Multiscale Convolutional Neural Network Architecture Research Articles

Multiscale 1-DCNN for Damage Localization and Quantification Using Guided Waves With Novel Data Fusion Technique and New Self-Attention Module

This article proposes an innovative damage localization and size quantification method named as MSCNNSAM based on new multiscale convolutional neural network (MSCNN) and novel small-weighted zero-setting self-attentive module (SAM) in carbon fiber reinforced plastic structures. Firstly, an improved piecewise aggregate approximation algorithm (IPAA) is developed to compress the guided wave signal and extract a series of damage indexes (DI). Considering the different effects of the damage location on the different sensing paths, a new method of damage information targeting enhancement and multipath data fusion is proposed. Then, a novel MSCNN architecture is also proposed for the inherent multiscale characteristics of the guided wave signal, which takes the multipath fused data as input and uses regression and classification methods to directly predict the location and size of the damage. Finally, to further improve the performance of the MSCNN, a SAM is proposed to effectively avoid the influence of low-information channel features and improve the damage feature extraction capability of the network. The proposed method is evaluated through experiments on a guided wave testing platform. Experimental results and comprehensive comparison analysis with respect to the state-of-the-art damage localization and quantification methods have demonstrated the superiority of the proposed MSCNNSAM.

A deep learning approach for inpatient length of stay and mortality prediction

PurposeAccurate prediction of the Length of Stay (LoS) and mortality in the Intensive Care Unit (ICU) is crucial for effective hospital management, and it can assist clinicians for real-time demand capacity (RTDC) administration, thereby improving healthcare quality and service levels. MethodsThis paper proposes a novel one-dimensional (1D) multi-scale convolutional neural network architecture, namely 1D-MSNet, to predict inpatients’ LoS and mortality in ICU. First, a 1D multi-scale convolution framework is proposed to enlarge the convolutional receptive fields and enhance the richness of the convolutional features. Following the convolutional layers, an atrous causal spatial pyramid pooling (SPP) module is incorporated into the networks to extract high-level features. The optimized Focal Loss (FL) function is combined with the synthetic minority over-sampling technique (SMOTE) to mitigate the imbalanced-class issue. ResultsOn the MIMIC-IV v1.0 benchmark dataset, the proposed approach achieves the optimum R-Square and RMSE values of 0.57 and 3.61 for the LoS prediction, and the highest test accuracy of 97.73% for the mortality prediction. ConclusionThe proposed approach presents a superior performance in comparison with other state-of-the-art, and it can effectively perform the LoS and mortality prediction tasks.

Superpixel-Based Multiscale CNN Approach Toward Multiclass Object Segmentation From UAV-Captured Aerial Images

Unmanned aerial vehicles (UAVs) are promising remote sensors capable of reforming remote sensing applications. However, for artificial intelligence (AI)-guided tasks, like land cover mapping and ground-object mapping, most deep learning-based architectures fail to extract scale-invariant features, resulting in poor performance accuracy. In this context, the article proposes a superpixel-aided multiscale convolutional neural network (CNN) architecture to avoid misclassification in complex urban aerial images.The proposed framework is a two-tier deep learning-based segmentation architecture. In the first stage, a superpixel-based simple linear iterative cluster (SLIC) algorithm produces superpixel images with crucial contextual information. The second stage comprises a multiscale CNN architecture that uses these information-rich superpixel images to extract scale-invariant features for predicting the object class of each pixel. Two UAV-image-based aerial image datasets: <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">NITRDrone</i> dataset and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">urban drone dataset</i> (UDD) are considered to perform the experimentation. The proposed model outperforms the considered state-of-the-art methods with an intersection of union (IoU) of 76.39% and 86.85% on UDD and NITRDrone datasets, respectively. Experimentally obtained results prove that the proposed architecture performs superior by achieving better performance accuracy in complex and challenging scenarios.

LMNet: Lightweight multi-scale convolutional neural network architecture for COVID-19 detection in IoMT environment

The COVID-19 disease, initially known as SARS-CoV-2, was first reported in early December 2019 and has caused immense damage to humans globally. The most widely used clinical screening method for COVID-19 is Reverse Transcription Polymerase Chain Reaction (RT-PCR). RT-PCR uses respiratory samples for testing, because of which, this manual technique becomes complicated, laborious and time-consuming. Even though it has a low sensitivity, it carries a considerable risk for the testing medical staff. Hence, there is a need for an automated diagnosis system that can provide quick and efficient diagnosis results. This research proposed a multi-scale lightweight CNN (LMNet) architecture for COVID-19 detection. The proposed model is computationally less expensive than previously available models and requires less memory space. The performance of the proposed LMNet model ensemble with DenseNet169 and MobileNetV2 is higher than the other state-of-the-art models. The ensemble model can be integrated at the backend of the smart devices; hence it is useful for the Internet of Medical Things (IoMT) environment.

2D CNN versus 3D CNN for false-positive reduction in lung cancer screening.

Purpose: To clarify whether and to what extent three-dimensional (3D) convolutional neural network (CNN) is superior to 2D CNN when applied to reduce false-positive nodule detections in the scenario of low-dose computed tomography (CT) lung cancer screening. Approach: We established a dataset consisting of 1600 chest CT examinations acquired on different subjects from various sources. There were in total 18,280 candidate nodules in these CT examinations, among which 9185 were nodules and 9095 were not nodules. For each candidate nodule, we extracted a number of cubic subvolumes with a dimension of by rotating the CT examinations randomly for 25 times prior to the extraction of the axis-aligned subvolumes. These subvolumes were split into three groups in a ratio of for training, validation, and independent testing purposes. We developed a multiscale CNN architecture and implemented its 2D and 3D versions to classify pulmonary nodules into two categories, namely true positive and false positive. The performance of the 2D/3D-CNN classification schemes was evaluated using the area under the receiver operating characteristic curves (AUC). The -values and the 95% confidence intervals (CI) were calculated. Results: The AUC for the optimal 2D-CNN model is 0.9307 (95% CI: 0.9285 to 0.9330) with a sensitivity of 92.70% and a specificity of 76.21%. The 3D-CNN model with the best performance had an AUC of 0.9541 (95% CI: 0.9495 to 0.9583) with a sensitivity of 89.98% and a specificity of 87.30%. The developed multiscale CNN architecture had a better performance than the vanilla architecture did. Conclusions: The 3D-CNN model has a better performance in false-positive reduction compared with its 2D counterpart; however, the improvement is relatively limited and demands more computational resources for training purposes.

A Unified Framework for Multilingual Speech Recognition in Air Traffic Control Systems

This work focuses on robust speech recognition in air traffic control (ATC) by designing a novel processing paradigm to integrate multilingual speech recognition into a single framework using three cascaded modules: an acoustic model (AM), a pronunciation model (PM), and a language model (LM). The AM converts ATC speech into phoneme-based text sequences that the PM then translates into a word-based sequence, which is the ultimate goal of this research. The LM corrects both phoneme- and word-based errors in the decoding results. The AM, including the convolutional neural network (CNN) and recurrent neural network (RNN), considers the spatial and temporal dependences of the speech features and is trained by the connectionist temporal classification loss. To cope with radio transmission noise and diversity among speakers, a multiscale CNN architecture is proposed to fit the diverse data distributions and improve the performance. Phoneme-to-word translation is addressed via a proposed machine translation PM with an encoder-decoder architecture. RNN-based LMs are trained to consider the code-switching specificity of the ATC speech by building dependences with common words. We validate the proposed approach using large amounts of real Chinese and English ATC recordings and achieve a 3.95% label error rate on Chinese characters and English words, outperforming other popular approaches. The decoding efficiency is also comparable to that of the end-to-end model, and its generalizability is validated on several open corpora, making it suitable for real-time approaches to further support ATC applications, such as ATC prediction and safety checking.

Divide-and-Conquer Dual-Architecture Convolutional Neural Network for Classification of Hyperspectral Images

Convolutional neural network (CNN) is well-known for its powerful capability on image classification. In hyperspectral images (HSIs), fixed-size spatial window is generally used as the input of CNN for pixel-wise classification. However, single fixed-size spatial architecture hinders the excellent performance of CNN due to the neglect of various land-cover distributions in HSIs. Moreover, insufficient samples in HSIs may cause the overfitting problem. To address these problems, a novel divide-and-conquer dual-architecture CNN (DDCNN) method is proposed for HSI classification. In DDCNN, a novel regional division strategy based on local and non-local decisions is devised to distinguish homogeneous and heterogeneous regions. Then, for homogeneous regions, a multi-scale CNN architecture with larger spatial window inputs is constructed to learn joint spectral-spatial features. For heterogeneous regions, a fine-grained CNN architecture with smaller spatial window inputs is constructed to learn hierarchical spectral features. Moreover, to alleviate the problem of insufficient training samples, unlabeled samples with high confidences are pre-labeled under adaptively spatial constraint. Experimental results on HSIs demonstrate that the proposed method provides encouraging classification performance, especially region uniformity and edge preservation with limited training samples.

Retinal vessel segmentation of color fundus images using multiscale convolutional neural network with an improved cross-entropy loss function

Abstract Retinal vessel analysis of fundus images is an indispensable method for the screening and diagnosis of related diseases. In this paper, we propose a novel retinal vessel segmentation method of the fundus images based on convolutional neural network (CNN) and fully connected conditional random fields (CRFs). The segmentation process is mainly divided into two steps. Firstly, a multiscale CNN architecture with an improved cross-entropy loss function is proposed to produce the probability map from image to image. We construct the multiscale network by combining the feature map of each middle layer to learn more detail information of the retinal vessels. Meanwhile, our proposed cross-entropy loss function ignores the slightest loss of relatively easy samples in order to take more attention to learn the hard examples. Secondly, CRFs is applied to get the final binary segmentation result which makes use of more spatial context information by taking into account the interactions among all of the pixels in the fundus images. The effectiveness of the proposed method has been evaluated on two public datasets, i.g., DRIVE and STARE with comparisons against eleven state-of-the-art approaches including five deep learning based methods. Results show that our method allows for detection of more tiny blood vessels and more precise locating of the edges.