1
|
Kassjański M, Kulawiak M, Przewoźny T, Tretiakow D, Kuryłowicz J, Molisz A, Koźmiński K, Kwaśniewska A, Mierzwińska-Dolny P, Grono M. Automated hearing loss type classification based on pure tone audiometry data. Sci Rep 2024; 14:14203. [PMID: 38902305 PMCID: PMC11190215 DOI: 10.1038/s41598-024-64310-2] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/29/2024] [Accepted: 06/07/2024] [Indexed: 06/22/2024] Open
Abstract
Hearing problems are commonly diagnosed with the use of tonal audiometry, which measures a patient's hearing threshold in both air and bone conduction at various frequencies. Results of audiometry tests, usually represented graphically in the form of an audiogram, need to be interpreted by a professional audiologist in order to determine the exact type of hearing loss and administer proper treatment. However, the small number of professionals in the field can severely delay proper diagnosis. The presented work proposes a neural network solution for classification of tonal audiometry data. The solution, based on the Bidirectional Long Short-Term Memory architecture, has been devised and evaluated for classifying audiometry results into four classes, representing normal hearing, conductive hearing loss, mixed hearing loss, and sensorineural hearing loss. The network was trained using 15,046 test results analysed and categorised by professional audiologists. The proposed model achieves 99.33% classification accuracy on datasets outside of training. In clinical application, the model allows general practitioners to independently classify tonal audiometry results for patient referral. In addition, the proposed solution provides audiologists and otolaryngologists with access to an AI decision support system that has the potential to reduce their burden, improve diagnostic accuracy, and minimise human error.
Collapse
Affiliation(s)
- Michał Kassjański
- Department of Geoinformatics, Faculty of Electronics, Telecommunications and Informatics, Gdańsk University of Technology, G. Narutowicza 11/12, 80-233, Gdańsk, Poland.
| | - Marcin Kulawiak
- Department of Geoinformatics, Faculty of Electronics, Telecommunications and Informatics, Gdańsk University of Technology, G. Narutowicza 11/12, 80-233, Gdańsk, Poland
| | - Tomasz Przewoźny
- Department of Otolaryngology, Medical University of Gdańsk, Gdańsk, Poland
| | - Dmitry Tretiakow
- Department of Otolaryngology, The Nicolaus Copernicus Hospital in Gdańsk, Copernicus Healthcare Entity, Gdańsk, Poland
| | - Jagoda Kuryłowicz
- Department of Otolaryngology, Medical University of Gdańsk, Gdańsk, Poland
| | - Andrzej Molisz
- Department of Otolaryngology, Medical University of Gdańsk, Gdańsk, Poland
| | | | - Aleksandra Kwaśniewska
- Department of Otolaryngology, Laryngological Oncology and Maxillofacial Surgery, University Hospital No. 2, Bydgoszcz, Poland
| | | | - Miłosz Grono
- Department of Otolaryngology, Medical University of Gdańsk, Gdańsk, Poland
| |
Collapse
|
2
|
Busch S, Lenarz T, Maier H. An easy method to determine crucial AMEI performance parameters from clinical routine data in individuals - Part 2: dynamic range. Int J Audiol 2024:1-8. [PMID: 38832702 DOI: 10.1080/14992027.2024.2360031] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/27/2023] [Accepted: 05/14/2024] [Indexed: 06/05/2024]
Abstract
OBJECTIVE The dynamic range (DR) available to the patient is a central parameter to determine speech intelligibility in quiet. DESIGN In this retrospective study, the DR for the Vibrant Soundbridge implanted in individual patients was calculated using in situ thresholds of the patients and technical data of the implant system. The average DR across frequencies (0.5, 1, 2, 4 kHz) was correlated with the patients' assigned word recognition score (WRS) in quiet. STUDY SAMPLE A data set of 66 cases (4 bilateral and 2 revised cases) from 60 implanted patients between 14.3-81.8 years were analysed. RESULTS The relationship between DR and WRS was described by a sigmoidal growth function with R2=0.6371 and a maximum WRS (upper asymptote) of 93.5%. Word recognition scores in quiet improved with increasing DR. A significant shift in performance was detected from DR bin 2 (10-20 dB, median WRS 55%) to bin 3 (20-30 dB, median WRS 80%) and from DR bin 4 (30-40 dB, median WRS 82.5%) to bin 5 (40-50 dB, median WRS 90%). CONCLUSION A minimum DR of 20 dB can yield sufficient speech intelligibility in quiet in implanted patients, however, an optimum DR is suggested to be 40 dB.
Collapse
Affiliation(s)
- Susan Busch
- Department of Otorhinolaryngology and Cluster of Excellence "Hearing4all", Hannover Medical School, Hannover, Germany
| | - Thomas Lenarz
- Department of Otorhinolaryngology and Cluster of Excellence "Hearing4all", Hannover Medical School, Hannover, Germany
| | - Hannes Maier
- Department of Otorhinolaryngology and Cluster of Excellence "Hearing4all", Hannover Medical School, Hannover, Germany
| |
Collapse
|
3
|
Cohen-Vaizer M, Dreyfuss M, Na'ara S, Shinnawi S, Laske R. The Impact of Surgical Expertise on the Cost-Effectiveness of Stapes Surgery. Audiol Neurootol 2023; 28:436-445. [PMID: 37343529 DOI: 10.1159/000530783] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/07/2022] [Accepted: 04/18/2023] [Indexed: 06/23/2023] Open
Abstract
INTRODUCTION Otosclerosis is the primary cause of conductive hearing loss with normal otoscopy. As the condition worsens, certain patients may develop a sensorineural component. Patients with successful surgeries may still need hearing aids, which creates a dilemma for health professionals as there are insufficient data to make informed decisions. This study investigated the influence of the surgeon's proficiency level, individual patient factors (e.g., age at the time of intervention and survival rates), and surgery costs on the cost-effectiveness of stapes surgery. METHODS We performed a cost-effectiveness analysis using an adapted Markov model incorporating annual all-cause mortalities. In addition, we introduced sensitivity analyses to address the effects of surgical expertise on adults with bilateral conductive hearing loss due to otosclerosis. A model was developed based on a decision tree with treatment options and complication scenarios for otosclerosis patients undergoing stapes surgery or receiving hearing aids. Annual all-cause mortality was considered. A sensitivity analysis was performed assigned to different training levels ("experts" and "less experienced") to simulate the effects of surgical experience on the cost-effectiveness of surgical outcomes. Successful surgery was defined as closing of the air-bone gap to 10 dB or less. Based on published data, "experts" were simulated with a 93.7% success rate, and "less experienced" were manufactured with a 68.9% success rate. RESULTS Stapes surgery provides improved quality of life (QoL) compared to hearing aids with lower cumulative costs up to 87 years of age in the case of "expert" surgeons and up to 78 years of age, when performed by "less experienced" surgeons. CONCLUSIONS Primary stapes surgery is highly cost-effective and delivers improved QoL compared to hearing aids with lower cumulative costs. Additionally, undergoing stapes surgical training remains highly cost-effective.
Collapse
Affiliation(s)
- Mauricio Cohen-Vaizer
- Department of Otolaryngology Head and Neck Surgery, Rambam Healthcare Campus, The Technion, Israel Institute of Technology, Haifa, Israel
| | - Michael Dreyfuss
- Department of Industrial Engineering and Management, Jerusalem College of Technology, Jerusalem, Israel
| | - Shoorok Na'ara
- Department of Otolaryngology Head and Neck Surgery, Rambam Healthcare Campus, The Technion, Israel Institute of Technology, Haifa, Israel
- Department of Otolaryngology, Head and Neck Surgery, University of California at San Francisco, San Francisco, California, USA
| | - Shadi Shinnawi
- Department of Otolaryngology Head and Neck Surgery, Rambam Healthcare Campus, The Technion, Israel Institute of Technology, Haifa, Israel
| | - Roman Laske
- Department of Otolaryngology, HNO Wiedikon, Zurich, Switzerland
| |
Collapse
|
4
|
Consensus Statement on Bone Conduction Devices and Active Middle Ear Implants in Conductive and Mixed Hearing Loss. Otol Neurotol 2022; 43:513-529. [PMID: 35383700 DOI: 10.1097/mao.0000000000003491] [Citation(s) in RCA: 17] [Impact Index Per Article: 8.5] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/26/2022]
Abstract
ABSTRACT Nowadays, several options are available to treat patients with conductive or mixed hearing loss. Whenever surgical intervention is not possible or contra-indicated, and amplification by a conventional hearing device (e.g., behind-the-ear device) is not feasible, then implantable hearing devices are an indispensable next option. Implantable bone-conduction devices and middle-ear implants have advantages but also limitations concerning complexity/invasiveness of the surgery, medical complications, and effectiveness. To counsel the patient, the clinician should have a good overview of the options with regard to safety and reliability as well as unequivocal technical performance data. The present consensus document is the outcome of an extensive iterative process including ENT specialists, audiologists, health-policy scientists, and representatives/technicians of the main companies in this field. This document should provide a first framework for procedures and technical characterization to enhance effective communication between these stakeholders, improving health care.
Collapse
|
5
|
Waldmann B, Rocha Félix T, Bento M, Miranda C, Peixoto MC, Pratas R, English R, Correia da Silva V. In-vivo characterisation of an implanted microphone and totally implantable active middle ear implant. Int J Audiol 2021; 61:948-955. [PMID: 34704862 DOI: 10.1080/14992027.2021.1994158] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 10/20/2022]
Abstract
OBJECTIVE To objectively evaluate acoustic sensitivity of the implanted microphone, and maximum stable gain of a totally implantable active middle ear implant. DESIGN Prospective, single centre evaluation. STUDY SAMPLE Fourteen adult patients. RESULTS Microphone sensitivity is approx. 10 dB lower than an externally worn conventional hearing aid, at frequencies up to 4000 Hz, and substantially lower at higher frequencies. The masking level due to microphone noise, which determines the softest test tones that can be detected, is estimated at <20 to <30 dB HL up to 1000 Hz, and <40 dB HL at higher frequencies. Maximum stable effective gain is the maximum amplification achievable without causing feedback whistling. In sensorineural hearing loss (SNHL) cases, it is 30-40 dB at frequencies up to 2000 Hz, allowing to compensate for even the maximum recommended hearing loss (60-70 dB HL). In both SNHL and mixed hearing loss (MHL) cases, maximum stable effective gain is lower (+20 to -30 dB) around 3000-6000 Hz. CONCLUSIONS Microphone sensitivity is high enough to achieve aided thresholds of 20-40 dB HL. A strong correlation between actuator coupling efficiency and maximum stable effective gain implies that any effort to improve actuator efficiency should also increase the available gain.
Collapse
|
6
|
Wardenga N, Diedrich V, Waldmann B, Lenarz T, Maier H. Hearing Aid Treatment in Patients with Mixed Hearing Loss. Part I: Expected Benefit and Limitations after Stapes Surgery. Audiol Neurootol 2020; 25:125-132. [PMID: 32045919 DOI: 10.1159/000502994] [Citation(s) in RCA: 4] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/06/2018] [Accepted: 12/10/2018] [Indexed: 12/26/2022] Open
Abstract
OBJECTIVE The purpose of the present study was to determine the fraction of patients with mixed hearing loss who can or cannot expect benefit from power hearing aids (HAs) after stapes surgery. DESIGN The audiological outcome of 374 stapes surgeries was used to calculate the patients' individual postoperative requirements in terms of gain and output of HAs. These requirements were compared to the available gain and output provided by state-of-the-art power HAs at 0.5, 1.0, 2.0, and 4.0 kHz. According to these comparisons, ears were divided into three groups. For G0, required gain and output lay within the corresponding technical limits of the HAs at all frequencies. In G1, one or both requirements could not be fulfilled at 1 frequency. G2 combined all ears where the requirements lay beyond the HA's technical limitations at 2 or more frequencies. RESULTS Stapes surgery resulted in an improvement of air-bone gap (ABG) in 84.5% of the cases by 15.7 dB on average. Based on pure-tone average (0.5, 1.0, 2.0, 4.0 kHz), 40.6% of all cases showed an ABG ≤10 dB. 44.9% of all cases did no longer need a HA after stapes surgery. A power HA would fulfill both audiological criteria at all 4 frequencies in 81.6% of cases that needed a HA postoperatively. However, 18.4% would not be sufficiently treatable at 1 or more frequencies (15.0% in G1, 3.4% in G2). CONCLUSIONS The present study identified a subset of patients with mixed hearing loss after stapes surgery that cannot be treated sufficiently with available power HAs. As the residual ABG is an important reason for this lack of treatment success, the advancement of alternative hearing devices that circumvent the middle ear, such as powerful active middle ear implants, is indicated.
Collapse
Affiliation(s)
- Nina Wardenga
- Cluster of Excellence Hearing4all, Hannover, Germany, .,Department of Otolaryngology, Hannover Medical School, Hannover, Germany,
| | - Victoria Diedrich
- Department of Otolaryngology, Hannover Medical School, Hannover, Germany
| | | | - Thomas Lenarz
- Cluster of Excellence Hearing4all, Hannover, Germany.,Department of Otolaryngology, Hannover Medical School, Hannover, Germany
| | - Hannes Maier
- Cluster of Excellence Hearing4all, Hannover, Germany.,Department of Otolaryngology, Hannover Medical School, Hannover, Germany
| |
Collapse
|