-
An Integrated Visual Analytics System for Studying Clinical Carotid Artery Plaques
Authors:
Chaoqing Xu,
Zhentao Zheng,
Yiting Fu,
Baofeng Chang,
Legao Chen,
Minghui Wu,
Mingli Song,
**song Jiang
Abstract:
Carotid artery plaques can cause arterial vascular diseases such as stroke and myocardial infarction, posing a severe threat to human life. However, the current clinical examination mainly relies on a direct assessment by physicians of patients' clinical indicators and medical images, lacking an integrated visualization tool for analyzing the influencing factors and composition of carotid artery p…
▽ More
Carotid artery plaques can cause arterial vascular diseases such as stroke and myocardial infarction, posing a severe threat to human life. However, the current clinical examination mainly relies on a direct assessment by physicians of patients' clinical indicators and medical images, lacking an integrated visualization tool for analyzing the influencing factors and composition of carotid artery plaques. We have designed an intelligent carotid artery plaque visual analysis system for vascular surgery experts to comprehensively analyze the clinical physiological and imaging indicators of carotid artery diseases. The system mainly includes two functions: First, it displays the correlation between carotid artery plaque and various factors through a series of information visualization methods and integrates the analysis of patient physiological indicator data. Second, it enhances the interface guidance analysis of the inherent correlation between the components of carotid artery plaque through machine learning and displays the spatial distribution of the plaque on medical images. Additionally, we conducted two case studies on carotid artery plaques using real data obtained from a hospital, and the results indicate that our designed carotid analysis system can effectively provide clinical diagnosis and treatment guidance for vascular surgeons.
△ Less
Submitted 8 August, 2023;
originally announced August 2023.
-
Surface-guided computing to analyze subcellular morphology and membrane-associated signals in 3D
Authors:
Felix Y. Zhou,
Andrew Weems,
Gabriel M. Gihana,
Bingying Chen,
Bo-Jui Chang,
Meghan Driscoll,
Gaudenz Danuser
Abstract:
Signal transduction and cell function are governed by the spatiotemporal organization of membrane-associated molecules. Despite significant advances in visualizing molecular distributions by 3D light microscopy, cell biologists still have limited quantitative understanding of the processes implicated in the regulation of molecular signals at the whole cell scale. In particular, complex and transie…
▽ More
Signal transduction and cell function are governed by the spatiotemporal organization of membrane-associated molecules. Despite significant advances in visualizing molecular distributions by 3D light microscopy, cell biologists still have limited quantitative understanding of the processes implicated in the regulation of molecular signals at the whole cell scale. In particular, complex and transient cell surface morphologies challenge the complete sampling of cell geometry, membrane-associated molecular concentration and activity and the computing of meaningful parameters such as the cofluctuation between morphology and signals. Here, we introduce u-Unwrap3D, a framework to remap arbitrarily complex 3D cell surfaces and membrane-associated signals into equivalent lower dimensional representations. The map**s are bidirectional, allowing the application of image processing operations in the data representation best suited for the task and to subsequently present the results in any of the other representations, including the original 3D cell surface. Leveraging this surface-guided computing paradigm, we track segmented surface motifs in 2D to quantify the recruitment of Septin polymers by blebbing events; we quantify actin enrichment in peripheral ruffles; and we measure the speed of ruffle movement along topographically complex cell surfaces. Thus, u-Unwrap3D provides access to spatiotemporal analyses of cell biological parameters on unconstrained 3D surface geometries and signals.
△ Less
Submitted 12 April, 2023;
originally announced April 2023.
-
A Two-Dimensional FFT Precoded Filter Bank Scheme
Authors:
R. Pereira Junior,
C. A. F. da Rocha,
B. S. Chang,
D. Le Ruyet
Abstract:
This work proposes a new precoded filter bank (FB) system via a two-dimensional (2D) fast Fourier transform (2D-FFT). Its structure is similar to Orthogonal Time Frequency Space (OTFS) systems, where the OFDM transmitter is changed to a filter bank multi-carrier (FBMC) one, thus obtaining a lower out-of-band emission. The complex orthogonality of the FBMC transmission is guaranteed by using precod…
▽ More
This work proposes a new precoded filter bank (FB) system via a two-dimensional (2D) fast Fourier transform (2D-FFT). Its structure is similar to Orthogonal Time Frequency Space (OTFS) systems, where the OFDM transmitter is changed to a filter bank multi-carrier (FBMC) one, thus obtaining a lower out-of-band emission. The complex orthogonality of the FBMC transmission is guaranteed by using precoding based on a discrete Fourier transform, which is also used to implement the two-dimensional fast Fourier transform. Through the use of a global transmission matrix, we propose a hybrid receiver for the new system. First, a frequency domain equalization is performed, followed by an interference cancellation on the delay-Doppler domain. The simulation results show that the proposed system obtains an error performance similar to other OTFS systems, and superior performance as compared to other precoded FBMC systems.
△ Less
Submitted 11 June, 2022;
originally announced June 2022.
-
Temporal Knowledge Distillation for On-device Audio Classification
Authors:
Kwanghee Choi,
Martin Kersner,
Jacob Morton,
Buru Chang
Abstract:
Improving the performance of on-device audio classification models remains a challenge given the computational limits of the mobile environment. Many studies leverage knowledge distillation to boost predictive performance by transferring the knowledge from large models to on-device models. However, most lack a mechanism to distill the essence of the temporal information, which is crucial to audio…
▽ More
Improving the performance of on-device audio classification models remains a challenge given the computational limits of the mobile environment. Many studies leverage knowledge distillation to boost predictive performance by transferring the knowledge from large models to on-device models. However, most lack a mechanism to distill the essence of the temporal information, which is crucial to audio classification tasks, or similar architecture is often required. In this paper, we propose a new knowledge distillation method designed to incorporate the temporal knowledge embedded in attention weights of large transformer-based models into on-device models. Our distillation method is applicable to various types of architectures, including the non-attention-based architectures such as CNNs or RNNs, while retaining the original network architecture during inference. Through extensive experiments on both an audio event detection dataset and a noisy keyword spotting dataset, we show that our proposed method improves the predictive performance across diverse on-device architectures.
△ Less
Submitted 5 February, 2022; v1 submitted 26 October, 2021;
originally announced October 2021.
-
Multiple Access for Transmissions Over Independent Fading Channels
Authors:
Bei-Hao Chang,
Chia-Fu Chang,
Pin-Wen Su,
I-Hsien Yeh,
Kai-Chuan Cheng,
Ying-Chen Lin,
Mao-Chao Lin
Abstract:
We propose to employ a multilevel detection (MLDT) technique to allow multiple users which respectively transmit messages over independent fading channels to share the same resource, e.g., the same signature sequence in the CDMA (code division multiple access) system. The users are separated by the different channel gains including amplitudes and phases resultant from the independent fading channe…
▽ More
We propose to employ a multilevel detection (MLDT) technique to allow multiple users which respectively transmit messages over independent fading channels to share the same resource, e.g., the same signature sequence in the CDMA (code division multiple access) system. The users are separated by the different channel gains including amplitudes and phases resultant from the independent fading channels. In a CDMA system with a fixed amount of available signature sequences, the number of users can be doubled or tripled by using MLDT although there is the cost of some BER performance degradation.
△ Less
Submitted 28 July, 2018;
originally announced July 2018.
