The detection of subsurface objects by near infrared ͑NIR͒ spectroscopy and imaging has usually been done with a large number of source positions and a corresponding large number of detector positions. Significant signals have been obtained with a multitude of sources and detectors, to be exact, 4 multiwavelength light emitting diodes ͑LEDs͒ and 16 nearby detectors photodiode silicon diode detectors. A great simplification is made by a dedicated device in which two out of phase sources and a single detector, used in a differential circuit, enable sensitive detection of the appearance of a functionally induced inhomogeneity, for example, a breast cancer or a brain functional signal. By using two LED NIR sources in antiphase at a wavelength appropriate to blood volume increment for the in detection of breast cancer angiogenesis, it is possible to design and construct a very efficient handheld scanner which will indicate the presence of a subsurface angiogenesis by creating imbalance of the optical patterns of the two 800 nm LED sources. Localization and an estimate of the size of the subsurface object may be obtained by scanning the device serially across the breast, as shown in a dynamic 1 cm 3 model tumor to be valid to a depth of 5 cm.

Progress in wearable techniques makes the long-term daily electrocardiogram (ECG) monitoring possible. However, the long-term wearable ECGs can be significantly contaminated by various noises, which affect the detection and diagnosis of cardiovascular diseases (CVDs). The situation becomes more serious for wearable ECG screening, where the data are huge, and doctors have no way to visually check the signal quality episode-by-episode. Therefore, automatic and accurate noise rejection for the wearable bigdata ECGs is craving. This paper addressed this issue and proposed a noise rejection method for wearable ECGs based on the combination of modified frequency slice wavelet transform (MFSWT) and convolutional neural network (CNN). Wearable ECGs were recorded using the newly developed 12-lead Lenovo smart ECG vest with a sample rate of 500 Hz and a resolution of 16 bit. One thousand 10-s ECG segments were picked up and were manually labeled into three quality types: clinically useful segments with good signal quality (type A), clinically useful segments with poor signal quality (type B), and clinically useless segments (pure noises, type C). Each of the 1,000 10-s ECG segments were transformed into a 2-D timefrequency (T-F) image using the MFSWT, with a pixel size of 200×50. Then, the 2-D grayscale images from MFSWT were fed into a 13-layer CNN model for training the classification models. Results from the standard 5-folder cross-validation showed that the proposed combination method of MFSWT and CNN achieved a highest classification accuracy of 86.3%, which was higher than the comparable methods from continuous wavelet transform (CWT) and artificial neural networks (ANN). The combination of MFSWT and CNN also had a good calculation efficiency. This paper indicated that the combination of MFSWT and CNN is a potential method for automatic identification of noisy segments from wearable ECG recordings. INDEX TERMS Wearable ECG, signal quality assessment (SQA), convolutional neural network (CNN), modified frequency slice wavelet transform (MFSWT).

We simulated the effects of compression of the breast on blood volume and tissue oxygenation. We sought to answer the question: how does the compression during breast examination impact on the circulatory systems of the breast tissue, namely blood flow, blood pooling, and oxygen concentration? We assumed that the blood was distributed in two compartments, arterial and venous. All the parameters were expressed with oxy- and deoxyhemoglobin quantities and were measured with a non-invasive method, Near Infrared Spectroscopy (NIRS). The simulated data showed that the blood volume pool in the breast decreased due to lower arterial flow and higher venous outflow, as the breast was squeezed under 100 cm H2O with a 10 cm diameter probe (or 78 cm2). The blood volume was reversed when the pressure was released. The breast venous oxygen saturation dropped, but overall tissue saturation (presenting NIRS signal, volume weighted average saturation) was increased. The results showed that simulation can be used to obtain venous and average oxygen saturation as well as blood flow in compressed breast tissues.

An analytical model governing the dynamics of the trapped-electron density in electron-trapping materials (ETM's) under simultaneous blue and near-IR illuminations is developed and studied in detail. Experimental results confirming the theoretical findings based on the model are presented, including a new method for experimentally determining a parameter of ETM's, β, which describes the rate of decay of electron-trap density under constant IR illumination and which is obtained by measurement of the phase shift of the ETM response to IR illumination containing a sinusoidally modulated temporal component. Issues concerning the use of ETM's as the synaptic connection weights in an optoelectronic neurocomputer are discussed; in particular, we propose a novel scheme for stabilizing the stored weight information in ETM's during readout and learning; this scheme is based on the dynamic equilibrium of the trapped-electron density established by simultaneous blue and uniform IR illuminations. It is shown thatETM's are deally suited for realizing dense, modifiable synapses that have the wide dynamic range needed in implementing large-scale programmable optoelectronic neural networks of pulsed neurons.

Progress in wearable techniques makes the long-term daily electrocardiogram (ECG) monitoring possible. Premature ventricular contraction (PVC) is one of the most common cardiac arrhythmias. This study proposed a method by combining the modified frequency slice wavelet transform (MFSWT) and convolutional neural network (CNN). Training data are from the 2018 China physiological signal challenge . The first 10-s ECG waveforms in each recording were transformed into 2-D time-frequency images (frequency range of 0-50 Hz and size of 300 × 100) using MFSWT. A 25-layer CNN structure was constructed, which includes five convolution layers with kernel size of 3×3, five dropout layers, five ReLU layers, five maximum pooling layers with kernel size of 2 × 2, a flatten layer, two fully connected layers, as well as the input and output layers. Test data were recorded from Results showed that, the proposed method achieved a high accuracy of 97.89% for PVC/non-PVC episodes classification, indicating that the combination of MFSWT and CNN provides new insight to accurately identify PVC from the wearable ECG recordings.