ABSTRACT
Deep Learning Automated Patient-Specific Quality Assurance (PSQA) aims to reduce clinical resource requirements. It is vital to ensure the safety and effectiveness of radiation therapy by predicting the dose difference metric (Gamma passing rate) and its distribution. However, current research overlooks uncertainty quantification in model predictions, limiting their trustworthiness in real clinical environments. This paper proposes a Multi-granularity Uncertainty Quantification (MGUQ) framework. A Bayesian framework that quantifies uncertainties at multiple granularities for multi-task PSQA, specifically Gamma Passing Rate (GPR) prediction and Dose Difference Prediction (DDP), integrates visualization-based interactive components. Using Bayesian theory, we derive a comprehensive multi-granularity loss function that comprises granularity-specific loss and coherence loss components. Additionally, we proposed Multi-granularity Prior Networks, a dual-stream network architecture, to infer the distributions of DDP (modeled as t-distributions) and GPR (modeled as Gaussian distributions) under specific statistical assumptions. Comprehensive evaluations are conducted on a dataset from ''Peeking Union Medical College Hospital'', and results show that our proposed method achieves a minimum MAE loss of 0.864 with a 2%/3 mm criterion and realizes the uncertainty visualization of dose difference. Further, it also achieves 100% Clinical Accuracy (CA) with a workload of 67.2%. Experiments demonstrate that the proposed framework can enhance the trustworthiness of deep learning applications in PSQA.
Subject(s)
Bayes Theorem , Quality Assurance, Health Care , Humans , Uncertainty , Deep LearningABSTRACT
Citrus peel is a commonly used food-medicine material in the production of fast-moving consumer goods (FMCGs). For instance, Ganpu tea is manufactured by combining the peel of Citri Reticulatae 'Chachi' (PCRC) with Pu-erh tea. The alleviated irritation of PCRC through years of aging makes Citri reticulatae Pericarpium a traditional Chinese medicine. Herein, we introduced short-term steaming into the processing of PCRC to favor the quick removal of its irritation while retaining its food-medicine properties. Sensory evaluation and volatile component analysis showed that 60-s steaming reduced irritation of freshly prepared PCRC. Biological evaluations indicated no effects of steaming on the neuroprotective activity of PCRC. The process increased the contents of several bioactive ingredients, including hesperidin, nobiletin, tangeretin, and synephrine. In addition, physical indications of accelerating PCRC aging were observed. Taken together, our findings suggest that short-term steaming may offer a promising new possibility for enhancing the quality of citrus peel.
Subject(s)
Citrus , Drugs, Chinese Herbal , Medicine, Chinese Traditional , Food , TeaABSTRACT
This article describes a power-efficient, high dynamic range (DR) incremental ADC (IADC) for wearable biopotential signals recording, where DC and low-frequency disturbances such as electrode offset, 50/60 Hz interference and motion artifact must be tolerated. To achieve a wide DR, the IADC performs a three-step conversion by combining zoom-SAR and extended counting (EC) on top of a second-order incremental delta-sigma modulator (ΔΣM). The hybrid architecture notably reduces the oversampling ratio (OSR) with respect to conventional incremental ΔΣMs, while using the EC further improves the Signal-to-Noise-and-Distortion Ratio (SNDR) by 7.4 to 25.6 dB. Fabricated in a 0.18-µm CMOS technology, the IADC achieves 107.6-dB DR, 104.9-dB peak SNR, and 99.3-dB peak SNDR at 2 kS/s while dissipating 130 µW from 1.8-V (analog) / 1.2-V (digital) supply. This translates to a highly competitive FoMDR of 176.5 dB. The high-DR IADC reduces the gain of the preceding instrumentation amplifier (IA) such that significant DC and low-frequency disturbances can be tolerated. The advantages of high DR have been demonstrated by wearable Electrocardiography (ECG) and Electroencephalography (EEG) recordings under motion artifact.
Subject(s)
Electrocardiography , Signal Processing, Computer-Assisted , Equipment Design , Amplifiers, Electronic , MotionABSTRACT
Convolutional Neural Networks (CNNs) have achieved remarkable progress in arbitrary artistic style transfer. However, the model size of existing state-of-the-art (SOTA) style transfer algorithms is immense, leading to enormous computational costs and memory demand. It makes real-time and high resolution hard for GPUs with limited memory and limits the application on mobile devices. This paper proposes a novel arbitrary artistic style transfer algorithm, KBStyle, whose model size is only 200 KB. Firstly, we design a style transfer network where the style encoder, content encoder, and corresponding decoder are custom designed to guarantee low computational cost and high shape retention. Besides, the weighted style loss function is presented to improve the performance of style migration. Then, we propose a novel knowledge distillation method (Symmetric Knowledge Distillation, SKD) for encoder-decoder-based style transfer models, which redefines the knowledge and symmetrically compresses the encoder and decoder. With the SKD, the proposed style transfer network is further compressed by 14 times to achieve the KBStyle. Experimental results demonstrate that the proposed SKD method achieves comparable results with other SOTA knowledge distillation algorithms for style transfer. Besides, the proposed KBStyle achieves high-quality stylized images. And the inference time of the KBStyle on an Nvidia TITAN RTX GPU is only 20 ms when the resolutions of the content image and style image are both 2k-resolution ( 2048×1080 ). Moreover, the 200 KB model size of KBStyle is much smaller than the SOTA models and facilitates style transfer on mobile devices.
ABSTRACT
Denoising is one of the most significant procedures in the image processing pipeline. Nowadays, deep-learning-based algorithms have achieved superior denoising quality than traditional algorithms. However, the noise becomes severe in the dark environment, where even the SOTA algorithms fail to achieve satisfactory performance. Besides, the high computational complexity of deep-learning-based denoising algorithms makes them hardware unfriendly and difficult to process high-resolution images in real-time. To address these issues, a novel low-light RAW denoising algorithm Two-Stage-Denoising (TSDN), is proposed in this paper. In TSDN, denoising consists of two procedures: noise removal and image restoration. Firstly, in the noise-removal stage, most noise is removed from the image, and an intermediate image that is easier for the network to recover the clean image is obtained. Then, in the restoration stage, the clean image is restored from the intermediate image. The TSDN is designed to be light-weight for real-time and hardware friendly. However, the tiny network will be insufficient for satisfactory performance if directly trained from scratch. Therefore, we present an Expand-Shrink-Learning (ESL) method to train the TSDN. In the ESL method, firstly, the tiny network is expanded to a larger one with similar architecture but more channels and layers, which enhances the learning ability of the network because of more parameters. Secondly, the larger network is shrunk and restored to the original small network in fine-grained learning procedures, including Channel-Shrink-Learning (CSL) and Layer-Shrink-Learning (LSL). Experimental results demonstrate that the proposed TSDN achieves better performance (PSNR and SSIM) than other SOTA algorithms in the dark environment. Besides, the model size of TSDN is one-eighth of that of the U-Net for denoising (a classical denoising network).
ABSTRACT
Although deep learning-based (DL-based) image processing algorithms have achieved superior performance, they are still difficult to apply on mobile devices (e.g., smartphones and cameras) due to the following reasons: 1) the high memory demand and 2) large model size. To adapt DL-based methods to mobile devices, motivated by the characteristics of image signal processors (ISPs), we propose a novel algorithm named LineDL. In LineDL, the default mode of the whole-image processing is reformulated as a line-by-line mode, eliminating the need to store large amounts of intermediate data for the whole image. An information transmission module (ITM) is designed to extract and convey the interline correlation and integrate the interline features. Furthermore, we develop a model compression method to reduce the model size while maintaining competitive performance; that is, knowledge is redefined, and compression is performed in two directions. We evaluate LineDL on general image processing tasks, including denoising and superresolution. The extensive experimental results demonstrate that LineDL achieves image quality comparable to that of state-of-the-art (SOTA) DL-based algorithms with a much smaller memory demand and competitive model size.
ABSTRACT
This brief presents a tutorial on multifaceted techniques for high efficiency piezoelectric energy harvesting. For the purpose of helping design piezoelectric energy harvesting system according to different application scenarios, we summarize and discuss the recent design trends and challenges. We divide the design focus into the following three categories, namely, (1) AC-DC rectifiers, (2) CP compensation circuits, (3) maximum power point tracking (MPPT) circuits. The features, problems encountered, and suitable systems of various AC-DC rectifier topologies are introduced and compared. The important role of non-linear methods for piezoelectric energy harvesting is illustrated from the perspective of impedance matching. Energy extraction techniques and voltage flipping techniques based on inductors, capacitors, and hybrid structures are analyzed. MPPT techniques with different features and targets are discussed.
ABSTRACT
Convolutional neural network (CNN)-based filters have achieved great success in video coding. However, in most previous works, individual models were needed for each quantization parameter (QP) band, which is impractical due to limited storage resources. To explore this, our work consists of two parts. First, we propose a frequency and spatial QP-adaptive mechanism (FSQAM), which can be directly applied to the (vanilla) convolution to help any CNN filter handle different quantization noise. From the frequency domain, a FQAM that introduces the quantization step (Qstep) into the convolution is proposed. When the quantization noise increases, the ability of the CNN filter to suppress noise improves. Moreover, SQAM is further designed to compensate for the FQAM from the spatial domain. Second, based on FSQAM, a QP-adaptive CNN filter called QA-Filter that can be used under a wide range of QP is proposed. By factorizing the mixed features to high-frequency and low-frequency parts with the pair of pooling and upsampling operations, the QA-Filter and FQAM can promote each other to obtain better performance. Compared to the H.266/VVC baseline, average 5.25% and 3.84% BD-rate reductions for luma are achieved by QA-Filter with default all-intra (AI) and random-access (RA) configurations, respectively. Additionally, an up to 9.16% BD-rate reduction is achieved on the luma of sequence BasketballDrill. Besides, FSQAM achieves measurably better BD-rate performance compared with the previous QP map method.
ABSTRACT
Due to the movement expressiveness and privacy assurance of human skeleton data, 3D skeleton-based action inference is becoming popular in healthcare applications. These scenarios call for more advanced performance in application-specific algorithms and efficient hardware support. Warnings on health emergencies sensitive to response speed require low latency output and action early detection capabilities. Medical monitoring that works in an always-on edge platform needs the system processor to have extreme energy efficiency. Therefore, in this paper, we propose the MC-LSTM, a functional and versatile 3D skeleton-based action detection system, for the above demands. Our system achieves state-of-the-art accuracy on trimmed and untrimmed cases of general-purpose and medical-specific datasets with early-detection features. Further, the MC-LSTM accelerator supports parallel inference on up to 64 input channels. The implementation on Xilinx ZCU104 reaches a throughput of 18 658 Frames-Per-Second (FPS) and an inference latency of 3.5 ms with the batch size of 64. Accordingly, the power consumption is 3.6 W for the whole FPGA+ARM system, which is 37.8x and 10.4x more energy-efficient than the high-end Titan X GPU and i7-9700 CPU, respectively. Meanwhile, our accelerator also keeps a 4 â¼ 5x energy efficiency advantage against the low-power high-performance Firefly-RK3399 board carrying an ARM Cortex-A72+A53 CPU. We further synthesize an 8-bit quantized version on the same hardware, providing a 48.8% increase in energy efficiency under the same throughput.
Subject(s)
Algorithms , Computers , Delivery of Health Care , Humans , MovementABSTRACT
In this work, a memristive spike-based computing in memory (CIM) system with adaptive neuron (MSPAN) is proposed to realize energy-efficient remote arrhythmia detection with high accuracy in edge devices by software and hardware co-design. A multi-layer deep integrative spiking neural network (DiSNN) is first designed with an accuracy of 93.6% in 4-class ECG classification tasks. Then a memristor-based CIM architecture and the corresponding mapping method are proposed to deploy the DiSNN. By evaluation, the overall system achieves an accuracy of over 92.25% on the MIT-BIH dataset while the area is 3.438 mm2 and the power consumption is 0.178 µJ per heartbeat at a clock frequency of 500 MHz. These results reveal that the proposed MSPAN system is promising for arrhythmia detection in edge devices.
ABSTRACT
The JPEG is one of the most widely used lossy image-compression standards, whose compression performance depends largely on a quantization table. In this work, we utilize a Convolutional Neural Network (CNN) to generate an image-adaptive quantization table in a standard-compliant way. We first build an image set containing more than 10,000 images and generate their optimal quantization tables through a classical genetic algorithm, and then propose a method that can efficiently extract and fuse the frequency and spatial domain information of each image to train a regression network to directly generate adaptive quantization tables. In addition, we extract several representative quantization tables from the dataset and train a classification network to indicate the optimal one for each image, which further improves compression performance and computational efficiency. Tests on diverse images show that the proposed method clearly outperforms the state-of-the-art method. Compared with the standard table at the compression rate of 1.0 bpp, the regression and classification network provide average Peak Signal-to-Noise Ratio (PSNR) gains of nearly 1.2 and 1.4 dB. For the experiment under Structural Similarity Index Measurement (SSIM), the improvements are 0.4% and 0.54%, respectively. The proposed method also has competitive computational efficiency, as the regression and classification network only take 15 and 6.25 milliseconds, respectively, to process a 768 W 512 image on a single CPU core at 3.20 GHz.
ABSTRACT
In consideration of the strong oxidizing power of hypochlorite (ClO-), which could cleave CN moiety, a cyclometalated iridium (III) complex (Ir-Ts) modified hydrazide group as the response unit was synthesized to sensitively and selectively detect ClO- under neutral condition. Upon addition of ClO-, a 21-fold emission enhancement at 574 nm was observed and phosphorescent product was formed due to the cleavage of CN moiety. The probe Ir-Ts displayed rapid response (<15 s) and high selectivity toward ClO- with a low detection limit of 86 nM. More importantly, bioimaging of ClO- was further studied in living cells.
Subject(s)
Coordination Complexes/chemistry , Hypochlorous Acid/analysis , HeLa Cells , Humans , Luminescent Measurements/methodsABSTRACT
Blockchain technology is increasingly being used in Internet of things (IoT) devices for information security and data integrity. However, it is challenging to implement complex hash algorithms with limited resources in IoT devices owing to large energy consumption and a long processing time. This paper proposes an RISC-V processor with memristor-based in-memory computing (IMC) for blockchain technology in IoT applications. The IMC-adapted instructions were designed for the Keccak hash algorithm by virtue of the extendibility of the RISC-V instruction set architecture (ISA). Then, an RISC-V processor with area-efficient memristor-based IMC was developed based on an open-source core for IoT applications, Hummingbird E200. The general compiling policy with the data allocation method is also disclosed for the IMC implementation of the Keccak hash algorithm. An evaluation shows that >70% improvements in both performance and energy saving were achieved with limited area overhead after introducing IMC in the RISC-V processor.
ABSTRACT
The need for continuous size downscaling of silicon transistors is driving the industrial development of strategies to enable further footprint reduction1,2. The atomic thickness of two-dimensional materials allows the potential realization of high-area-efficiency transistor architectures. However, until now, the design of devices composed of two-dimensional materials has mimicked the basic architecture of silicon circuits3-6. Here, we report a transistor based on a two-dimensional material that can realize photoswitching logic (OR, AND) computing in a single cell. Unlike the conventional transistor working mechanism, the two-dimensional material logic transistor has two surface channels. Furthermore, the material thickness can change the logic behaviour-the architecture can be flexibly expanded to achieve in situ memory such as logic computing and data storage convergence in the same device. These devices are potentially promising candidates for the construction of new chips that can perform computing and storage with high area-efficiency and unique functions.
ABSTRACT
The primary aim of this study was to investigate the impact of the angiotensin I-converting enzyme gene (ACE) on general cognitive ability, specific cognitive ability and psychomotor function in Chinese children. In total, 450 children completed both C-WISC tests and ACE I/D genotyping. Of these, 320 children were examined using psychomotor tests. The quantitative traits of psychometric IQ and psychomotor abilities were calculated to determine whether there were any significant differences related to their ACE genotypes on the basis of an analysis of variance. F test results showed no significant differences with regard to any aspect of intelligence or psychomotor performance relative to the various ACE I/D genotypes (all p > 0.05). Our study suggests that ACE I/D do not have a measurable impact on any aspects of IQ or psychomotor ability and that psychomotor ability correlates well with IQ in Chinese children.