Introduction: Monitoring electromyograms (EMGs) during skull base surgeries is crucial to prevent cranial nerve injuries, which are common complications of skull base surgery. In order to enhance the safety and efficacy of surgical procedures, using machine learning (ML) algorithms to classify EMG signals can improve the recognition of muscle activation patterns.

Methods: This research utilized a public dataset (DOI: 10.17632/7hyptcbkkd.2) to monitor the EMG obtained from five cranial nerves of 11 patients during cerebellopontine angle tumour surgery. Specifically, the EMG data were collected using the Neuromaster G1 MEE-2000 (Nihon Kohden, Inc., Tokyo, Japan) from the V, VII, XI, X, and XII cranial nerves. An ML model was developed using MATLAB 2023b, based on an ensemble of decision trees, to classify EMG signals into 'Injury', 'Artifact', and 'Healthy' categories. The features used include the amplitude of the rectified value, root mean square value, median frequency, total power, and mean normalized frequency. The data were split using holdout with 80% for training and 20% for testing. Synthetic minority oversampling was applied to the training data to balance the classes; 800 maximum splits per tree were configured with limits of 5 observations per leaf and 10 per parent node. The model was trained through 250 learning cycles with pruning enabled to improve generalization. Subsequently, the model was validated using 5-fold cross-validation, ensuring a robust evaluation of its performance.

Results and discussion: The model achieved on the test set an overall accuracy of 81.12%, with 32.49% precision and 81.01% recall for Injury, 70.00% precision and 75.68% recall for Artifact, and 97.54% precision and 82.12% recall for Healthy, with F1-scores of 46.38%, 72.73%, and 89.17%, respectively.

Conclusion: This study demonstrated the potential of ML in EMG for intraoperative monitoring of cranial nerves, suggesting future optimizations and the integration of advanced algorithms to further improve diagnostic accuracy and clinical utility.

Introduction: Surface electromyography (sEMG) is a non-invasive technique applied to the skin that is able to record the muscles’ electrical activity. In sports science, sEMG is used to assess muscle activation, analyse performance, and monitor muscle fatigue. This study aims to examine muscle activation, namely the electrical potentials produced by the contraction of the gluteus maximus, during a back squat in two groups who underwent different training modalities.

Methods: A total of 20 healthy subjects were selected: 9 participants (age: 21.7±1.8 years, 7 males) followed a training based on box squats, while 11 (age: 21.3±0.7 years, 9 males) observed a program based on pin squats. The study was conducted in two sessions, each involving a single repetition of back squat, with a 3-month interval between sessions, during which participants followed their respective training programs. Muscle activation was measured through sEMG (Encephalan Mini AP-10 system). An analysis of signals was performed using Matlab2023b, computing the median frequency, ARV, and sample entropy. RM-ANOVA and post hoc analysis with Bonferroni correction were performed to identify differences between groups and over time.

Results and discussion: The RM-ANOVA assessed statistical differences over time for the ARV (p<0.0001), and the post-hoc analysis showed an increase in the ARV for both the box squat (p<0.0001) and the pin squat (p=0.015) groups. Notably, no significant differences were assessed between the two groups (p=0.808). Concerning the median frequency and the sample entropy, no statistical differences were assessed. These results suggest that both trainings had a positive effect on muscle activation. However, the box squats group showed a higher increase over time with respect to the pin squat group.

Conclusion: This study analyzed the activation of the gluteus maximus muscle during the back squat after two different types of training. The results demonstrated the effectiveness of sEMG in assessing differences between training methods, providing insights to optimize training strategies.

Introduction : Titanium (Ti)-based coatings are widely used to enhance the surface properties of 316L stainless steel (SS) in biomedical applications. Notably, titanium nitride (TiN) and titanium oxynitride (TiON) are considered advanced ceramics due to their exceptional technical properties. This study explores the properties of TiN and TiON coatings deposited via magnetron sputtering, with a particular focus on how substrate bias voltage influences wettability and corrosion behaviour, driven by the surface's physico-chemical traits.

Methods : A pure titanium nitride target was sputtered in a mixed gas atmosphere of argon and oxygen onto various substrates, such as stainless steel, silicon, and glass. The substrate--target distance was set to 30 mm, and the working pressure was maintained at 10^-2 Torr. A negative DC bias voltage (0V and -100V) was applied during a 30-minute deposition period. Wettability was assessed and electrochemical behaviour was evaluated in a physiological simulated solution over 12 hours of immersion in Hanks' solution

Results : Substrate bias voltage did not affect the thickness and grain size of the coatings, except for the roughness in TiN coatings, which decreased with a -100V bias. Wettability tests indicated that TiN coatings had low contact angles and better wettability, whereas TiON coatings showed high contact angles and poorer wettability. The Icorr values for TiN (0V and -100V) were significantly lower than those for 316L steel. TiON coatings demonstrated even greater corrosion resistance, significantly outperforming uncoated steel.

Conclusions : TiN and TiON coatings, applied with and without substrate bias (0V and -100V) using magnetron sputtering, were evaluated for potential biomedical applications. These coatings exhibited distinct surface characteristics, particularly in terms of contact angle, wettability, and corrosion resistance. The use of negative substrate bias voltage significantly improved corrosion properties. Based on the observed data, TiON coatings are highly appropriate for medical implants.

Stress, as defined by psycho-biologists, is a multifaceted response that encompasses both physiological and psychological components. Chronic stress poses a substantial risk to an individual’s well-being, especially for older adults residing in assisted living facilities.

Bio-signal processing at the output combination of biosensors, such as a heart rate sensor, temperature sensor, and GSR (Galvanic Skin Response) sensor, has been shown to indicate the stress level of human beings. The use of machine learning is crucial in detecting the stress level, while the use of Internet of Things (IoT) makes it easier to share the collected data for analysis and decision making. The objective of this work was the design of IoT- and ML-based wearable stress detection devices encompassing biosensors, using bio-signal processing.

The system was evaluated for its performance in terms of finding the stress level by taking a sufficiently large range of samples. Training and testing were conducted on the samples taken from an old age home named ‘SHEOWS’ (Saint Hardyal Educational and Orphans Welfare Society), which is situated at Okhla, New Delhi, India.

Fuzzy logic algorithms were applied to classify stress levels into four distinct categories, 'Relax,' 'Calm,' 'Anxious,' or 'Stressed’, based on the collected sensor data. Machine learning techniques were employed for stress prediction using the collected sensor data and stress level labels were obtained from the fuzzy logic classification. Among the various machine learning algorithms evaluated, the Random Forest model demonstrated superior accuracy compared to other models, achieving an accuracy of 95.06% in detecting the level of stress. The available device needs to be translated into an industrial physical form so that it can be used as an aesthetic wearable device by users, collecting data continuously and transmitting the stress level to the doctor’s dashboard.

Introduction: Recent studies show that the combination of biomaterials and 3D bioprinting is a promising approach for treating extensive bone injuries. The aim of this study was to develop a bone bioink containing nanosilicate, which enhances the biomaterial's mechanical and biological properties, and pluronic p123 due to its hydrophilic potential, which make this a good candidate for drug delivery. Methodology: Nanosilicate was prepared with tetraethyl orthosilicate (TEOS) and the hydrogel was composed of alginate 5%, pluronic p123 20%, and nanosilicate 2%; it was characterized using rheological tests, scanning electron microscopy (SEM), and degradation and swelling tests. The hydrogel was then mixed with 106 mesenchymal cells/mL and was bioprinted in a 3D bioprinter. In order to evaluate the biocompatibility, a live/dead assay was performed on day one. Results: The nanosilicate showed an average diameter of 392.78 ± 85.08 nm, a zeta potential of -39.65 ± 6.1 mV, and a PDI of 0.105 ± 0.09. The SEM images showed the highly porous structure of the hydrogel with distribution of the nanosilicate throughout the surface of the pores, suggesting an optimal structure for cell adhesion. The hydrogel swelled 1718% over 24 hours, indicating a good capacity for nutrient exchange and cell migration. The degradation rate was 54.21% after one month, suggesting a good clearance of the biomaterial during the regeneration of the natural tissue. Rheological characterization showed a suitable G´/ G´´ ratio, indicating good mechanical properties to biomimic the bone tissue. The live/dead assay revealed a cell viability of around 70% after one day, being non-cytotoxic for the mesenchymal cells. Conclusion: The results of this study showed that the described bioink is a promising material for bone tissue engineering and repair.

Introduction: Both urinary tract infections and primary bloodstream infections result from the use of PDMS catheters and are reported to be the most common nosocomial infections. Despite disinfection procedures used in hospitals, infections still occur, causing delays in discharge and potential mortalities. Bimetallic AgCu nanoparticles (AgCuNPs) are found to exhibit antimicrobial activity against pathogens. Hence, this work reports a manufacturing procedure for fabricating antimicrobial AgCu-PDMS films, which was shown to effectively reduce microbial growth of bacteria and fungi.

Methods: AgCuNP (0.1 % wt/v) was dispersed in Sylgard™ silicone elastomer curing agent and then added to the silicone elastomer base. The mixture was heat-cured and cast, forming 0.1 mm thin films. The fabricated AgCu-PDMS films were treated with a UV₂₅₄ or UV₃₆₅ light source for 15 min. SEM was used to characterise the films, and the population of NP exposured was semi-quantified using ImageJ. Microbial broths (x10⁴) were loaded onto the AgCu-PDMS films, and kinetic growth measurements were conducted to evaluate the antimicrobial properties against E. coli, S. aureus, and C. albicans.

Results and Discussion: Despite the higher frequency exhibited by the UV₂₅₄, more than double of the AgCuNPs were found exposed on the PDMS surface after etching the film using the UV₃₆₅ source. This can be explained by the photothermal property owned by this bimetallic nanomaterial, which shows a board surface plasmonic resonance at λ_max (410 nm) and resulted in excessive polymer degradation due to the excitation of the nanoparticles. Therefore, as expected, the antimicrobial results obtained from the UV₃₆₅-treated AgCu-PDMS films achieved the optimal 72% reduction of microbial growth in relation to the control.

Conclusions: PDMS polymer with incorporated AgCuNPs exhibited antimicrobial properties after UV subjection to expose nanoparticles. Whilst it shows potential to challenge catheter infections, toxicity tests are required to research the suitability of these PDMS films for biomedical applications.

Introduction: Language disorder is the most frequent developmental disorder in childhood, impacting various aspects of language processing. Approximately 11-18% of children aged 18-36 months exhibit language delays, often improving by age 3, but some persist into developmental language disorder (DLD). The early detection of DLD is crucial, as it allows for timely intervention, improving long-term outcomes. This study aims to assess DLD though machine learning (ML) techniques applied to Mel-Frequency Cepstral Coefficients (MFCCs), which are features commonly used in voice analysis.

Materials and methods: This study utilized a freely available dataset (DOI: https://doi.org/10.6084/m9.figshare.2360626.v2) that comprises voice recordings from 54 children (35 boys, 19 girls) aged 6 to 12 years diagnosed with DLD, recorded with MD SONY MZ-N710 (fs = 44.1 kHz, 16-bit resolution), and 44 typically developing children (15 boys, 29 girls) aged 4 to 12 years, recorded with a SONY digital Dictaphone (fs = 16kHz, 16-bit resolution). Specifically, the dataset includes recordings of each letter of the alphabet. For the purposes of this study, MFCCs and their first and second derivatives were extracted from the recordings of the letter "A" to perform the classification task. Moreover, the mean and median frequencies of the power spectrum were computed, aiming to investigate eventual spectral distinctions between the two groups.

Results: The best performance was achieved by employing a set of 15 features selected through the MRMR procedure, which resulted in a test accuracy of 96.7% and an AUC of 0.98, utilizing a Cubic SVM. Moreover, a t-test assessed differences between the two groups concerning the mean frequency (p = 0.024) and the median frequency (p = 0.022).

Conclusion: This study demonstrated the feasibility of employing ML algorithms in diagnosing DLD through the analysis of MFCCs extracted from voice recordings. This approach could significantly enhance long-term outcomes for individuals affected by this disorder.

Introduction: Brain--computer interfaces (BCIs) based on steady-state visual evoked potentials (SSVEPs) are widely used alternative communication modalities due to their high information transfer rate and systematic responses in occipital cortical areas. In particular, single-flicker modalities have been of great interest in recent years. These have mainly been studied using many recording channels in the occipital area. To minimise the number of recording electrodes, we analysed the clustering of SSVEPs under different stimulation conditions (gaze directions) using a single channel.

Methods: Using a publicly available dataset, EEG signals were recorded from 7 subjects exposed to a central flickering visual stimulus (15 Hz) surrounded by four static targets in the cardinal directions. Participants focused their gaze sequentially on these targets while their cortical responses were recorded. We discretized the energy of the SSVEP, analysed it according to its principal components, and quantified the resulting clustering using the Calinski--Harabasz (CH) index.

Results and Discussion: The energy patterns showed specific characteristics that allowed them to be grouped according to the different stimulation conditions. Principal component analysis revealed that the first three components explained, on average, 93% of the total variance of the data. Quantification of separability using the CH index showed that up to four different stimuli were effectively grouped by a single channel, with an average CH index of 296.36 for the best channel, indicating excellent separability between states. Pairwise comparisons showed that the N--S and N--W directions were the most discriminable, with an average CH above 200 in all cases. A between-subjects analysis revealed that this clustering efficiency was mainly located in a specific region of the occipital cortex (O1, O2, and Oz), where CH indices were consistently higher.

Conclusions: This approach demonstrates the feasibility of significantly reducing the number of recording channels in the implementation of a single-flicker SSVEP BCI.

Introduction: Human–Machine Interaction is a research area that has been gaining increasing attention due to the search for new, more natural, and intuitive interfaces based on the use of new technologies which facilitate the interaction of users with smart devices. In this context, there have been attempts to develop gesture-based interfaces. However, a fundamental question still needs to be addressed centred around whether the gestures used are indeed intuitive. To this end, questionnaires answered by volunteers are usually used, but this type of response is quite subjective. The use of neurophysiological indicators may be an alternative to finding an objective and efficient metric of intuitiveness. This study aimed to test the hypothesis that the improved coordination of muscle synergies serves as a reliable indicator of gesture intuitiveness.

Methods: EMG signals from 16 muscles were monitored, and muscle networks were constructed from the extraction of muscle synergies obtained using Non-Negative Matrix Factorization (NMF) and also from Intermuscular Coherence (IMC). The muscle networks obtained using both approaches in four frequency bands were analyzed in their spatial structure and also using metrics (such as Weighted Global Efficiency (WGE) and Effective Average Strength (EAS)). The correlation of these metrics with the Intuitiveness Level (IL) associated with each gesture was then calculated.

Results and Discussion: The networks from muscle synergies show denser connectivity levels than IMC. Notably, WGE values of synergy muscle networks in the Beta and Gamma2 bands, as well as EAS values of IMC muscle networks in the Gamma1 band, positively correlate with IL values.

Conclusions: The results provide substantial evidence supporting a significant correlation between the intuitiveness level and muscle synergies analyzed using both NMF and IMC approaches.

Introduction: Postural stability results from an effective interaction between sensory feedback integration and muscle modulation. However, the propensity for falls increases with age and has been extensively described for neuromuscular disorders. The comprehension of the muscle modulation mechanisms involved in postural control could help for handle balance deficits and fall prevention. In this study, we investigate the muscle contractile dynamics generated by challenging postures and how the balance is affected by the feedback visual privation.

Methods: This exploratory study examined muscle activation during different postural stability conditions. Postural conditions included three items of the Berg Balance Scale and were administrated in increasing difficulty order with both open and close eyes. Electromyographic (EMG) signals were obtained bilaterally from lower and upper leg muscles (tibialis anterior, gastrocnemius medialis, vastus medialis, rectus femoris, biceps femoris, and tensor fasciae latae) and from one trunk muscle. EMG recordings were synchronized with plantar sensors (F-Scan64 system) to extract Center of Force and pressure data. EMG amplitude, spectral analysis and intramuscular coherence were examined in relation to body movement during destabilizing postures.

Results and Discussion: This experimental approach allowed us to characterize muscle dynamic contraction under different postural conditions with and without visual information. We have found that different strategies and specific modulations were required for controlling balance in highly demanding postures. The amplification of muscle activity and coactivation of lower leg muscles were observed during the most challenging posture and without visual information. Also, specific behaviors consisting of intermittent muscle activity along the postural tasks were observed, especially in the muscles of the dominant leg.

Conclusions: In this study, we describe characteristic oscillatory modulations and synergistic activations as motor strategies for maintaining the balance after a demanding postural condition. These results are discussed in relation to the possibility of accurately assess the efficiency of postural motor strategies.