Search

Latest developments in acoustic research suggest that using surveying methods based on artificial intelligence (AI) could improve the effectiveness of underwater monitoring. Passive acoustic monitoring (PAM) has proven to be a cost-effective approach for gathering information about the acoustic behavior of dolphins and plays a crucial role in studying their vocalizations, particularly whistles. This study investigates the efficiency of a binary convolutional neural network (CNN) in detecting dolphin whistles amidst high-density vocalizations in an aquatic environment. Specifically, this analysis intends to determine whether a properly trained CNN can recognize a single whistle even in challenging condition, including situations where multiple dolphins vocalize simultaneously, resulting in overlapping whistles that may have different shapes and durations. To this aim, experimental trials were conducted at Oltremare marine park, Riccione, Italy, where underwater recordings of seven-dolphin vocalizations were collected over 22 consecutive hours. The CNN was trained on labeled whistle spectrograms. The model, comprising three convolutional layers followed by max pooling layers and rectified linear unit (ReLU) activation functions, was evaluated using a 10-fold cross-validation approach. Confusion matrix and performance metrics indicate that the proposed approach achieves results comparable to those reported in the literature, despite the more challenging working conditions. The study supports the potential of AI models in enhancing passive acoustic monitoring techniques.

Abstract Globally, interactions between fishing activities and dolphins are cause for concern due to their negative effects on both mammals and fishermen. The recording of acoustic emissions could aid in detecting the presence of dolphins in close proximity to fishing gear, elucidating their behavior, and guiding potential management measures designed to limit this harmful phenomenon. This data descriptor presents a dataset of acoustic recordings (WAV files) collected during interactions between common bottlenose dolphins (Tursiops truncatus) and fishing activities in the Adriatic Sea. This dataset is distinguished by the high complexity of its repertoire, which includes various different typologies of dolphin emission. Specifically, a group of free-ranging dolphins was found to emit frequency-modulated whistles, echolocation clicks, and burst pulse signals, including feeding buzzes. An analysis of signal quality based on the signal-to-noise ratio was conducted to validate the dataset. The signal digital files and corresponding features make this dataset suitable for studying dolphin behavior in order to gain a deeper understanding of their communication and interaction with fishing gear (trawl).

The assessment of muscle-recruitment timing from electromyography (EMG) signal is relevant in different fields, including clinical gait analysis and robotic systems to interpret user’s motion intention. However, available methods typically provide only information in time domain without evaluating muscle-activation frequency content. This study aims to propose a novel adaptative algorithm for detecting muscle activation in time-frequency domain based on continuous wavelet transform (CWT) analysis. Precisely, the novel contribution of the proposed algorithm consists of evaluating the frequency range of every muscle activations detected in time domain. Performances are evaluated on a test bench of 720 simulated and 105 real surface EMG signals, stratified for signal-to-noise ratio (SNR), and then validated against different reference algorithms. Outcomes indicate that the proposed approach can provide an accurate prediction of muscle onset and offset timing in both simulated (mean absolute error, MAE $\approx 10$ ms) and real datasets (MAE < 30 ms), minimally affected by the SNR variability and compatible with the timing of EMG-driven assistive devices. Concomitantly, the maximum frequency of the activations is computed, ranging from around 100 Hz up to almost 500 Hz. This suggests a large within-muscle between-muscle variability of the frequency range. In conclusion, the current study introduces a novel reliable wavelet-based algorithm to detect both time and frequency content of muscle activation, suitable in different conditions of signal quality.

Passive acoustics are widely used to monitor the presence of dolphins in the marine environment. This study aims to introduce a low-cost and homemade approach for assembling a complete underwater microphone (i.e., the hydrophone), employing cheap and easy to obtain components. The hydrophone was assembled with two piezo disks connected in a balanced configuration and encased in a plastic container filled with plastic foam. The hydrophone’s performance was validated by direct comparison with the commercially available AS-1 hydrophone (Aquarian Hydrophones, Anacortes, U.S.) on different underwater acoustic signals: artificial acoustic signals (ramp and multitone signals) and various dolphin vocalizations (whistle, echolocation clicks, and burst pulse signals). The sensitivity of the device’s performance to changes in the emission source position was also tested. The results of the validation procedure on both artificial signals and real dolphin vocalizations showed that the significant cost savings associated with cheap technology had a minimal effect on the recording device’s performance within the frequency range of 0–35 kHz. At this stage of experimentation, the global cost of the hydrophone could be estimated at a few euros, making it extremely price competitive when compared to more expensive commercially available models. In the future, this effective and low-cost technology would allow for continuous monitoring of the presence of free-ranging dolphins, significantly lowering the total cost of autonomous monitoring systems. This would permit broadening the monitored areas and creating a network of recorders, thus improving the acquisition of data.

Abstract Background Machine learning models were satisfactorily implemented for estimating gait events from surface electromyographic (sEMG) signals during walking. Most of them are based on inter-subject approaches for data preparation. Aim of the study is to propose an intra-subject approach for binary classifying gait phases and predicting gait events based on neural network interpretation of sEMG signals and to test the hypothesis that the intra-subject approach is able to achieve better performances compared to an inter-subject one. To this aim, sEMG signals were acquired from 10 leg muscles in about 10.000 strides from 23 healthy adults, during ground walking, and a multi-layer perceptron (MLP) architecture was implemented. Results Classification/prediction accuracy was tested vs. the ground truth, represented by the foot–floor-contact signal provided by three foot-switches, through samples not used during training phase. Average classification accuracy of 96.1 ± 1.9% and mean absolute value (MAE) of 14.4 ± 4.7 ms and 23.7 ± 11.3 ms in predicting heel-strike (HS) and toe-off (TO) timing were provided. Performances of the proposed approach were tested by a direct comparison with performances provided by the inter-subject approach in the same population. Comparison results showed 1.4% improvement of mean classification accuracy and a significant (p

Background: Muscle co-contraction plays a significant role in motion control. Available detection methods typically only provide information in the time domain. The current investigation proposed a novel approach for muscle co-contraction detection in the time–frequency domain, based on continuous wavelet transform (CWT). Methods: In the current study, the CWT-based cross-energy localization of two surface electromyographic (sEMG) signals in the time–frequency domain, i.e., the CWT coscalogram, was adopted for the first time to characterize muscular co-contraction activity. A CWT-based denoising procedure was applied for removing noise from the sEMG signals. Algorithm performances were checked on synthetic and real sEMG signals, stratified for signal-to-noise ratio (SNR), and then validated against an approach based on the acknowledged double-threshold statistical algorithm (DT). Results: The CWT approach provided an accurate prediction of co-contraction timing in simulated and real datasets, minimally affected by SNR variability. The novel contribution consisted of providing the frequency values of each muscle co-contraction detected in the time domain, allowing us to reveal a wide variability in the frequency content between subjects and within stride. Conclusions: The CWT approach represents a relevant improvement over state-of-the-art approaches that provide only a numerical co-contraction index or, at best, dynamic information in the time domain. The robustness of the methodology and the physiological reliability of the experimental results support the suitability of this approach for clinical applications.

Background: Muscular-activity timing is useful information that is extractable from surface EMG signals (sEMG). However, a reference method is not available yet. The aim of this study is to investigate the reliability of a novel machine-learning-based approach (DEMANN) in detecting the onset/offset timing of muscle activation from sEMG signals. Methods: A dataset of 2880 simulated sEMG signals, stratified for signal-to-noise ratio (SNR) and time support, was generated to train a hidden single-layer fully-connected neural network. DEMANN’s performance was evaluated on simulated sEMG signals and two different datasets of real sEMG signals. DEMANN was validated against different reference algorithms, including the acknowledged double-threshold statistical algorithm (DT). Results: DEMANN provided a reliable prediction of muscle onset/offset in simulated and real sEMG signals, being minimally affected by SNR variability. When directly compared with state-of-the-art algorithms, DEMANN introduced relevant improvements in prediction performances. Conclusions: These outcomes support DEMANN’s reliability in assessing onset/offset events in different motor tasks and the condition of signal quality (different SNR), improving reference-algorithm performances. Unlike other works, DEMANN’s adopts a machine learning approach where a neural network is trained by only simulated sEMG signals, avoiding the possible complications and costs associated with a typical experimental procedure, making this approach suitable to clinical practice.

Background: Surface electromyography (SEMG) is widely used in clinics for assessing muscle functionality. All procedures proposed for noise reduction alter SEMG spectrum, especially in the low-frequency band (below 30 Hz). Indeed, low-frequency band is generally addressed to motion artifacts and electrocardiogram (ECG) interference without any further investigation on the possibility of SEMG having significant spectral content. The aim of the present study was evaluating SEMG frequency content to understand if low-frequency spectral content is negligible or, on the contrary, represents a significant SEMG portion potentially providing relevant clinical information. Method: Isometric recordings of five muscles (sternocleidomastoideus, erectores spinae at L4, rectus abdominis, rectus femoris and tibialis anterior) were acquired in 10 young healthy voluntary subjects. These recordings were not affected by motion artifacts by construction and were pre-processed by the Segmented-Beat Modulation Method for ECG deletion before performing spectral analysis. Results: Results indicated that SEMG frequency content is muscle and subject dependent. Overall, the 50th[25th;75th] percentiles spectrum median frequency and spectral power below 30 Hz were 74[54; 87] Hz and 18[10; 31] % of total (0–450 Hz) spectral power. Conclusions: Low-frequency spectral content represents a significant SEMG portion and should not be neglected. Keywords: Surface electromyographic signal, Electromyographic spectrum, Segmented-Beat Modulation Method, Non-linear filtering, Spectral analysis

Denture stomatitis (DS) is an inflammatory status of oral mucosae frequently observed in denture wearers, and mainly associated with oral overgrowth of Candida albicans. DS is the cause of multiple visits to the dental office and is thought to enhance the risk of systemic infections. The treatment of DS mainly relies upon improvement of oral hygiene measures and prescription of topical or systemic antifungal agents, and disinfectants that, although effective, are not without drawbacks. Since, in recent years, some probiotics were investigated as a means to contrast oral colonization by Candida spp., this study was designed to preliminarily evaluate the effects of probiotic strain Streptococcus salivarius K12, in subjects affected by DS, and the duration of these effects. Fifty adult denture wearers affected by DS were enrolled and randomly divided into two groups: the experimental group was instructed to perform careful oral and denture hygiene and to assume the probiotic preparation for 30 days; the control group received only oral hygiene instructions. Patients were evaluated for signs of DS at the beginning of the study, at the end of treatment and 30 days later. Microbiological samples were obtained at the beginning of the study and at the end of treatment to quantify Candida albicans cells. Experimental treatment reduced clinical signs and symptoms of DS and the count of C. albicans. The clinical effects of experimental treatment were still evident after 30 days, suggesting that administration of probiotic strain Streptococcus salivarius K12 could be a promising approach in the treatment of DS.

For the assessment of glucose tolerance from IVGTT data in Zucker rat, minimal model methodology is reliable but time- and money-consuming. This study aimed to validate for the first time in Zucker rat, simple surrogate indexes of insulin sensitivity and secretion against the glucose-minimal-model insulin sensitivity index (SI) and against first- (Φ1) and second-phase (Φ2) β-cell responsiveness indexes provided by C-peptide minimal model. Validation of the surrogate insulin sensitivity index (ISI) and of two sets of coupled insulin-based indexes for insulin secretion, differing from the cut-off point between phases (FPIR3-SPIR3, t = 3 min and FPIR5-SPIR5, t = 5 min), was carried out in a population of ten Zucker fatty rats (ZFR) and ten Zucker lean rats (ZLR). Considering the whole rat population (ZLR+ZFR), ISI showed a significant strong correlation with SI (Spearman's correlation coefficient, r = 0.88; P

Hemiplegia is a neurological disorder that is often detected in children with cerebral palsy. Although many studies have investigated muscular activity in hemiplegic legs, few EMG-based findings focused on unaffected limb. This study aimed to quantify the asymmetric behavior of lower-limb-muscle recruitment during walking in mild-hemiplegic children from surface-EMG and foot-floor contact features. sEMG signals from tibialis anterior (TA) and gastrocnemius lateralis and foot-floor contact data during walking were analyzed in 16 hemiplegic children classified as W1 according to Winter’ scale, and in 100 control children. Statistical gait analysis, a methodology achieving a statistical characterization of gait by averaging surface-EMG-based features, was performed. Results, achieved in hundreds of strides for each child, indicated that in the hemiplegic side with respect to the non-hemiplegic side, W1 children showed a statistically significant: decreased number of strides with normal foot-floor contact; decreased stance-phase length and initial-contact sub-phase; curtailed, less frequent TA activity in terminal swing and a lack of TA activity at heel-strike. The acknowledged impairment of anti-phase eccentric control of dorsiflexors was confirmed in the hemiplegic side, but not in the contralateral side. However, a modified foot-floor contact pattern is evinced also in the contralateral side, probably to make up for balance requirements.

A C-peptide-based assessment of β-cell function was performed here in the Zucker fatty rat, a suitable animal model of human metabolic syndrome. To this aim, a 90-min intravenous glucose tolerance test (IVGTT) was performed in seven Zucker fatty rats (ZFR), 7-to-9 week-old, and seven age-matched Zucker lean rats (ZLR). The minimal model of C-peptide (CPMM), originally introduced for humans, was adapted to Zucker rats and then applied to interpret IVGTT data. For a comprehensive evaluation of glucose tolerance in ZFR, CPMM was applied in combination with the minimal model of glucose kinetics (GKMM). Our results showed that the present CPMM-based interpretation of data is able to: 1) provide a suitable fit of C-Peptide data; 2) achieve a satisfactory estimation of parameters of interest 3) quantify both insulin secretion by estimating the time course of pre-hepatic secretion rate, SR(t), and total insulin secretion, TIS, and pancreatic sensitivity by means of three specific indexes of β-cell responsiveness to glucose stimulus (first-phase, Ф(1), second-phase, Ф(2), and steady-state, Ф(ss), never assessed in Zucker rats before; 4) detect the significant enhancement of insulin secretion in the ZFR, in face of a severe insulin-resistant state, previously observed only using a purely experimental approach. Thus, the methodology presented here represents a reliable tool to assess β-cell function in the Zucker rat, and opens new possibilities for the quantification of further processes involved in glucose homeostasis such as the hepatic insulin degradation.