The important role of angiogenesis in cancer development has driven many researchers to investigate the prospects of noninvasive cancer diagnosis based on the technology of contrast-enhanced ultrasound (CEUS) imaging. This paper presents a deep learning framework to detect prostate cancer in the sequential CEUS images. The proposed method uniformly extracts features from both the spatial and the temporal dimensions by performing three-dimensional convolution operations, which captures the dynamic information of the perfusion process encoded in multiple adjacent frames for prostate cancer detection. The deep learning models were trained and validated against expert delineations over the CEUS images recorded using two types of contrast agents, i.e. the anti-PSMA based agent targeted to prostate cancer cells and the non-targeted blank agent. Experiments showed that the deep learning method achieved over 91% specificity and 90% average accuracy over the targeted CEUS images for prostate cancer detection, which was superior ($p<0.05$) than previously reported approaches and implementations.

In this paper, a high-efficiency Doherty power amplifier (DPA) using the complex combining load (CCL) with noninfinity peaking impedance for broadband operation is presented. By applying the CCL, the load impedance of the carrier amplifier at saturation power level can be compensated at different operation frequencies. Considering the influences of the output impedance of the peaking PA on the carrier PA, a detailed analysis of the optimal noninfinity peaking output impedance with the CCL is carried out on the whole operation frequency range. Theoretical analysis of the output power and drain efficiency (DE) reveals that the CCL can offer a new degree of freedom for extending the bandwidth of the DPA. Based on the proposed theory, a modified DPA working over the frequency band of 1.1-2.4 GHz (bandwidth of 74%) is designed and fabricated based on GaN HEMT devices. Experimental results exhibit 6-dB output back-off drain efficiencies of 43.8%-54.9% and saturated drain efficiencies of 55.4%-68% can be obtained. The saturated output power of the DPA is 43.3-45.4 dBm with a gain of 9.5-11.1 dB across the whole operation band. When driven with long-term evolution modulated signals with 6.5-dB peak-to-average power ratio, the DPA exhibits an adjacent channel leakage ratio better than -49 dBc while maintaining a DE of 44% after digital predistortion linearization.

This paper presents a high-precision bandgap voltage reference (BGR) with high-order temperature compensation. The compensation signal is generated by using both strong-inversion MOSFETs and Bipolar Junction transistors (BJTs), which cancels the high-order nonlinear term $T\ln (T)$ in the BJT base-emitter voltage (V _BE ), and thus a low temperature coefficient (TC) over a wide temperature range is achieved. The proposed BGR circuit is fabricated in a 0.18- $\mu \text{m}$ CMOS process with an active area of $0.256 m{m^{2}}$ and a max power consumption of 1.35 mW. A minimum TC of 0.706 ${\mathrm{ppm}}/{}^ \circ C$ from $- 25\,\,{}^ \circ C$ to 125 ${}^ \circ C$ is achieved after an 8-bit resistance trimming. The line sensitivity is 0.0146%/V operating from 3.2 V to 3.7 V. The BGR achieves a power supply rejection (PSR) of −63.4 dB and a noise spectrum density of $0.92 ~\mu \text{V}/\sqrt {Hz} $ at 10 Hz.

This letter presents a novel deep learning algorithm for feature extraction from the hyperspectral images. The proposed method takes advantage of the knowledge that the features of the spatial-spectral data naturally fall into an array of groups with respect to different spectral bands. Aiming to reduce the influence of redundant spectral bands adaptively using unlabeled hyperspectral data, we incorporate the group information in the training algorithm of the deep neural network via a regularized weight-decay process. Experiments over different benchmarks of hyperspectral images show that the proposed method provides competitive solution with the state-of-the-art approaches.

In this article, a recessed source trench silicon carbide (SiC) MOSFET with integrated MOS-channel diode (MCD) is proposed and investigated by TCAD simulations. The MCD features a short channel, and the channel length could be adjusted by varying the recessed depth. Owing to the drain-induced barrier lowering effect, a low potential barrier for electrons to flow through the JFET region to the N $+$ source region is formed, which successfully eliminates the bipolar degradation of the parasitic body p-i-n diode. Besides, the recessed source trench introduces an additional depletion region and homogenizes the distribution of the OFF-state electric field. As a result, a low gate-to-drain capacitance and a high breakdown voltage (BV) are obtained. Simulation results indicate that compared with the SiC trench MOSFET with integrated self-assembled three-level protection Schottky barrier diode, a 78.7% reduction in gate-to-drain capacitance and a 24.4% improvement in BV could be achieved in the proposed SiC MOSFET.

This article presents an analytical and coherent design method for the generalized high-efficiency broadband parallel-circuit (PC) Class-E/F₃ power amplifier (PA). By importing the initial phase shift of the third harmonic and adding series reactive element to the load network, a free design parameter can be defined and the design space of the load network parameters are expanded. The design equations of the idealized waveforms and the optimum values of the load network components are derived in detail, and the performance of the peak drain voltage, maximum operating frequency, and power output capability can be improved. In addition, with the expanding space of load network, the generalized PC Class-E/F₃ structure exhibits broadband capability. A matching method with the real frequency technique (RFT) and the de-embedding technique is presented to design the optimal broadband matching network. To verify the validity of the proposed method, a broadband high-efficiency PA working from 1.7 to 2.6 GHz (fractional bandwidth of 42%) is designed. The experimental results of the fabricated PA present high-efficiency characteristics of 68.5%-81% drain efficiency and 62%-75.8% power-added efficiency with 38.8-40.5 dBm output power over the whole operation bandwidth.