Multicentre clinical trial shows benefits of CrystalisXDP

A recent multicentre study evaluated the benefits of CrystalisXDP, the new signal processing strategy for cochlear implants introduced by the Saphyr® neo sound processor. Compared to former versions of the signal processing strategy, CrystalisXDP is correlated with improved speech perception in quiet and in noise, as well as with a subjective improvement in sound quality.

Speech perception score: Bozorg-Grayeli and colleagues from five different university hospitals in France showed that Oticon Medical cochlear implant users (20 adults) benefited from a sound processor upgrade introducing CrystalisXDP

Bozorg-Grayeli and colleagues from five different university hospitals in France1 showed that Oticon Medical cochlear implant users (20 adults) benefited from a sound processor upgrade introducing CrystalisXDP.
Improvements were observed in speech perception scores in quiet at different intensities, in speech perception in noise, and in a global improvement in sound quality. In particular, when compared to the former signal processing strategy, speech perception in noise had improved by 30–40% after a 30-day habituation period. 
Speech recognition was first assessed in quiet at 40, 55, 70 and 85 dB SPL. CrystalisXDP was set to its medium preset, optimised for sounds with an average intensity of 70 dB SPL. As seen in Figure 1, all scores were higher with CrystalisXDP, either on the day of renewal or after the habituation period had ended. Scores gradually increased for speech sounds presented at 55 dB SPL, reaching a significant +8.5% compared to the standard strategy after the habituation period. Improvements were also observed for higher intensities, with +7.8% at 70 dB SPL (significant at Day 0), and a +17.7% at 85 dB SPL at Day 30, suggesting a better representation of loud sounds.

Figure 1: Vocal audiometry in quiet at 40, 55, 70 and 85 dB (n = 20)

Vocal audiometry in noise

Speech-in-noise perception was assessed by presenting isolated words together with noise from a cocktail party, at a signal-to-noise ratio (SNR) of +10 dB. Group results showed that scores increased from the standard strategy (27.5%) to CrystalisXDP (32.7%) on Day 0 and on Day 30 (54.6%), leading to an average improvement of +27.1% improvement for the 20 participants after the habituation period. 
To verify that this effect was really due to the change in signal processing strategy and other factors present during the habituation period, the authors obtained supplementary data on Day 30 for the standard strategy on nine participants (Figure 2). Within this subgroup, speech-in-noise performance appeared to already be better on Day 0 with CrystalisXDP than with standard, with a +10.4 % advantage for CrystalisXDP; these findings became statistically significant after the habituation period and culminated with a +40% improvement for speech perception in noise compared to the standard strategy.

Figure 2. Vocal audiometry in noise (10-dB SNR cocktail party noise, n = 9) 

Subjective sound quality was assessed using a short questionnaire (10 items) derived from the APHAB questionnaire, asking users to rate on a numerical scale from 0 (very bad) to 10 (very good) their perceived sound quality in ten different everyday life listening situations (perception of speech in quiet and in noise, sound quality in quiet and in noise, experienced listening effort sound quality when listening to the TV or using the phone, quality of music sounds, quality in social interaction settings and overall impression.

Subjective sound quality appraisal

As shown in Figure 3, all subjective sound quality ratings where higher with CrystalisXDP, with a statistically significant change in main effect of strategy. Marked differences (*significant differences after planned comparisons) were observed in the pooled items: Noise: Speech intelligibility, sound quality in noise, subjective listening effort and social interaction; Leisure: TV, music and telephone. These results demonstrate that a signal processing strategy optimised for speech perception and associating a wide input dynamic range to a back-end multi-channel compression system like CrystalisXDP can improve speech perception and overall satisfaction in cochlear implant users. 

Figure 3. Subjective sound quality appraisal (score out of 10) 

Reference List 

1. Bozorg-Grayeli, A., Guevara, N., Bébéar, J.P., Ardoint, M., Saaï, S., Hoen, M., Gnansia, D., Romanet, P., Lavieille, J.P. (2015) Clinical evaluation of the XDP output compression strategy for cochlear implants. European Archives of Oto-Rhino-Laryngology. 17 Oct 2015.