|
Dear list,
3 PhD positions are opened in Oldenburg, Germany (
https://uol.de/stellen/?stelle=67452 ). If you know of anyone who might be interested in such a position, please forward the information below.
Thank you!
Best regards,
Anna Warzybok
Description of PhD research projects The overarching aim of the three collaborating PhD projects is to contribute to the construction of a "virtual hearing clinic" for mobile personal devices that combines self-controlled auditory diagnostics, machine-learning supported classification of hearing disorders, and provisions to fit and test "virtual" hearing devices to derive predictions about the individual benefit from a real hearing aid. Depending on the skills and interests of the successful candidates, the three projects will combine an individual mixture of basic research, audiological evaluations with normal and hearing-impaired volunteers, data science, auditory modelling, algorithm and software development, and project management as follows: Position A: Psychophysics and medical classification The psychophysical and psychometrical methods involved in multimodal tests (including auditory tests) with mobile devices need to be optimized and tested for robustness, reproducibility, and validity for classifying the underlying hearing disorder. One goal is to create an auditory profile with a minimum set of highly controlled and efficient clinical-audiological tests that allow the fitting of relevant individual model parameters. A second goal is to develop medical classification methods operating on these sets of tests in combination with auditory models and with machine learning methods such as Bayesian nets for interpreting the results and prediction of missing data. Position B: Speech recognition with multilingual tests with normal and hearing-impaired listeners The self-controlled, optimized tests to be run on the virtual hearing clinic include speech recognition tests in noise that will be further developed together with international partners to address more languages (covering >90 % of EU and > 65% of world population), and to be more realistic in real life (e.g. using virtual acoustic technology) in order to assess the individual benefit from advanced hearing solutions. One goal is to systematically assess the differences across tonal and non-tonal languages in collaboration with our Chinese partners and to both empirically and theoretically (i.e., using auditory models) assess the influence of speaker, language, the Lombard effect, and reverberation on speech recognition. Position C: Development of a virtual (hearing) clinic for mobile devices The aim of this project is to integrate model-based audiological, multilingual diagnostics, statistical classification and virtual hearing aid simulation into consistent, user-centered building blocks for the "virtual hearing clinic" that can be used for mobile Health "virtual clinics" solutions in other clinical disciplines as well. By the use of a fixed, optimised test set, even a reduced, screening-test-like approach within the virtual (hearing) clinic should provide enough information to sufficiently characterize and at least coarsely classify the indivudual (hearing) disorder as a means for deriving a treatment recommendation. The emphasis in this project will be put on data management, statistical learning methods, usability engineering, and empirical studies on the feasibility, strengths and shortcomings of the "virtual (hearing) clinic" with a growing set of voluntary test subjects and patients. Research Environment: |