Head of group: Abigail Anne Kressner
We believe that by recognizing the cross-disciplinary disconnects between engineers and clinicians and by working to remove these disconnects through the translation of research between DTU, the Copenhagen University Hospital, and our industrial collaborators, opportunities for improving the treatment of hearing impairment will emerge. We are especially well-suited for this cross-disciplinary work, given that we spend 50% of our time at DTU within the Hearing Systems Section and 50% of our time immersed among clinicians and their patients at the Copenhagen Hearing and Balance Center (CHBC) at Rigshospitalet. In this group, we are ultimately interested in improving the quality of life for hearing-impaired individuals. To accomplish this, we engage in both basic and applied research to better understand hearing impairment and how we can use technology to treat it.
Speech and spatial perception with hearing device technology
In this line of research, we are interested in investigating how hearing device technologies influence speech and spatial perception. On the basic science side of the spectrum, our focus lies in understanding how hearing-impaired individuals understand speech in realistic, noisy environments—for example, in restaurants, classrooms, or in meeting rooms. On the more applied side, we are interested in understanding how hearing device technologies can be designed to improve speech intelligibility and spatial perception in these situations. We approach these questions using several different methodologies, including virtual sound environments (Figure 1), real-time hearing aid and cochlear implant simulators, and patient-reported outcomes.
Figure 1 Virtual sound environments are a tool for bringing realism into the laboratory, such that a listener placed within a loudspeaker array will have the feeling that they are in an authentic real-world listening environment rather than in the laboratory. We are using virtual sound environments to assess hearing device technology under more realistic conditions than was previously possible with more traditional laboratory setups.
Selected publications:
• Niels West, Abigail Anne Kressner, Line Husted Baungaard, Matilde Grønborg Sandvej, Michael Bille, and Per Cayé-Thomasen. Nordic results of cochlear implantation in adults: Speech reception and patient reported outcomes, Acta Oto-Laryngologica 140(11):939-947, 2020.
• Thomas Bentsen, Stefan J. Mauger, Abigail Anne Kressner, Tobias May, and Torsten Dau. The impact of noise power estimation on speech intelligibility in cochlear-implant speech coding strategies, Journal of the Acoustical Society of America, 145(2):818-821, 2019.
• Abigail Anne Kressner, Adam Westermann, and Jörg Matthias Buchholz. The impact of reverberation on speech intelligibility in cochlear implant recipients, Journal of the Acoustical Society of America, 144(2):1113-1122, 2018.
• Abigail Anne Kressner and Christopher John Rozell. Structure in time-frequency binary masking errors and its impact on speech intelligibility. Journal of the Acoustical Society of America, 137(4):2025-2035, 2015.
Tools for assessing and treating hearing loss
A second, broader research stream includes the development of tools that facilitate better assessment and treatment of hearing loss. An underlying theme throughout this research effort is to maintain focus on clinical relevance. In doing so, we hope to advance clinical care, and thereby, optimize hearing healthcare overall. Topics in this direction include tools that facilitate better diagnosis and rehabilitation of patients (e.g., measuring listening effort with pupillometry and measuring speech discrimination using functional near-infrared spectroscopy), as well as tools that help facilitate further research and development (e.g., creating a new Danish speech corpus and developing a speech intelligibility metric for predicting the effect of noise reduction algorithms in cochlear implants; see Figure 2).
Figure 2 Cochlear implants aid profoundly deafened individuals by directly stimulating the auditory nerve through a set of electrodes that sit along the cochlear. The two plots above are depictions of stimulation patterns for an example Danish sentence. Few tools exist, however, for predicting the resulting perceptual outcomes of these two stimulation patterns. We are working to develop such tools.
Selected publications:
• Abigail Anne Kressner, Tobias May, and Christopher John Rozell. Outcome measures based on classification performance fail to predict the intelligibility of binary-masked speech. Journal of the Acoustical Society of America, 139(6):3033-3036, 2016.
• Abigail Anne Kressner, David V. Anderson, and Christopher John Rozell. Evaluating the generalization of the Hearing Aid Speech Quality Index (HASQI). IEEE Transactions in Audio, Speech and Language Processing, 21(2):407-415, 2013.