RESUMO
With the successful development in computer vision, building a deep convolutional neural network (CNNs) has been mainstream, considering the character of shared parameters in a convolutional layer. Stacking convolutional layers into a deep structure improves performance, but over-stacking also ramps up the needed resources for GPUs. Seeing another surge of Transformers in computer vision, the issue has aroused severely. A resource-hungry model is hardly implemented for limited hardware or single-customers-based GPU. Therefore, this work focuses on these concerns and proposes an efficient but robust backbone, which equips with channel and spatial direction attentions, so the attentions help to expand receptive fields in shallow convolutional layers and pass the information to every layer. An attention-boosted network based on already efficient CNNs, Universal Pixel Attention Networks (UPANets), is proposed. Through a series of experiments, UPANets fulfil the purposes of learning global information with less needed resources and outshine many existing SOTAs in CIFAR-{10, 100}.
RESUMO
In the era of bathing in big data, it is common to see enormous amounts of data generated daily. As for the medical industry, not only could we collect a large amount of data, but also see each data set with a great number of features. When the number of features is ramping up, a common dilemma is adding computational cost during inferring. To address this concern, the data rotational method by PCA in tree-based methods shows a path. This work tries to enhance this path by proposing an ensemble classification method with an AdaBoost mechanism in random, automatically generating rotation subsets termed Random RotBoost. The random rotation process has replaced the manual pre-defined number of subset features (free pre-defined process). Therefore, with the ensemble of the multiple AdaBoost-based classifier, overfitting problems can be avoided, thus reinforcing the robustness. In our experiments with real-world medical data sets, Random RotBoost reaches better classification performance when compared with existing methods. Thus, with the help from our proposed method, the quality of clinical decisions can potentially be enhanced and supported in medical tasks.
RESUMO
Motion mode (M-mode) echocardiography is essential for measuring cardiac dimension and ejection fraction. However, the current diagnosis is time-consuming and suffers from diagnosis accuracy variance. This work resorts to building an automatic scheme through well-designed and well-trained deep learning to conquer the situation. That is, we proposed RAMEM, an automatic scheme of real-time M-mode echocardiography, which contributes three aspects to address the challenges: 1) provide MEIS, the first dataset of M-mode echocardiograms, to enable consistent results and support developing an automatic scheme; For detecting objects accurately in echocardiograms, it requires big receptive field for covering long-range diastole to systole cycle. However, the limited receptive field in the typical backbone of convolutional neural networks (CNN) and the losing information risk in non-local block (NL) equipped CNN risk the accuracy requirement. Therefore, we 2) propose panel attention embedding with updated UPANets V2, a convolutional backbone network, in a real-time instance segmentation (RIS) scheme for boosting big object detection performance; 3) introduce AMEM, an efficient algorithm of automatic M-mode echocardiography measurement, for automatic diagnosis; The experimental results show that RAMEM surpasses existing RIS schemes (CNNs with NL & Transformers as the backbone) in PASCAL 2012 SBD and human performances in MEIS. The implemented code and dataset are available at https://github.com/hanktseng131415go/RAMEM.