Skeleton-Based Attention Mask for Pedestrian Attribute Recognition Network

This paper presents an extended model for a pedestrian attribute recognition network utilizing skeleton data as a soft attention model to extract a local feature corresponding to a specific attribute. This technique helped keep valuable information surrounding the target area and handle the variation of human posture. The attention masks were designed to focus on the partial and the whole-body regions. This research utilized an augmented layer for data augmentation inside the network to reduce over-fitting errors. Our network was evaluated in two datasets (RAP and PETA) with various backbone networks (ResNet-50, Inception V3, and Inception-ResNet V2). The experimental result shows that our network improves overall classification performance with a mean accuracy of about 2–3% in the same backbone network, especially local attributes and various human postures.

Download Full-text

Deep Fish

SLAS DISCOVERY Advancing Life Sciences ◽

10.1177/1087057116667894 ◽

2016 ◽

Vol 22 (1) ◽

pp. 102-107 ◽

Cited By ~ 6

Author(s):

Omer Ishaq ◽

Sajith Kecheril Sadanandan ◽

Carolina Wählby

Keyword(s):

Deep Learning ◽

Data Augmentation ◽

Expert Knowledge ◽

Model Organism ◽

Image Data ◽

Classification Performance ◽

Whole Body ◽

Head Region ◽

Data Set ◽

Vertebrate Model

Zebrafish ( Danio rerio) is an important vertebrate model organism in biomedical research, especially suitable for morphological screening due to its transparent body during early development. Deep learning has emerged as a dominant paradigm for data analysis and found a number of applications in computer vision and image analysis. Here we demonstrate the potential of a deep learning approach for accurate high-throughput classification of whole-body zebrafish deformations in multifish microwell plates. Deep learning uses the raw image data as an input, without the need of expert knowledge for feature design or optimization of the segmentation parameters. We trained the deep learning classifier on as few as 84 images (before data augmentation) and achieved a classification accuracy of 92.8% on an unseen test data set that is comparable to the previous state of the art (95%) based on user-specified segmentation and deformation metrics. Ablation studies by digitally removing whole fish or parts of the fish from the images revealed that the classifier learned discriminative features from the image foreground, and we observed that the deformations of the head region, rather than the visually apparent bent tail, were more important for good classification performance.

Download Full-text

Radioimmunszintigraphie mit SPECT: Methodik, Probleme und klinische Erfahrungen

Nuklearmedizin ◽

10.1055/s-0038-1628889 ◽

1987 ◽

Vol 26 (05) ◽

pp. 202-205 ◽

Cited By ~ 1

Author(s):

J. Fass ◽

S. Truong ◽

U. Büll ◽

V. Schumpelick ◽

R. Bares

Keyword(s):

Clinical Practice ◽

The Other ◽

Whole Body ◽

Blood Clearance ◽

Planar Scintigraphy ◽

Gastrointestinal Carcinomas ◽

Body Regions ◽

Other Hand ◽

Body Retention ◽

Klinische Erfahrungen

Radioimmunoscintigraphy (RIS) with 111ln- and 131 I-labelled monoclonal anti bodies (MAbs) against CEA and/or CA 19-9 was performed in 83 patients with various gastrointestinal carcinomas. A total of 276 body regions could be examined. The results of planar scintigraphy and SPECT were compared intraindividually. Using 111 In-labelled MAbs the sensitivity of RIS was significantly improved by SPECT (88.9 vs. 52.4% with planar scintigraphy, p <0.01). For131 l-labelled MAbs the effect was smaller (83.9 vs. 65.6% with planar scintigraphy, n.s.). This finding can be explained by different kinetics and biodistribution of the used MAb preparations.111 In-labelled MAbs with long whole-body retention and rapid blood clearance reveal ideal qualities for SPECT; on the other hand, the short whole-body retention of131 l-labelled MAbs leads to small count rates and therefore long counting times that make delayed SPECT unsuitable in clinical practice

Download Full-text

Kinetic analysis and optimisation of 18F-rhPSMA-7.3 PET imaging of prostate cancer

European Journal of Nuclear Medicine and Molecular Imaging ◽

10.1007/s00259-021-05346-8 ◽

2021 ◽

Author(s):

Simona Malaspina ◽

Vesa Oikonen ◽

Anna Kuisma ◽

Otto Ettala ◽

Kalle Mattila ◽

...

Keyword(s):

Prostate Cancer ◽

Metastatic Disease ◽

Uptake Kinetics ◽

Whole Body ◽

Target Area ◽

Post Injection ◽

The Mean ◽

Pet Radiopharmaceutical ◽

Time Frames ◽

Over Time

Abstract Purpose This phase 1 open-label study evaluated the uptake kinetics of a novel theranostic PET radiopharmaceutical, 18F-rhPSMA-7.3, to optimise its use for imaging of prostate cancer. Methods Nine men, three with high-risk localised prostate cancer, three with treatment-naïve hormone-sensitive metastatic disease and three with castration-resistant metastatic disease, underwent dynamic 45-min PET scanning of a target area immediately post-injection of 300 MBq 18F-rhPSMA-7.3, followed by two whole-body PET/CT scans acquired from 60 and 90 min post-injection. Volumes of interest (VoIs) corresponding to prostate cancer lesions and reference tissues were recorded. Standardised uptake values (SUV) and lesion-to-reference ratios were calculated for 3 time frames: 35–45, 60–88 and 90–118 min. Net influx rates (Ki) were calculated using Patlak plots. Results Altogether, 44 lesions from the target area were identified. Optimal visual lesion detection started 60 min post-injection. The 18F-rhPSMA-7.3 signal from prostate cancer lesions increased over time, while reference tissue signals remained stable or decreased. The mean (SD) SUV (g/mL) at the 3 time frames were 8.4 (5.6), 10.1 (7) and 10.6 (7.5), respectively, for prostate lesions, 11.2 (4.3), 13 (4.8) and 14 (5.2) for lymph node metastases, and 4.6 (2.6), 5.7 (3.1) and 6.4 (3.5) for bone metastases. The mean (SD) lesion-to-reference ratio increases from the earliest to the 2 later time frames were 40% (10) and 59% (9), respectively, for the prostate, 65% (27) and 125% (47) for metastatic lymph nodes and 25% (19) and 32% (30) for bone lesions. Patlak plots from lesion VoIs signified almost irreversible uptake kinetics. Ki, SUV and lesion-to-reference ratio estimates showed good agreement. Conclusion 18F-rhPSMA-7.3 uptake in prostate cancer lesions was high. Lesion-to-background ratios increased over time, with optimal visual detection starting from 60 min post-injection. Thus, 18F-rhPSMA-7.3 emerges as a very promising PET radiopharmaceutical for diagnostic imaging of prostate cancer. Trial Registration NCT03995888 (24 June 2019).

Download Full-text

Classification of Space Objects by Using Deep Learning with Micro-Doppler Signature Images

Sensors ◽

10.3390/s21134365 ◽

2021 ◽

Vol 21 (13) ◽

pp. 4365

Author(s):

Kwangyong Jung ◽

Jae-In Lee ◽

Nammoon Kim ◽

Sunjin Oh ◽

Dong-Wook Seo

Keyword(s):

Transfer Learning ◽

Data Augmentation ◽

Doppler Frequency ◽

Extraction Methods ◽

Classification Performance ◽

Reconstruction Method ◽

Generative Adversarial Network ◽

Adversarial Network ◽

Space Objects ◽

Target Characteristics

Radar target classification is an important task in the missile defense system. State-of-the-art studies using micro-doppler frequency have been conducted to classify the space object targets. However, existing studies rely highly on feature extraction methods. Therefore, the generalization performance of the classifier is limited and there is room for improvement. Recently, to improve the classification performance, the popular approaches are to build a convolutional neural network (CNN) architecture with the help of transfer learning and use the generative adversarial network (GAN) to increase the training datasets. However, these methods still have drawbacks. First, they use only one feature to train the network. Therefore, the existing methods cannot guarantee that the classifier learns more robust target characteristics. Second, it is difficult to obtain large amounts of data that accurately mimic real-world target features by performing data augmentation via GAN instead of simulation. To mitigate the above problem, we propose a transfer learning-based parallel network with the spectrogram and the cadence velocity diagram (CVD) as the inputs. In addition, we obtain an EM simulation-based dataset. The radar-received signal is simulated according to a variety of dynamics using the concept of shooting and bouncing rays with relative aspect angles rather than the scattering center reconstruction method. Our proposed model is evaluated on our generated dataset. The proposed method achieved about 0.01 to 0.39% higher accuracy than the pre-trained networks with a single input feature.

Download Full-text

Data Augmentation and Spectral Structure Features for Limited Samples Hyperspectral Classification

Remote Sensing ◽

10.3390/rs13040547 ◽

2021 ◽

Vol 13 (4) ◽

pp. 547

Author(s):

Wenning Wang ◽

Xuebin Liu ◽

Xuanqin Mou

Keyword(s):

Classification Accuracy ◽

Data Augmentation ◽

Classification Problem ◽

Classification Performance ◽

Spectral Structure ◽

Limited Sample ◽

Sample Classification ◽

Training Samples ◽

Traditional Classification ◽

Hyperspectral Classification

For both traditional classification and current popular deep learning methods, the limited sample classification problem is very challenging, and the lack of samples is an important factor affecting the classification performance. Our work includes two aspects. First, the unsupervised data augmentation for all hyperspectral samples not only improves the classification accuracy greatly with the newly added training samples, but also further improves the classification accuracy of the classifier by optimizing the augmented test samples. Second, an effective spectral structure extraction method is designed, and the effective spectral structure features have a better classification accuracy than the true spectral features.

Download Full-text

Combining a convolutional neural network with autoencoders to predict the survival chance of COVID-19 patients

Scientific Reports ◽

10.1038/s41598-021-93543-8 ◽

2021 ◽

Vol 11 (1) ◽

Author(s):

Fahime Khozeimeh ◽

Danial Sharifrazi ◽

Navid Hoseini Izadi ◽

Javad Hassannataj Joloudari ◽

Afshin Shoeibi ◽

...

Keyword(s):

Clinical Data ◽

Data Augmentation ◽

Clinical Information ◽

Ct Images ◽

Classification Performance ◽

Survival Chance ◽

Average Accuracy ◽

Novel Method ◽

Aided Diagnosis ◽

Augmentation Procedure

AbstractCOVID-19 has caused many deaths worldwide. The automation of the diagnosis of this virus is highly desired. Convolutional neural networks (CNNs) have shown outstanding classification performance on image datasets. To date, it appears that COVID computer-aided diagnosis systems based on CNNs and clinical information have not yet been analysed or explored. We propose a novel method, named the CNN-AE, to predict the survival chance of COVID-19 patients using a CNN trained with clinical information. Notably, the required resources to prepare CT images are expensive and limited compared to those required to collect clinical data, such as blood pressure, liver disease, etc. We evaluated our method using a publicly available clinical dataset that we collected. The dataset properties were carefully analysed to extract important features and compute the correlations of features. A data augmentation procedure based on autoencoders (AEs) was proposed to balance the dataset. The experimental results revealed that the average accuracy of the CNN-AE (96.05%) was higher than that of the CNN (92.49%). To demonstrate the generality of our augmentation method, we trained some existing mortality risk prediction methods on our dataset (with and without data augmentation) and compared their performances. We also evaluated our method using another dataset for further generality verification. To show that clinical data can be used for COVID-19 survival chance prediction, the CNN-AE was compared with multiple pre-trained deep models that were tuned based on CT images.

Download Full-text

Deep Spectral Spatial Inverted Residual Network for Hyperspectral Image Classification

Remote Sensing ◽

10.3390/rs13214472 ◽

2021 ◽

Vol 13 (21) ◽

pp. 4472

Author(s):

Tianyu Zhang ◽

Cuiping Shi ◽

Diling Liao ◽

Liguo Wang

Keyword(s):

Image Classification ◽

Data Augmentation ◽

Hyperspectral Image ◽

Classification Performance ◽

Hyperspectral Data ◽

Hyperspectral Images ◽

Hyperspectral Image Classification ◽

Feature Maps ◽

Spatial Feature ◽

Global Context

Convolutional neural networks (CNNs) have been widely used in hyperspectral image classification in recent years. The training of CNNs relies on a large amount of labeled sample data. However, the number of labeled samples of hyperspectral data is relatively small. Moreover, for hyperspectral images, fully extracting spectral and spatial feature information is the key to achieve high classification performance. To solve the above issues, a deep spectral spatial inverted residuals network (DSSIRNet) is proposed. In this network, a data block random erasing strategy is introduced to alleviate the problem of limited labeled samples by data augmentation of small spatial blocks. In addition, a deep inverted residuals (DIR) module for spectral spatial feature extraction is proposed, which locks the effective features of each layer while avoiding network degradation. Furthermore, a global 3D attention module is proposed, which can realize the fine extraction of spectral and spatial global context information under the condition of the same number of input and output feature maps. Experiments are carried out on four commonly used hyperspectral datasets. A large number of experimental results show that compared with some state-of-the-art classification methods, the proposed method can provide higher classification accuracy for hyperspectral images.

Download Full-text

On-Road Driver Emotion Recognition Using Facial Expression

Applied Sciences ◽

10.3390/app12020807 ◽

2022 ◽

Vol 12 (2) ◽

pp. 807

Author(s):

Huafei Xiao ◽

Wenbo Li ◽

Guanzhong Zeng ◽

Yingzhang Wu ◽

Jiyong Xue ◽

...

Keyword(s):

Facial Expression ◽

Emotion Recognition ◽

Data Augmentation ◽

Recognition Task ◽

Superior Performance ◽

Expression Recognition ◽

Emotion Detection ◽

Backbone Networks ◽

Driving Task ◽

Machine Systems

With the development of intelligent automotive human-machine systems, driver emotion detection and recognition has become an emerging research topic. Facial expression-based emotion recognition approaches have achieved outstanding results on laboratory-controlled data. However, these studies cannot represent the environment of real driving situations. In order to address this, this paper proposes a facial expression-based on-road driver emotion recognition network called FERDERnet. This method divides the on-road driver facial expression recognition task into three modules: a face detection module that detects the driver’s face, an augmentation-based resampling module that performs data augmentation and resampling, and an emotion recognition module that adopts a deep convolutional neural network pre-trained on FER and CK+ datasets and then fine-tuned as a backbone for driver emotion recognition. This method adopts five different backbone networks as well as an ensemble method. Furthermore, to evaluate the proposed method, this paper collected an on-road driver facial expression dataset, which contains various road scenarios and the corresponding driver’s facial expression during the driving task. Experiments were performed on the on-road driver facial expression dataset that this paper collected. Based on efficiency and accuracy, the proposed FERDERnet with Xception backbone was effective in identifying on-road driver facial expressions and obtained superior performance compared to the baseline networks and some state-of-the-art networks.

Download Full-text

Influence of diet and exercise on skeletal muscle and visceral adipose tissue in men

Journal of Applied Physiology ◽

10.1152/jappl.1996.81.6.2445 ◽

1996 ◽

Vol 81 (6) ◽

pp. 2445-2455 ◽

Cited By ~ 222

Author(s):

Robert Ross ◽

John Rissanen ◽

Heather Pedwell ◽

Jennifer Clifford ◽

Peter Shragge

Keyword(s):

Skeletal Muscle ◽

Body Weight ◽

Adipose Tissue ◽

Visceral Adipose Tissue ◽

Whole Body ◽

Relative Reduction ◽

Body Regions ◽

Diet And Exercise ◽

Magnetic Resonance Imaging Mri ◽

Visceral Adipose

Ross, Robert, John Rissanen, Heather Pedwell, Jennifer Clifford, and Peter Shragge. Influence of diet and exercise on skeletal muscle and visceral adipose tissue in men. J. Appl. Physiol. 81(6): 2445–2455, 1996.—The effects of diet only (DO) and diet combined with either aerobic (DA) or resistance (DR) exercise on subcutaneous adipose tissue (SAT), visceral adipose tissue (VAT), lean tissue (LT), and skeletal muscle (SM) tissue were evaluated in 33 obese men (DO, n= 11; DA, n = 11; DR, n = 11). All tissues were measured by using a whole body multislice magnetic resonance imaging (MRI) model. Within each group, significant reductions were observed for body weight, SAT, and VAT ( P < 0.05). The reductions in body weight (∼10%) and SAT (∼25%) and VAT volume (∼35%) were not different between groups ( P > 0.05). For all treatments, the relative reduction in VAT was greater than in SAT ( P < 0.05). For the DA and DR groups only, the reduction in abdominal SAT (∼27%) was greater ( P < 0.05) than that observed for the gluteal-femoral region (∼20%). Conversely, the reduction in VAT was uniform throughout the abdomen regardless of treatment ( P > 0.05). MRI-LT and MRI-SM decreased both in the upper and lower body regions for the DO group alone ( P < 0.05). Peak O2 uptake (liters) was significantly improved (∼14%) in the DA group as was muscular strength (∼20%) in the DR group ( P< 0.01). These findings indicate that DA and DR result in a greater preservation of MRI-SM, mobilization of SAT from the abdominal region, by comparison with the gluteal-femoral region, and improved functional capacity when compared with DO in obese men.

Download Full-text