New article online: Robust S1 and S2 heart sound recognition based on spectral restoration and multi-style training

Robust S1 and S2 heart sound recognition based on spectral restoration and multi-style training

Biomedical Signal Processing and Control, 49: 173-180 (2019)
https://www.sciencedirect.com/science/article/pii/S1746809418302787?via%3Dihub

Yu Tsao, Tzu-Hao Lin
Research Center for Information Technology Innovation (CITI) at Academia Sinica, Taipei, Taiwan

Fei Chen
Department of Electrical and Electronic Engineering, Southern University of Science and Technology, Xueyuan Road 1088#, Xili, Nanshan District, Shenzhen, China

Yun-Fan Chang, Chui-Hsuan Cheng, Kun-Hsi Tsai
iMediPlus Inc., Hsinchu, Taiwan

Recently, we have proposed a deep learning based heart sound recognition framework, which can provide high recognition performance under clean testing conditions. However, the recognition performance can notably degrade when noise is present in the recording environments. This study investigates a spectral restoration algorithm to reduce noise components from heart sound signals to achieve robust S1 and S2 recognition in real-world scenarios. In addition to the spectral restoration algorithm, a multi-style training strategy is adopted to train a robust acoustic model, by incorporating acoustic observations from both original and restored heart sound signals. We term the proposed method as SRMT (spectral restoration and multi-style training). The experimental procedure in this study is described as follows: First, an electronic stethoscope was used to record actual heart sounds, and the noisy signals were artificially generated at different signal-to-noise-ratios (SNRs). Second, an acoustic model based on deep neural networks (DNNs) was trained using original heart sounds and heart sounds processed through spectral restoration. Third, the performance of the trained model was evaluated using the following metrics: accuracy, precision, recall, specificity, and F-measure. The results confirm the effectiveness of the proposed method for recognizing heart sounds in noisy environments. The recognition results of an acoustic model trained on SRMT outperform that trained on clean data with a 2.36% average accuracy improvement (from 85.44% and 87.80%), over clean, 20dB, 15dB, 10dB, 5dB, and 0dB SNR conditions; the improvements are more notable in low SNR conditions: the average accuracy improvement is 3.87% (from 82.83% to 86.70%) in the 0dB SNR condition.

Before February 03, 2019, you can download the pdf copy from this link.

Advertisements

New article online: The Effects of Continuous Acoustic Stress on ROS Levels and Antioxidant-related Gene Expression in the Black Porgy

 

The Effects of Continuous Acoustic Stress on ROS Levels and Antioxidant-related Gene Expression in the Black Porgy (Acanthopagrus schlegelii)

Zoological Studies 57: 59 (2018)
http://zoolstud.sinica.edu.tw/Journals/57/57-59.html

Hao-Yi Chang, Yi Ta Shao
Institute of Marine Biology, National Taiwan Ocean University

Tzu-Hao Lin
Department of Marine Biodiversity Research, Japan Agency for Marine-Earth Science and Technology

Kazuhiko Anraku
Fisheries Department, Kagoshima University

Short-term exposure to strong underwater noise is known to seriously impact fish. However, the chronic physiological effects of continuous exposure to weak noise, i.e. the operation noise from offshore wind farms (OWF), remain unclear. Since more and more OWF will be built in the near future, their operation noise is an emerging ecological issue. To investigate the long-term physiological effects of such underwater noise on fish, black porgies (Acanthopagrus schlegelii) were exposed to two types of simulated wind farm noise—quiet (QC: 109 dB re 1 μPa / 125.4 Hz; approx. 100 m away from the wind turbine) and noisy (NC: 138 dB re 1 μPa / 125.4 Hz; near the turbine)—for up to 2 weeks. Measurement of auditory-evoked potentials showed that black porgies can hear sound stimuli under both NC and QC scenarios. Although no significant difference was found in plasma cortisol levels, the fish under NC conditions exhibited higher plasma reactive oxygen species (ROS) levels than the control group at week 2. Moreover, alterations were found in mRNA levels of hepatic antioxidant-related genes (sod1, cat and gpx), with cat downregulated and gpxupregulated after one week of QC exposure. Our results suggest that the black porgy may adapt to QC levels of noise by modulating the antioxidant system to keep ROS levels low. However, such antioxidant response was not observed under NC conditions; instead, ROS accumulated to measurably higher levels. This study suggests that continuous OWF operation noise represents a potential stressor to fish. Furthermore, this is the first study to demonstrate that chronic exposure to noise could induce ROS accumulation in fish plasma.

New article online: The effects of continuously acoustical stress on cortisol in milkfish

The effects of continuously acoustical stress on cortisol in milkfish (Chanos Chanos)

General and Comparative Endocrinology (2017)
https://doi.org/10.1016/j.ygcen.2017.07.018

Chih An Wei, Yi-Ta Shao*
Institute of Marine Biology, National Taiwan Ocean University

Tzu-Hao Lin
Research Center for Information Technology Innovation, Academia Sinica

Ruo Dong Chen
Institute of Cellular and Organismic Biology, Academia Sinica

Yung-Che Tseng
Marine Research Station, Institute of Cellular and Organismic Biology, Academia Sinica

Strong underwater acoustic noise has been known that may cause hearing loss and actual stress in teleost. However, the long-term physiological effects of relatively quiet but continuously noise on fish were less understood. In present study, milkfish, Chanos chanos, were exposed to the simulated-wind farm noise either quiet (109 dB re 1 μPa / 125.4 Hz; approx. 10-100m distant from the wind farm) or noisy (138 dB re 1 μPa / 125.4 Hz; nearby the wind farm) conditions for 24 hr, 3 days and 1 week. Comparing to the control group (80 dB re 1 μPa / 125.4 Hz), the fish exposed to noisy conditions had higher plasma cortisol levels in the first 24 hr. However, the cortisol levels of 24 hr spot returned to the resting levels quickly. The fish exposed under noisy condition had significantly higher head kidney star (steroidogenic acute regulatory) and hsd11b2 (11-β-hydroxysteroid dehydrogenase 2) mRNA levels at the following treatment time points. In addition, noise exposure did not change hypothalamus crh (Corticotropin-releasing hormone) mRNA levels in this experiment. The results implied that the weak but continuously noise was a potential stressor to fish, but the impacts may be various depending on the sound levels and exposure time. Furthermore, this study showed that the continuous noise may up-regulate the genes that are related to cortisol synthesis and possibly make the fish more sensitive to ambient stressors, which may influence the energy allocation appearance in long-term exposures.