated audiogram the clinician can then identify the overall quality of the audiogram and then make further investigations or management decisions depending on that quality.