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.

The objective of the present study was to characterize the effectiveness of the audiological and radiological diagnostic techniques as well as the results of the surgical treatment of the patients presenting with otosclerosis based at the hospital clinic of the Pacific State Medical University. The analysis included 49 surgical interventions for stapedoplasty carried out on 42 patients during the period from 2015 to 2017. The tympanic form of the disorder was diagnosed in 51.02% of the cases, the mixed form I in 16.28%, and the mixed form II in 32.7%. The excellent results within two months after surgery were obtained in 87.2% of the treated patients, the good and satisfactory results in 8.72% and 4.08% of them respectively.
Цель исследования - характеристика аудиологических, рентгенологических методов диагностики и результатов 49 операций при отосклерозе, выполненных с 2015 по 2017 г. в клинике Тихоокеанского государственного медицинского университета. Тимпанальная форма диагностирована у 51,02% пациентов, смешанная I - у 16,28%, смешанная II - у 32,7%. Через 2 мес отличные результаты были достигнуты у 87,2% больных, хорошие - у 8,72%, удовлетворительные - у 4,08%.

OBJECTIVE: Newborns from Neonatal intensive care units (NICU) are at high-risk for sensorineural hearing loss (SNHL) a follow-up is needed for early diagnosis and intervention. Our objective here was to describe the features and changes of SNHL at different periods during a follow-up of almost 20 years.
METHODS: Risk factors for SNHL during development were analyzed. The audiological examination included: Brainstem auditory evoked potentials (BAEP), and Transient evoked otoacoustic emissions (TEOAE). At birth; tonal audiometry (between 125 and 8000 Hz), and tympanometry were performed at 5, 10, 15, and 20 years of age.
RESULTS: Sixty-five percent of cases presented bilateral absence of BAEP. At 5 years of age, the most frequent SNHL level was severe (42.5%), followed by moderate (22.5%), and profound level (20%), in all cases, the SNHL was symmetrical with a predominance of lesion for the high frequencies. Exchange transfusion was associated with a higher degree of SNHL (OR = 6.00, CI = 1.11-32.28, p < 0.02). In 55%, SNHL remained stable, but in 40% of the cases it was progressive. At the end of the study six cases with moderate loss progressed to the severe level and seven cases with severe level progressed to profound.
CONCLUSIONS: Forty percent of infants with SNHL discharged from NICU may present a progression in the hearing loss. Exchange transfusion was associated with a higher degree of SNHL. NICU graduates with SNHL merit a long-term audiological follow-up throughout their lifespan.

This study aimed to quantify outcomes in a group of patients who were implanted with an Oticon Medical Neurelec (Vallauris, France) cochlear implant system, the Digisonic(®) SP/Saphyr(®) Neo. Ten participants took part in this preliminary study. Their speech perception capacities were evaluated at 3, 6, and 12-months after cochlear implant activation and compared to pre-implantation scores and to scores observed with former versions of the sound processor. Compared to former versions of the sound processor, patients using the Saphyr(®) Neo processor obtained better speech perception scores for sentences in silence at each tests session (3 months: 79%, 6 months: 82% and 12 months: 94%) compared to Digisonic(®) users (respectively: 58%, 69% and 75%) and Convex sound processor users (resp. 39%, 59% and 51%). These observations confirm that the technological improvements made in the Saphyr(®) Neo sound processor coupled with the Digisonic(®) implant, provided quantifiable benefits in speech perception in Quiet compared to former versions of the processor Convex and Digisonic(®) SP.

OBJECTIVE: Occupational exposure to anaestethic gases has been suggested to induce auditory damages. The aim of this study is to investigate high-frequency audiometric responses in subjects exposed to anaesthetic gases, in order to highlight the possible effects on auditory system.
METHODS: The study was performed on a sample of 30 medical specialists of Messina University Anaesthesia and Intensive care. We have used tonal audiometry as well as high-frequency one. We have compared the responses with those obtained in a similar control group not exposed to anaesthetic gases. Results were compared statistically.
RESULTS: Results show a strong correlation (p = 0.000) between left and right ear responses to all the audiometric tests. The exposed and the control group run though the standard audiometry analysis plays different audiometric responses up only to higher frequencies (2000 HZ p = 0.009 and 4000 Hz p = 0.04); in high-frequency audiometry, as all other frequencies, the attention is drew to the fact that the sample groups distinguish themselves in a significantly statistic way (10,000 Hz p = 0.025, 12,000 Hz p = 0.008, 14,000 Hz p = 0.026, 16,000 Hz p = 0.08). The highest values are the ones related to exposed subjects both in standard (2000 Hz p = 0.01, 4000 Hz p = 0.02) and in high-frequency audiometry (10,000 Hz p = 0.011, 12,000 Hz p = 0.004, 14,000 Hz p = 0.012, 16,000 Hz p = 0.004).
CONCLUSIONS: Results, even if preliminary and referred to a low-range sample, show an involvement of the anatomic structure responsible for the perception of high-frequency audiometric responses in subjects exposed to anaesthetic gases.