Speech Perception Improvement Algorithm Based on a Dual-Path Long Short-Term Memory Network.

Current deep learning-based speech enhancement methods focus on enhancing the time-frequency representation of the signal. However, conventional methods can lead to speech damage due to resolution mismatch problems that emphasize only specific information in the time or frequency domain. To address these challenges, this paper introduces a speech enhancement model designed with a dual-path structure that identifies key speech characteristics in both the time and time-frequency domains. Specifically, the time path aims to model semantic features hidden in the waveform, while the time-frequency path attempts to compensate for the spectral details via a spectral extension block. These two paths enhance temporal and spectral features via mask functions modeled as LSTM, respectively, offering a comprehensive approach to speech enhancement. Experimental results show that the proposed dual-path LSTM network consistently outperforms conventional single-domain speech enhancement methods in terms of speech quality and intelligibility.

AI Model for Detection of Abdominal Hemorrhage Lesions in Abdominal CT Images.

Information technology has been actively utilized in the field of imaging diagnosis using artificial intelligence (AI), which provides benefits to human health. Readings of abdominal hemorrhage lesions using AI can be utilized in situations where lesions cannot be read due to emergencies or the absence of specialists; however, there is a lack of related research due to the difficulty in collecting and acquiring images. In this study, we processed the abdominal computed tomography (CT) database provided by multiple hospitals for utilization in deep learning and detected abdominal hemorrhage lesions in real time using an AI model designed in a cascade structure using deep learning, a subfield of AI. The AI model was used a detection model to detect lesions distributed in various sizes with high accuracy, and a classification model that could screen out images without lesions was placed before the detection model to solve the problem of increasing false positives owing to the input of images without lesions in actual clinical cases. The developed method achieved 93.22% sensitivity and 99.60% specificity.

Distinct volumetric features of cerebrospinal fluid distribution in idiopathic normal-pressure hydrocephalus and Alzheimer's disease.

OBJECTIVE: The aims of the study were to measure the cerebrospinal fluid (CSF) volumes in the lateral ventricle, high-convexity subarachnoid space, and Sylvian fissure region in patients with idiopathic normal-pressure hydrocephalus (INPH) and Alzheimer's disease (AD), and to evaluate differences in these volumes between INPH and AD groups and healthy controls. METHODS: Forty-nine INPH patients, 59 AD patients, and 26 healthy controls were imaged with automated three-dimensional volumetric MRI. RESULTS: INPH patients had larger lateral ventricles and CSF spaces of the Sylvian fissure region and smaller high-convexity subarachnoid spaces than other groups, and AD patients had larger lateral ventricles and CSF spaces of the Sylvian fissure region than the control group. The INPH group showed a negative correlation between lateral ventricle and high-convexity subarachnoid space volumes, while the AD group showed a positive correlation between lateral ventricle volume and volume for CSF spaces of the Sylvian fissure region. The ratio of lateral ventricle to high-convexity subarachnoid space volumes yielded an area under the curve of 0.990, differentiating INPH from AD. CONCLUSIONS: Associations between CSF volumes suggest that there might be different mechanisms between INPH and AD to explain their respective lateral ventricular dilations. The ratio of lateral ventricle to high-convexity subarachnoid space volumes distinguishes INPH from AD with good diagnostic sensitivity and specificity. We propose to refer to this ratio as the VOSS (ventricle over subarachnoid space) index.

A New Type of Wireless Transmission Based on Digital Direct Modulation for Use in Partially Implantable Hearing Aids.

In this study, we developed a new type of wireless transmission system for use in partially implantable hearing aids. This system was designed for miniaturization and low distortion, and features direct digital modulation. The sigma-delta output, which has a high SNR due to oversampling and noise shaping technology, is used as the data signal and is transmitted using a wireless transmission system to the implant unit through OOK without restoration as an audio signal, thus eliminating the need for additional circuits (i.e., LPF and a reference voltage supply circuit) and improving the ease of implantation and reliability of the circuit. We selected a carrier frequency of 27 MHz after analysis of carrier attenuation by human tissue, and designed the communication coil with reference to both the geometry and required communication distance. Circuit design and simulation for wireless transmission were performed using Multisim 13.0. The system was fabricated based on the circuit design; the size of the device board was 13 mm × 13 mm, the size of the implanted part was 9 mm × 9 mm, the diameter of the transmitting/receiving coil was 26 mm, and the thicknesses of these coils were 0.5 and 0.3 mm, respectively. The difference (error) between the detected and simulation waveforms was about 5%, and was thought to be due to the tolerances of the fabricated communication coil and elements (resistors, capacitors, etc.) used in the circuit configuration of the system. The number of windings was reduced more than 9-fold compared to the communication coil described by Taghavi et al. The measured THD was <1% in the frequency band from 100 Hz to 10 kHz, thus easily meeting the standard specification for hearing aids.

A method for enhancing speech and warning signals based on parallel convolutional neural networks in a noisy environment.

BACKGROUND: Digital hearing aids are based on technology that amplifies sound and removes noise according to the frequency of hearing loss in hearing loss patients. However, within the noise removed is a warning sound that alert the listener; the listener may be exposed to danger because the warning sound is not recognized. OBJECTIVE: In this paper, a deep learning model was used to improve these limits and propose a method to distinguish the warning sound in speech signals mixed with noise. In addition, the improved speech and warning sound were derived by removing noise present in the classification sound signals. METHODS: To classify the sound dataset, an adaptive convolution filter that changes according to two signals is proposed. The proposed convolution filter is applied to the PCNNs model to analyze the characteristics of the time and frequency domains of the dataset and classify the presence or absence of warning sound. In addition, the CEDN model was used to improve the intelligibility of the warning and the speech in the signal based on the warning sound classification from the proposed PCNNs model. RESULTS: Experimental results show that the PCNNs model using the proposed multiplicative filters is efficient for analyzing sound signals with complex frequencies. In addition, the CEDN model was used to improve the intelligibility of the warning and the speech in the signal based on the warning sound classification from the proposed PCNNs model. CONVLUSION: We confirmed that the PCNN model with the proposed filter showed the highest training rate, lowest error rate, and the most stable results. In addition, the CEDN model confirmed that speech and warning sounds were recognized, but it was confirmed that there was a limitation in clearly recognizing speech as the noise ratio increased.

Implementation of a 3D position detection system for a medical simulator.

BACKGOROUND AND OBJECTIVE: Cardiovascular disorders are increasing because of poor eating habits, excessive drinking, and lack of exercise. Some of the typical cardiovascular surgical procedures utilize catheters. Catheter-based procedures require the surgeons to have extensive experience and high proficiency at performing vascular interventions. However, the learning period to acquire such proficiency is lengthy and the opportunities for practical training and mastery are insufficient. Therefore, due to insufficient skill, dangerous situations with damage or rupture of the patient's blood vessels may occur, thereby increasing the risk of medical accidents. Hence, it is necessary to have experience and proficiency for performing vascular interventions. Thus, it is necessary to develop a simulator to shorten learning time and reduce medical accidents. METHODS: In this study, we developed a position detection system for the simulator to use physical models to learn cardiovascular surgical intervention techniques. The developed system uses changes in the output values of a Hall sensor based on the position of a permanent magnet. RESULTS AND CONCLUSIONS: From the changing output values, the distance calculation equation is derived, and the position of the permanent magnet is effectively estimated from the calculations. The performance of the position detecting system was tested, and the results proved that the system could be sufficiently used as a simulator.

Implementation of a fully implantable middle-ear hearing device chip.

BACKGROUND AND OBJECTIVE: Recently, with the increase in the population of hearing impaired people, various types of hearing aids have been rapidly developed. In particular, a fully implantable middle ear hearing device (F-IMEHD) is developed for people with sensorineural hearing loss. The F-IMEHD system comprises an implantable microphone, a transducer, and a signal processor. The signal processor should have a small size and consume less power for implantation in a human body. METHODS: In this study, we designed and fabricated a signal-processing chip using the modified FFT algorithm. This algorithm was developed focusing on eliminating time delay and system complexity in the transform process. The designed signal-processing chip comprises a 4-channel WDRC, a fitting memory, a communication 1control part, and a pulse density modulator. Each channel is separated using a 64-point fast Fourier transform (FFT) method and the gain value is matched using the fitting table in the fitting memory. RESULTS AND CONCLUSION: The chip was designed by Verilog-HDL and the designed HDL codes were verified by Modelsim-PE 10.3 (Mentor graphics, USA). The chip was fabricated using a 0.18 µm CMOS process (SMIC, China). Experiments were performed on a cadaver to verify the performance of the fabricated chip.

Noise reduction algorithm with the soft thresholding based on the Shannon entropy and bone-conduction speech cross- correlation bands.

BACKGROUND: The conventional methods of speech enhancement, noise reduction, and voice activity detection are based on the suppression of noise or non-speech components of the target air-conduction signals. However, air-conduced speech is hard to differentiate from babble or white noise signals. OBJECTIVE: To overcome this problem, the proposed algorithm uses the bone-conduction speech signals and soft thresholding based on the Shannon entropy principle and cross-correlation of air- and bone-conduction signals. METHODS: A new algorithm for speech detection and noise reduction is proposed, which makes use of the Shannon entropy principle and cross-correlation with the bone-conduction speech signals to threshold the wavelet packet coefficients of the noisy speech. RESULTS: The proposed method can be get efficient result by objective quality measure that are PESQ, RMSE, Correlation, SNR. CONCLUSION: Each threshold is generated by the entropy and cross-correlation approaches in the decomposed bands using the wavelet packet decomposition. As a result, the noise is reduced by the proposed method using the MATLAB simulation. To verify the method feasibility, we compared the air- and bone-conduction speech signals and their spectra by the proposed method. As a result, high performance of the proposed method is confirmed, which makes it quite instrumental to future applications in communication devices, noisy environment, construction, and military operations.

Implementation of automatic external defibrillator using real time ventricular fibrillation detecting algorithm based on time domain analysis.

The increase in mortality associated with arrhythmia is an inevitable problem of modern society such as westernized eating habits and an increase in stress due to industrialization, and the related social costs are increasing. In this regard, the supply of automatic external defibrillator (AED) used outside hospitals is increasing mainly in public institutions, and AED is a medical practice performed by non-medical personnel. Therefore, studies on arrhythmia detection algorithm to make accurate clinical judgment for proper use are increasing. In this paper, we propose a time domain analysis method to detect arrhythmia in real time and implement AED by porting it to programmable gate array and digital signal processor. The analysis of the phase domain improves the detection rate of R-peak using the differentiated electrocardiogram (ECG) waveform rather than the existing ECG waveform and makes it easy to distinguish the normal ECG from the arrhythmia signal in the phase domain. The proposed algorithm was verified by simulation using Labview and ModelSim, and it was verified that the proposed algorithm works effectively by performing animal experiments using the implemented AED.

Mastication noise reduction method for fully implantable hearing aid using piezo-electric sensor.

BACKGROUND: Fully implantable hearing devices (FIHDs) can be affected by generated biomechanical noise such as mastication noise. OBJECTIVE: To reduce the mastication noise using a piezo-electric sensor, the mastication noise is measured with the piezo-electric sensor, and noise reduction is practiced by the energy difference. METHODS: For the experiment on mastication noise, a skull model was designed using artificial skull model and a piezo-electric sensor that can measure the vibration signals better than other sensors. A 1 kHz pure-tone sound through a standard speaker was applied to the model while the lower jawbone of the model was moved in a masticatory fashion. RESULTS: The correlation coefficients and signal-to-noise ratio (SNR) before and after application of the proposed method were compared. It was found that the signal-to-noise ratio and correlation coefficients increased by 4.48 dB and 0.45, respectively. CONCLUSION: The mastication noise is measured by piezo-electric sensor as the mastication noise that occurred during vibration. In addition, the noise was reduced by using the proposed method in conjunction with MATLAB. In order to confirm the performance of the proposed method, the correlation coefficients and signal-to-noise ratio before and after signal processing were calculated. In the future, an implantable microphone for real-time processing will be developed.

Difference of auditory brainstem responses by stimulating to round and oval window in animal experiments.

ABSTACT To ensure the safety and efficacy of implantable hearing aids, animal experiments are an essential developmental procedure, in particular, auditory brainstem responses (ABRs) can be used to verify the objective effectiveness of implantable hearing aids. This study measured and compared the ABRs generated when applying the same vibration stimuli to an oval window and round window. The ABRs were measured using a TDT system 3 (TDT, USA), while the vibration stimuli were applied to a round window and oval window in 4 guinea pigs using a piezo-electric transducer with a proper contact tip. A paired t-test was used to determine any differences between the ABR amplitudes when applying the stimulation to an oval window and round window. The paired t-test revealed a significant difference between the ABR amplitudes generated by the round and oval window stimulation (t = 10.079, α < .0001). Therefore, the results confirmed that the biological response to round window stimulation was not the same as that to oval window stimulation.

Wireless charing pillow for a fully implantable hearing aid: Design of a circular array coil based on finite element analysis for reducing magnetic weak zones.

Many types of fully implantable hearing aids have been developed. Most of these devices are implanted behind the ear. To maintain the implanted device for a long period of time, a rechargeable battery and wireless power transmission are used. Because inductive coupling is the most renowned method for wireless power transmission, many types of fully implantable hearing aids are transcutaneously powered using inductively coupled coils. Some patients with an implantable hearing aid require a method for conveniently charging their hearing aid while they are resting or sleeping. To address this need, a wireless charging pillow has been developed that employs a circular array coil as one of its primary parts. In this device, all primary coils are simultaneously driven to maintain an effective charging area regardless of head motion. In this case, however, there may be a magnetic weak zone that cannot be charged at the specific secondary coil's location on the array coil. In this study, assuming that a maximum charging distance is 4 cm, a circular array coil-serving as a primary part of the charging pillow-was designed using finite element analysis. Based on experimental results, the proposed device can charge an implantable hearing aid without a magnetic weak zone within 4 cm of the perpendicular distance between the primary and secondary coils.

Voice activity detection algorithm using perceptual wavelet entropy neighbor slope.

This paper presents a voice activity detection (VAD) approach using a perceptual wavelet entropy neighbor slope (PWENS) in a low signal-to-noise (SNR) environment and with a variety of noise types. The basis for our study is to use acoustic features that have large entropy variance for each wavelet critical band. The speech signal is decomposed by the proposed perceptual wavelet packet decomposition (PWPD), and the VAD function is extracted by PWENS. Finally, VAD is decided by the proposed VAD decision rule using two memory buffers. In order to evaluate the performance of the VAD decision, many speech samples and a variety of SNR conditions were used in the experiment. The performance of the VAD decision is confirmed using objective indexes such as a graph of the VAD decision and the relative error rate.

Individual tooth region segmentation using modified watershed algorithm with morphological characteristic.

In this paper, a new method for individual tooth segmentation was proposed. The proposed method is composed of enhancement and extraction of boundary and seed of watershed algorithm using trisection areas by morphological characteristic of teeth. The watershed algorithm is one of the conventional methods for tooth segmentation; however, the method has some problems. First, molar region detection ratio is reduced because of oral structure features that is low intensities in molar region. Second, inaccurate segmentation occurs in incisor region owing to specular reflection. To solve the problems, the trisection method using morphological characteristic was proposed, where three tooth areas are made using ratio of entire tooth to each tooth. Moreover, the enhancement is to improve the intensity of molar using the proposed method. In addition, boundary and seed of watershed are extracted using trisection areas applied other parameters each area. Finally, individual tooth segmentation was performed using extracted boundary and seed. Furthermore, the proposed method was compared with conventional methods to confirm its efficiency. As a result, the proposed method was demonstrated to have higher detection ratio, better over segmentation, and overlap segmentation than conventional methods.

Speech quality evaluation of subcutaneously implanted microphone using in vivo experiment.

The microphone in a fully implantable hearing device (FIHD) is generally implanted under the skin covering the temporal bone. However, the implanted microphone can be affected by the skin, which causes both sound attenuation and distortion, particularly at high frequencies. As the degree of attenuation and distortion through the skin is severe, speech quality evaluation parameters are needed for the received signal when designing an implantable microphone. However, the performance of most implantable microphones is only assessed based on the sensitivity and frequency response. Thus, practical indicators based on human auditory characteristics are needed for an objective evaluation of the performance of implantable microphones. In this study, a subcutaneously implantable microphone was designed, and its frequency response investigated using an in vivo experiment. Plus, to evaluate the objective indicators, the speech quality of the signals measured by the implanted microphone was calculated using a MATLAB program, and the indicators compared before and after implantation.

Measurement of stapes vibration in Human temporal bones by round window stimulation using a 3-coil transducer.

Round window placement of a 3-coil transducer offers a new approach for coupling an implantable hearing aid to the inner ear. The transducer exhibits high performance at low-frequencies. One remarkable feature of the 3-coil transducer is that it minimizes leakage flux. Thus, the transducer, which consists of two permanent magnets and three coils, can enhance vibrational displacement. In human temporal bones, stapes vibration was observed by laser Doppler vibrometer in response to round window stimulation using the 3-coil transducer. Coupling between the 3-coil transducer and the round window was connected by a wire-rod. The stimulation created stapes velocity when the round window stimulated. Performance evaluation was conducted by measuring stapes velocity. To verify the performance of the 3-coil transducer, stapes velocity for round window and tympanic membrane stimulation were compared, respectively. Stapes velocity by round window stimulation using the 3-coil transducer was approximately 14 dB higher than that achieved by tympanic membrane stimulation. The study shows that 3-coil transducer is suitable for implantable hearing aids.

In vivo evaluation of mastication noise reduction for dual channel implantable microphone.

Input for fully implantable hearing devices (FIHDs) is provided by an implantable microphone under the skin of the temporal bone. However, the implanted microphone can be affected when the FIHDs user chews. In this paper, a dual implantable microphone was designed that can filter out the noise from mastication. For the in vivo experiment, a fabricated microphone was implanted in a rabbit. Pure-tone sounds of 1 kHz through a standard speaker were applied to the rabbit, which was given food simultaneously. To evaluate noise reduction, the measured signals were processed using a MATLAB program based adaptive filter. To verify the proposed method, the correlation coefficients and signal to-noise ratio before and after signal processing were calculated. By comparing the results, signal-to-noise ratio and correlation coefficients are enhanced by 6.07dB and 0.529 respectively.

Alterations in cortical activity of male methamphetamine abusers performing an empathy task: fMRI study.

OBJECTIVES: We investigate possible differences in neural correlates of empathy processing between abstinent methamphetamine (MA) abusers and healthy subjects using functional magnetic resonance imaging (fMRI). METHODS: Nineteen abstinent MA abusers (mean age of 36.06 years, range 31-52 years) and 19 healthy subjects (mean age of 37.05 years, range 33-42 years) participated in this study. A visual fMRI activation paradigm was used, comprising a series of cartoons, each depicting a short story. There were two categories of stories: empathy (Empathy) and Physical causality (Physical). fMRI images were acquired using a 3.0 T whole-body scanner. All fMRI data were analyzed using MATLAB v. 7.2 and SPM5. RESULTS: Both MA subjects and controls exhibited activation in the dorsomedial prefrontal cortex. Despite this similarity in activation patterns, we found that the two groups differed in the activation of several cortical regions associated with the processing of empathy information. Hypoactivations of the orbitofrontal cortex, temporal poles, and hippocampus in MA abusers relative to healthy subjects suggests that the ability of empathic response could be compromised in abstinent MA abusers (p < 0.05, corrected for a small volume). CONCLUSIONS: Functional impairments in the empathic neural network caused by MA may contribute to the misunderstanding of others and to the erosion of social interactions in MA abusers.

Reduced corpus callosum white matter microstructural integrity revealed by diffusion tensor eigenvalues in abstinent methamphetamine addicts.

The purpose of the current study was to determine whether abstinent methamphetamine (MA) abusers demonstrate differences in white matter (WM) integrity of the corpus callosum (CC) due to possible neurotoxic effects of long-term MA abuse, compared with control subjects. In addition to fractional anisotropy (FA), the eigenvalues of the diffusion ellipsoid were used to evaluate the microstructural source of abnormal change in abstinent MA abusers if there occurred a difference in white matter integrity of the CC between healthy controls and abstinent MA abusers. Results showed significantly reduced FA in the genu of the corpus callosum in MA-dependent subjects compared with controls. Furthermore, the eigenvalues offered a unique opportunity to assess the microstructural source of abnormal changes in the genu of the CC. The relationships between Wisconsin Card Sorting Test (WCST) performance and the values of tensor measures also suggest that altered myelination is a possible source of FA reduction observed in the genu of the CC in MA abusers.

Manganese-enhanced auditory tract-tracing MRI with cochlear injection.

Recent studies have demonstrated the use of manganese ion (Mn2+)) as an in vivo neuronal tract tracer. In contrast to histological approaches, manganese tracing can be performed repeatedly on the same living animal. In this study, we describe the neuroaxonal tracing of the auditory pathway in the living guinea pig, relying on the fact that Mn2+ ion enters excitable cells through voltage-gated calcium channels and is an excellent MRI paramagnetic tract-tracing agent. Small focal injections of Mn2+ ion into the cochlea produced significant contrast enhancement along the known neuronal circuitry. This in vivo approach, allowing repeated measures, is expected to open new vistas to study auditory physiology and to provide new insights on in vivo axonal transport and neuronal activity in the central auditory system.