Wyniki wyszukiwania - Biblioteka Nauki

1

Hearing tester - mobile application for hearing loss screening

100%

Krzyżek K. , Kłaczyński M.

|

tom Vol. 20, No. 3

87--95

EN

The article presents results of the work on creation of an application for Android mobile devices to perform hearing screening using pure-tone air conduction audiometry. The application was created in accordance with the procedure defined in the ISO 8253-1: 2010 standard and is to serve the widest possible group of people in a domestic environment, using headphones included with mobile devices. The effectiveness of the application was verified by performing comparative hearing tests on a representative group of patients with different hearing status. Validation was performed using the Interacoustics AD629 clinical audiometer and procedures adopted in the standards. The results of the test comparison were satisfactory. The mean error of the hearing threshold tested using the mobile application, compared to the clinical audiometer was 5.2 dBHL. In addition, the hearing threshold curves for the mobile application are similar in shape to the hearing thresholds for the clinical audiometer and coincide with the auditory experiences of the patients.

PL

Artykuł przedstawia wyniki prac nad stworzeniem aplikacji na urządzenia mobilne z systemem Android w celu wykonania przesiewowych badań słuchu w zakresie progowej audiometrii tonalnej przy przewodnictwie powietrznym. Aplikacja została stworzona zgodnie z procedurą określoną w normie ISO 8253-1:2010 i ma służyć jak najszerszej grupie osób w domowych warunkach badań przy wykorzystaniu słuchawek dedykowanych urządzeniom mobilnym. Działanie aplikacji zostało zweryfikowane poprzez wykonanie porównawczych badań słuchu na reprezentatywnej grupie pacjentów o zróżnicowanym stanie słuchu. Walidację przeprowadzono przy użyciu audiometru klinicznego Interacoustics AD629 i procedur przyjętych w normach. Wyniki badań porównawczych okazały się zadowalające. Średni błąd zbadanego progu słyszenia przy pomocy aplikacji mobilnej, w porównaniu do badań przeprowadzonych przy pomocy audiometru klinicznego, wyniósł 5.2 dBHL. Dodatkowo, krzywe progu słyszenia dla aplikacji mobilnej są zbliżone w kształcie do krzywych progu słyszenia dla audiometru klinicznego i pokrywają się z wrażeniami słuchowymi pacjentów.

2

Usefulness of the matching pursuit method in phonocardiographic signal analysis

100%

Damps N. , Kłaczyński M.

Vibrations in Physical Systems

|

2021

|

tom Vol. 32, nr 1

art. no. 2021102

EN

This paper presents a phonocardiographic signal analysis with special emphasis on the Matching Pursuit method. To the knowledge of the authors, this method has not been used before to analyze PCG (phonocardiogram) signals. For this reason, its usefulness for this signal type was tested and a dictionary of Gabor atoms was created. Based on these findings, PCG signal analysis was performed as a Wigner-Ville distribution and compared with a spectrogram. Observing the obtained graphs, it was found that the Wigner-Ville map gives more detailed information about the frequencies which make up the given signal and the time of their occurrence. This method can be used to detect anomalies and pathologies of the heart.

3

Parkinson’s disease diagnostics using AI and natural language knowledge transfer

80%

Chronowski M. , Kłaczyński M. , Dec-Ćwiek M. , Porębska K.

|

tom Vol. 25, No. 1

art. no. 2024103

EN

With global life expectancy rising every year, ageing-associated diseases are becoming an increasingly important problem. Very often, successful treatment relies on early diagnosis. In this work, the issue of Parkinson's disease (PD) diagnostics is tackled. It is particularly important, as there are no certain antemortem methods of diagnosing PD - meaning that the presence of the disease can only be confirmed after the patient's death. In our work, we propose a non-invasive approach for classification of raw speech recordings for PD recognition using deep learning models. The core of the method is an audio classifier using knowledge transfer from a pretrained natural language model, namely wav2vec 2.0. The model was tested on a group of 38 PD patients and 10 healthy persons above the age of 50. A dataset of speech recordings acquired using a smartphone recorder was constructed and the recordings were labelled as PD/non-PD with the severity of the disease additionally rated using Hoehn-Yahr scale. We then benchmarked the classification performance against baseline methods. Additionally, we show an assessment of human-level performance with neurology professionals.

4

Experimental verification of the usefulness of selected infrasound and low-frequency noise (ILFN) indicators in assessing the noise annoyance of wind turbines

61%

Wszołek T. , Mleczko D. , Pawlik P. , Kłaczyński M. , Małecki P. , Stępień B.

|

2022

|

tom Vol. 33, nr 3

art. no. 2022315

EN

This paper presents an overview of the indices used in evaluating ILFN noise, based on C and G weighting curves and LC-LA difference parameter, as well as curves compared to the loudness threshold curve. The research section includes measurement results of wind turbine (WT) noise along with proposed indicators for evaluating this noise in the infrasound and low-frequency bands at distances of 250 m, 500 m and 1000 m from the turbine. The results obtained indicate low noise levels in the infrasound band, lower than the threshold curves from a dozen or so dB in the upper part of this band to nearly 60 dB in the lower part. The LC-LA indicator has been shown to be of poor utility for evaluating low-frequency noise, with the LG indicator reasonably useful for evaluating infrasound noise.