Improvement in Performance and Stability of PbS QD/IGZO Phototransistors Through the Introduction of Ga2O3 Film for Broadband Sensor Applications.

The development of broadband photosensors has become crucial in various fields. Indium-gallium-zinc oxide (IGZO, In:Ga:Zn = 1:1:1) phototransistors with PbS quantum dots (QDs) have shown promising features for such sensors, such as reasonable mobility, low leakage current, good photosensitivity, and low-cost fabrication. However, the instability of PbS QD/IGZO phototransistors under an air atmosphere and prolonged storage remain serious concerns. In this article, two concepts to improve the reliability of PbS QD/IGZO phototransistors were implemented. P-type doping in the PbS QD layer through oxidation allows increasing the built-in potential between IGZO and PbS QDs, leading to enhancement in photoinduced electron-hole pair creation. Second, agglomeration and fusion of a PbS QDs layer were controlled via thermal annealing, which facilitated the transport of photocreated carriers. The p-type doping and interconnection of a PbS QD layer can be achieved by deposition and subsequent thermal annealing of gallium oxide (Ga2O3) on PbS QD/IGZO stacks. The resulting Ga2O3/PbS QD/IGZO phototransistors exhibited high-performance switching characteristics under dark conditions. Notably, they showed a remarkable photoresponsivity of 196.69 ± 4.05 A/W and a detectivity of (5.47 ± 1.4) × 1012 Jones even at a long-wavelength illumination of 1550 nm. While the unpassivated PbS/IGZO phototransistor suffered serious degradation in optical performance after 2 weeks of storage, the Ga2O3/PbS QD/IGZO phototransistor demonstrated enhanced stability, maintaining high performance for over 5 weeks. These findings suggest that Ga2O3/PbS QD/IGZO phototransistors offer a feasible approach for the fabrication of large-scale active matrix broadband photosensor arrays, potentially revolutionizing optical sensing in various cutting-edge applications.

A Multi-modal Teacher-student Framework for Improved Blood Pressure Estimation.

Blood pressure (BP) is a critical vital sign that hypertensive patients regularly measure. In this study, we propose a novel BP estimation framework to distill the knowledge from a multi-modal model to a uni-modal BP estimation model through teacher-student training. The multi-modal BP estimation model consists of three components: first, a gated recurrent unit network that generates features from photoplethysmogram, electrocardiogram, age, height, and weight; second, an attention mechanism that integrates each feature into joint embeddings; and third, a regression layer that estimates BP from the joint embeddings. The uni-modal BP estimation model has similar components to the multi-modal model but uses only PPG signal. BP is predicted by the embeddings extracted from the uni-modal model, and these embeddings are trained to be as similar as possible to the joint embeddings extracted from the multi-modal model. The proposed method demonstrates absolute prediction errors of 5.24±6.41 and 3.49±3.85 mmHg for systolic blood pressure and diastolic blood pressure in the MIMIC-III dataset, satisfying the AAMI standard.

Deep-learning-based blood pressure estimation using multi channel photoplethysmogram and finger pressure with attention mechanism.

Recently, several studies have proposed methods for measuring cuffless blood pressure (BP) using finger photoplethysmogram (PPG) signals. This study presents a new BP estimation system that measures PPG signals under progressive finger pressure, making the system relatively robust to errors caused by finger position when using the cuffless oscillometric method. To reduce errors caused by finger position, we developed a sensor that can simultaneously measure multi-channel PPG and force signals in a wide field of view (FOV). We propose a deep-learning-based algorithm that can learn to focus on the optimal PPG channel from multi channel PPG using an attention mechanism. The errors (ME ± STD) of the proposed multi channel system were 0.43±9.35 mmHg and 0.21 ± 7.72 mmHg for SBP and DBP, respectively. Through extensive experiments, we found a significant performance difference depending on the location of the PPG measurement in the BP estimation system using finger pressure.

High-Performance Broadband Phototransistor Based on TeOx/IGTO Heterojunctions.

Ultraviolet to infrared broadband spectral detection capability is a technological challenge for sensing materials being developed for high-performance photodetection. In this work, we stacked 9 nm-thick tellurium oxide (TeOx) and 8 nm-thick InGaSnO (IGTO) into a heterostructure at a low temperature of 150 °C. The superior photoelectric characteristics we achieved benefit from the intrinsic optical absorption range (300-1500 nm) of the hexagonal tellurium (Te) phase in the TeOx film, and photoinduced electrons are driven effectively by band alignment at the TeOx/IGTO interface under illumination. A photosensor based on our optimized heterostructure exhibited a remarkable detectivity of 1.6 × 1013 Jones, a responsivity of 84 A/W, and a photosensitivity of 1 × 105, along with an external quantum efficiency of 222% upon illumination by blue light (450 nm). Simultaneously, modest detection properties (responsivity: ∼31 A/W, detectivity: ∼6 × 1011 Jones) for infrared irradiation at 970 nm demonstrate that this heterostructure can be employed as a broadband phototransistor. Furthermore, its low-temperature processability suggests that our proposed concept might be used to design array optoelectronic devices for wide band detection with high sensitivity, flexibility, and stability.

Attention-Based Joint Training of Noise Suppression and Sound Event Detection for Noise-Robust Classification.

Sound event detection (SED) recognizes the corresponding sound event of an incoming signal and estimates its temporal boundary. Although SED has been recently developed and used in various fields, achieving noise-robust SED in a real environment is typically challenging owing to the performance degradation due to ambient noise. In this paper, we propose combining a pretrained time-domain speech-separation-based noise suppression network (NS) and a pretrained classification network to improve the SED performance in real noisy environments. We use group communication with a context codec method (GC3)-equipped temporal convolutional network (TCN) for the noise suppression model and a convolutional recurrent neural network for the SED model. The former significantly reduce the model complexity while maintaining the same TCN module and performance as a fully convolutional time-domain audio separation network (Conv-TasNet). We also do not update the weights of some layers (i.e., freeze) in the joint fine-tuning process and add an attention module in the SED model to further improve the performance and prevent overfitting. We evaluate our proposed method using both simulation and real recorded datasets. The experimental results show that our method improves the classification performance in a noisy environment under various signal-to-noise-ratio conditions.

Luminance-Degradation Compensation Based on Multistream Self-Attention to Address Thin-Film Transistor-Organic Light Emitting Diode Burn-In.

We propose a deep-learning algorithm that directly compensates for luminance degradation because of the deterioration of organic light-emitting diode (OLED) devices to address the burn-in phenomenon of OLED displays. Conventional compensation circuits are encumbered by high cost of the development and manufacturing processes because of their complexity. However, given that deep-learning algorithms are typically mounted onto systems on chip (SoC), the complexity of the circuit design is reduced, and the circuit can be reused by only relearning the changed characteristics of the new pixel device. The proposed approach comprises deep-feature generation and multistream self-attention, which decipher the importance of the variables, and the correlation between burn-in-related variables. It also utilizes a deep neural network that identifies the nonlinear relationship between extracted features and luminance degradation. Thereafter, luminance degradation is estimated from burn-in-related variables, and the burn-in phenomenon can be addressed by compensating for luminance degradation. Experiment results revealed that compensation was successfully achieved within an error range of 4.56%, and demonstrated the potential of a new approach that could mitigate the burn-in phenomenon by directly compensating for pixel-level luminance deviation.

Multi-TALK: Multi-Microphone Cross-Tower Network for Jointly Suppressing Acoustic Echo and Background Noise.

In this paper, we propose a multi-channel cross-tower with attention mechanisms in latent domain network (Multi-TALK) that suppresses both the acoustic echo and background noise. The proposed approach consists of the cross-tower network, a parallel encoder with an auxiliary encoder, and a decoder. For the multi-channel processing, a parallel encoder is used to extract latent features of each microphone, and the latent features including the spatial information are compressed by a 1D convolution operation. In addition, the latent features of the far-end are extracted by the auxiliary encoder, and they are effectively provided to the cross-tower network by using the attention mechanism. The cross tower network iteratively estimates the latent features of acoustic echo and background noise in each tower. To improve the performance at each iteration, the outputs of each tower are transmitted as the input for the next iteration of the neighboring tower. Before passing through the decoder, to estimate the near-end speech, attention mechanisms are further applied to remove the estimated acoustic echo and background noise from the compressed mixture to prevent speech distortion by over-suppression. Compared to the conventional algorithms, the proposed algorithm effectively suppresses the acoustic echo and background noise and significantly lowers the speech distortion.

Joint Optimization of Deep Neural Network-Based Dereverberation and Beamforming for Sound Event Detection in Multi-Channel Environments.

In this paper, we propose joint optimization of deep neural network (DNN)-supported dereverberation and beamforming for the convolutional recurrent neural network (CRNN)-based sound event detection (SED) in multi-channel environments. First, the short-time Fourier transform (STFT) coefficients are calculated from multi-channel audio signals under the noisy and reverberant environments, which are then enhanced by the DNN-supported weighted prediction error (WPE) dereverberation with the estimated masks. Next, the STFT coefficients of the dereverberated multi-channel audio signals are conveyed to the DNN-supported minimum variance distortionless response (MVDR) beamformer in which DNN-supported MVDR beamforming is carried out with the source and noise masks estimated by the DNN. As a result, the single-channel enhanced STFT coefficients are shown at the output and tossed to the CRNN-based SED system, and then, the three modules are jointly trained by the single loss function designed for SED. Furthermore, to ease the difficulty of training a deep learning model for SED caused by the imbalance in the amount of data for each class, the focal loss is used as a loss function. Experimental results show that joint training of DNN-supported dereverberation and beamforming with the SED model under the supervision of focal loss significantly improves the performance under the noisy and reverberant environments.

Recognition, classification, and prediction of the tactile sense.

The emulation of the tactile sense is presented with the encoding of a complex surface texture through an electrical sensor device. To achieve a functional capability comparable to a human mechanoreceptor, a tactile sensor is designed by employing a naturally formed porous structure of a graphene film. The inherent tactile patterns are achievable by means of proper analysis of the electrical signals that the sensor provides during the event of touching the interacting objects. It is confirmed that the pattern-recognition method using machine learning is suitable for quantifying human tactile sensations. The classification accuracy of the tactile sensor system is better than that of human touch for the tested fabric samples, which have a delicate surface texture.

Deep learning ensemble with asymptotic techniques for oscillometric blood pressure estimation.

BACKGROUND AND OBJECTIVE: This paper proposes a deep learning based ensemble regression estimator with asymptotic techniques, and offers a method that can decrease uncertainty for oscillometric blood pressure (BP) measurements using the bootstrap and Monte-Carlo approach. While the former is used to estimate SBP and DBP, the latter attempts to determine confidence intervals (CIs) for SBP and DBP based on oscillometric BP measurements. METHOD: This work originally employs deep belief networks (DBN)-deep neural networks (DNN) to effectively estimate BPs based on oscillometric measurements. However, there are some inherent problems with these methods. First, it is not easy to determine the best DBN-DNN estimator, and worthy information might be omitted when selecting one DBN-DNN estimator and discarding the others. Additionally, our input feature vectors, obtained from only five measurements per subject, represent a very small sample size; this is a critical weakness when using the DBN-DNN technique and can cause overfitting or underfitting, depending on the structure of the algorithm. To address these problems, an ensemble with an asymptotic approach (based on combining the bootstrap with the DBN-DNN technique) is utilized to generate the pseudo features needed to estimate the SBP and DBP. In the first stage, the bootstrap-aggregation technique is used to create ensemble parameters. Afterward, the AdaBoost approach is employed for the second-stage SBP and DBP estimation. We then use the bootstrap and Monte-Carlo techniques in order to determine the CIs based on the target BP estimated using the DBN-DNN ensemble regression estimator with the asymptotic technique in the third stage. RESULTS: The proposed method can mitigate the estimation uncertainty such as large the standard deviation of error (SDE) on comparing the proposed DBN-DNN ensemble regression estimator with the DBN-DNN single regression estimator, we identify that the SDEs of the SBP and DBP are reduced by 0.58 and 0.57  mmHg, respectively. These indicate that the proposed method actually enhances the performance by 9.18% and 10.88% compared with the DBN-DNN single estimator. CONCLUSION: The proposed methodology improves the accuracy of BP estimation and reduces the uncertainty for BP estimation.

Noncontact Sleep Study by Multi-Modal Sensor Fusion.

Polysomnography (PSG) is considered as the gold standard for determining sleep stages, but due to the obtrusiveness of its sensor attachments, sleep stage classification algorithms using noninvasive sensors have been developed throughout the years. However, the previous studies have not yet been proven reliable. In addition, most of the products are designed for healthy customers rather than for patients with sleep disorder. We present a novel approach to classify sleep stages via low cost and noncontact multi-modal sensor fusion, which extracts sleep-related vital signals from radar signals and a sound-based context-awareness technique. This work is uniquely designed based on the PSG data of sleep disorder patients, which were received and certified by professionals at Hanyang University Hospital. The proposed algorithm further incorporates medical/statistical knowledge to determine personal-adjusted thresholds and devise post-processing. The efficiency of the proposed algorithm is highlighted by contrasting sleep stage classification performance between single sensor and sensor-fusion algorithms. To validate the possibility of commercializing this work, the classification results of this algorithm were compared with the commercialized sleep monitoring device, ResMed S+. The proposed algorithm was investigated with random patients following PSG examination, and results show a promising novel approach for determining sleep stages in a low cost and unobtrusive manner.

Oscillometric blood pressure estimation by combining nonparametric bootstrap with Gaussian mixture model.

BACKGROUND: Blood pressure (BP) is one of the most important vital indicators and plays a key role in determining the cardiovascular activity of patients. METHODS: This paper proposes a hybrid approach consisting of nonparametric bootstrap (NPB) and machine learning techniques to obtain the characteristic ratios (CR) used in the blood pressure estimation algorithm to improve the accuracy of systolic blood pressure (SBP) and diastolic blood pressure (DBP) estimates and obtain confidence intervals (CI). The NPB technique is used to circumvent the requirement for large sample set for obtaining the CI. A mixture of Gaussian densities is assumed for the CRs and Gaussian mixture model (GMM) is chosen to estimate the SBP and DBP ratios. The K-means clustering technique is used to obtain the mixture order of the Gaussian densities. RESULTS: The proposed approach achieves grade "A" under British Society of Hypertension testing protocol and is superior to the conventional approach based on maximum amplitude algorithm (MAA) that uses fixed CR ratios. The proposed approach also yields a lower mean error (ME) and the standard deviation of the error (SDE) in the estimates when compared to the conventional MAA method. In addition, CIs obtained through the proposed hybrid approach are also narrower with a lower SDE. CONCLUSIONS: The proposed approach combining the NPB technique with the GMM provides a methodology to derive individualized characteristic ratio. The results exhibit that the proposed approach enhances the accuracy of SBP and DBP estimation and provides narrower confidence intervals for the estimates.

Estimated confidence interval from single blood pressure measurement based on algorithmic fusion.

BACKGROUND: Current oscillometric blood pressure measurement devices generally provide only single-point estimates for systolic and diastolic blood pressures and rarely provide confidence ranges for these estimates. A novel methodology to obtain confidence intervals (CIs) for systolic blood pressure (SBP) and diastolic blood pressure (DBP) estimates from a single oscillometric blood pressure measurement is presented. METHODS: The proposed methodology utilizes the multiple regression technique to fuse optimally a set of SBP and DBP estimates obtained through different algorithms. However, the set of SBP and DBP estimates is a small number to determine the CI of each individual subject. To address this issue, the weighted bootstrap approach based on the multiple regression technique was used to generate a pseudo sample set for the SBP and the DBP. In this paper, the multiple regression technique can estimate the best-fitting surface of an efficient function that relates the input sample set as an independent vector to the auscultatory nurse measurement as a dependent vector to estimate regression coefficients. Consequently, the coefficients are assigned to an eight-sample set obtained from the fusion of different algorithms as optimally weighted parameters. CIs are also estimated using the conventional methods on the set of fused SBP and DBP estimates for comparison purposes. RESULTS: The proposed method was applied to an experimental dataset of 85 patients. The results indicated that the proposed approach provides better blood pressure estimates than the existing algorithms and, in addition, is able to provide CIs for a single measurement. CONCLUSIONS: The CIs derived from the proposed scheme are much smaller than those calculated by conventional methods except for the pseudo maximum amplitude-envelope algorithm for both the SBP and the DBP, probably because of the decrease in the standard deviation through the increase in the pseudo measurements using the weighted bootstrap method for each subject. The proposed methodology is likely the only one currently available that can provide CIs for single-sample blood pressure measurements.

Enhanced voice activity detection in kernel subspace domain.

This paper proposes a voice activity detection (VAD) method in a kernel subspace domain to improve the performance of the kernel-based VAD. A linear transform matrix that can simultaneously diagonalize the two covariance matrices using kernel principal component analysis is presented to generate the kernel subspace. The likelihood ratio test based on Gaussian distributions is applied for the VAD in the kernel subspace. Experimental results show that the proposed VAD algorithm outperforms the conventional approaches under various noise conditions.

Statistical voice activity detection in kernel space.

This paper proposes a statistical voice activity detection method in a high-dimensional kernel feature space by a nonlinear mapping. A Gaussian density model is presented using kernel principal component analysis to represent the nonlinear characteristics of the speech signal. The proposed approach offers a decision rule based on a multiple observation likelihood ratio test in the kernel space.

A subspace approach based on embedded prewhitening for voice activity detection.

This paper presents a subspace approach for voice activity detection (VAD). The proposed approach is based on an embedded prewhitening scheme for the simultaneous diagonalization of the clean speech and noise covariance matrices to provide a decision rule based on likelihood ratio test in signal subspace domain. Experimental results show that the proposed subspace-based VAD algorithm outperforms the method using a Gaussian model in a conventional discrete Fourier transform domain at the low signal-to-noise conditions.