scholarly journals Evaluation of Enhanced Learning Techniques for Segmenting Ischaemic Stroke Lesions in Brain Magnetic Resonance Perfusion Images using a Convolutional Neural Network Scheme

2019 ◽  
Author(s):  
Carlos Uziel Perez Malla ◽  
Maria del C. Valdes Hernandez ◽  
Muhammad Febrian Rachmadi ◽  
Taku Komura
Keyword(s):  
Magnetic Resonance ◽  
Ischaemic Stroke ◽  
Data Augmentation ◽  
Imaging Modality ◽  
Brain Mri ◽  
Machine Learning Techniques ◽  
Magnetic Resonance Perfusion ◽  
Mr Perfusion Imaging ◽  
Learning Techniques ◽  
The Impact

ABSTRACTMagnetic resonance (MR) perfusion imaging non-invasively measures cerebral perfusion, which describes the blood’s passage through the brain’s vascular network. Therefore it is widely used to assess cerebral ischaemia. Convolutional Neural Networks (CNN) constitute the state-of-the-art method in automatic pattern recognition and hence, in segmentation tasks. But none of the CNN architectures developed to date have achieved high accuracy when segmenting ischaemic stroke lesions, being the main reasons their heterogeneity in location, shape, size, image intensity and texture, especially in this imaging modality. We use a freely available CNN framework, developed for MR imaging lesion segmentation, as core algorithm to evaluate the impact of enhanced machine learning techniques, namely data augmentation, transfer learning and post-processing, in the segmentation of stroke lesions using the ISLES 2017 dataset, which contains expert annotated diffusion-weighted perfusion and diffusion brain MRI of 43 stroke patients. Of all the techniques evaluated, data augmentation with binary closing achieved the best results, improving the mean Dice score in 17% over the baseline model. Consistent with previous works, better performance was obtained in the presence of large lesions.

A Method of Environmental Sound Classification Based on Residual Networks and Data Augmentation

2021 ◽  
pp. 2150018
Author(s):  
Jinfang Zeng ◽  
Youming Li ◽  
Yu Zhang ◽  
Da Chen
Keyword(s):  
Neural Networks ◽  
Data Augmentation ◽  
Machine Learning Techniques ◽  
Deep Convolutional Neural Networks ◽  
Mel Frequency Cepstral Coefficients ◽  
Environmental Sound ◽  
Sound Classification ◽  
Learning Techniques ◽  
Proposed Model ◽  
Audio Data

Environmental sound classification (ESC) is a challenging problem due to the complexity of sounds. To date, a variety of signal processing and machine learning techniques have been applied to ESC task, including matrix factorization, dictionary learning, wavelet filterbanks and deep neural networks. It is observed that features extracted from deeper networks tend to achieve higher performance than those extracted from shallow networks. However, in ESC task, only the deep convolutional neural networks (CNNs) which contain several layers are used and the residual networks are ignored, which lead to degradation in the performance. Meanwhile, a possible explanation for the limited exploration of CNNs and the difficulty to improve on simpler models is the relative scarcity of labeled data for ESC. In this paper, a residual network called EnvResNet for the ESC task is proposed. In addition, we propose to use audio data augmentation to overcome the problem of data scarcity. The experiments will be performed on the ESC-50 database. Combined with data augmentation, the proposed model outperforms baseline implementations relying on mel-frequency cepstral coefficients and achieves results comparable to other state-of-the-art approaches in terms of classification accuracy.

Brain abnormality detection using template matching

2018 ◽  
Vol 14 (4) ◽  
Author(s):  
G.B. Praveen ◽  
Anita Agrawal ◽  
Shrey Pareek ◽  
Amalin Prince
Keyword(s):  
Magnetic Resonance ◽  
Template Matching ◽  
Imaging Modality ◽  
Brain Mri ◽  
Tumor Segmentation ◽  
Processing Unit ◽  
Brain Disorders ◽  
Brain Tumor Segmentation ◽  
Central Processing ◽  
Wide Range

Abstract Magnetic resonance imaging (MRI) is a widely used imaging modality to evaluate brain disorders. MRI generates huge volumes of data, which consist of a sequence of scans taken at different instances of time. As the presence of brain disorders has to be evaluated on all magnetic resonance (MR) sequences, manual brain disorder detection becomes a tedious process and is prone to inter- and intra-rater errors. A technique for detecting abnormalities in brain MRI using template matching is proposed. Bias filed correction is performed on volumetric scans using N4ITK filter, followed by volumetric registration. Normalized cross-correlation template matching is used for image registration taking into account, the rotation and scaling operations. A template of abnormality is selected which is then matched in the volumetric scans, if found, the corresponding image is retrieved. Post-processing of the retrieved images is performed by the thresholding operation; the coordinates and area of the abnormality are reported. The experiments are carried out on the glioma dataset obtained from Brain Tumor Segmentation Challenge 2013 database (BRATS 2013). Glioma dataset consisted of MR scans of 30 real glioma patients and 50 simulated glioma patients. NVIDIA Compute Unified Device Architecture framework is employed in this paper, and it is found that the detection speed using graphics processing unit is almost four times faster than using only central processing unit. The average Dice and Jaccard coefficients for a wide range of trials are found to be 0.91 and 0.83, respectively.

scholarly journals Machine learning analysis of lifeguard flag decisions and recorded rescues

2019 ◽  
Vol 19 (11) ◽  
pp. 2541-2549
Author(s):  
Chris Houser ◽  
Jacob Lehner ◽  
Nathan Cherry ◽  
Phil Wernette
Keyword(s):  
Machine Learning ◽  
The United States ◽  
Machine Learning Techniques ◽  
Public Health Issue ◽  
Rip Currents ◽  
Effective Strategies ◽  
Wave Forcing ◽  
Learning Techniques ◽  
Yellow Flag ◽  
The Impact

Abstract. Rip currents and other surf hazards are an emerging public health issue globally. Lifeguards, warning flags, and signs are important, and to varying degrees they are effective strategies to minimize risk to beach users. In the United States and other jurisdictions around the world, lifeguards use coloured flags (green, yellow, and red) to indicate whether the danger posed by the surf and rip hazard is low, moderate, or high respectively. The choice of flag depends on the lifeguard(s) monitoring the changing surf conditions along the beach and over the course of the day using both regional surf forecasts and careful observation. There is a potential that the chosen flag is not consistent with the beach user perception of the risk, which may increase the potential for rescues or drownings. In this study, machine learning is used to determine the potential for error in the flags used at Pensacola Beach and the impact of that error on the number of rescues. Results of a decision tree analysis indicate that the colour flag chosen by the lifeguards was different from what the model predicted for 35 % of days between 2004 and 2008 (n=396/1125). Days when there is a difference between the predicted and posted flag colour represent only 17 % of all rescue days, but those days are associated with ∼60 % of all rescues between 2004 and 2008. Further analysis reveals that the largest number of rescue days and total number of rescues are associated with days where the flag deployed over-estimated the surf and hazard risk, such as a red or yellow flag flying when the model predicted a green flag would be more appropriate based on the wind and wave forcing alone. While it is possible that the lifeguards were overly cautious, it is argued that they most likely identified a rip forced by a transverse-bar and rip morphology common at the study site. Regardless, the results suggest that beach users may be discounting lifeguard warnings if the flag colour is not consistent with how they perceive the surf hazard or the regional forecast. Results suggest that machine learning techniques have the potential to support lifeguards and thereby reduce the number of rescues and drownings.

scholarly journals Monitoring the Impact of Air Quality on the COVID-19 Fatalities in Delhi, India: Using Machine Learning Techniques

2020 ◽  
pp. 1-8
Author(s):  
Jasleen Kaur Sethi ◽  
Mamta Mittal
Keyword(s):  
Machine Learning ◽  
Air Quality ◽  
Air Pollutants ◽  
Machine Learning Techniques ◽  
Environmental Restoration ◽  
The Novel ◽  
Ozone Pollution ◽  
Learning Techniques ◽  
Novel Coronavirus ◽  
The Impact

ABSTRACT Objective: The focus of this study is to monitor the effect of lockdown on the various air pollutants due to the coronavirus disease (COVID-19) pandemic and identify the ones that affect COVID-19 fatalities so that measures to control the pollution could be enforced. Methods: Various machine learning techniques: Decision Trees, Linear Regression, and Random Forest have been applied to correlate air pollutants and COVID-19 fatalities in Delhi. Furthermore, a comparison between the concentration of various air pollutants and the air quality index during the lockdown period and last two years, 2018 and 2019, has been presented. Results: From the experimental work, it has been observed that the pollutants ozone and toluene have increased during the lockdown period. It has also been deduced that the pollutants that may impact the mortalities due to COVID-19 are ozone, NH3, NO2, and PM10. Conclusions: The novel coronavirus has led to environmental restoration due to lockdown. However, there is a need to impose measures to control ozone pollution, as there has been a significant increase in its concentration and it also impacts the COVID-19 mortality rate.

scholarly journals Knocking and Listening: Learning Mechanical Impulse Response for Understanding Surface Characteristics

Sensors ◽  
2020 ◽  
Vol 20 (2) ◽  
pp. 369 ◽  
Author(s):  
Semin Ryu ◽  
Seung-Chan Kim
Keyword(s):  
Intelligent System ◽  
Surface Characteristics ◽  
Machine Learning Techniques ◽  
Test Accuracy ◽  
Learning Approaches ◽  
Well Test ◽  
Mechanical Responses ◽  
Learning Techniques ◽  
The Impact ◽  
Mechanical Impulse

Inspired by spiders that can generate and sense vibrations to obtain information regarding a substrate, we propose an intelligent system that can recognize the type of surface being touched by knocking the surface and listening to the vibrations. Hence, we developed a system that is equipped with an electromagnetic hammer for hitting the ground and an accelerometer for measuring the mechanical responses induced by the impact. We investigate the feasibility of sensing 10 different daily surfaces through various machine-learning techniques including recent deep-learning approaches. Although some test surfaces are similar, experimental results show that our system can recognize 10 different surfaces remarkably well (test accuracy of 98.66%). In addition, our results without directly hitting the surface (internal impact) exhibited considerably high test accuracy (97.51%). Finally, we conclude this paper with the limitations and future directions of the study.

scholarly journals Predictive Modeling for Occupational Safety Outcomes and Days Away from Work Analysis in Mining Operations

2020 ◽  
Vol 17 (19) ◽  
pp. 7054
Author(s):  
Anurag Yedla ◽  
Fatemeh Davoudi Kakhki ◽  
Ali Jannesari
Keyword(s):  
Machine Learning ◽  
Data Augmentation ◽  
Predictive Analytics ◽  
Machine Learning Algorithms ◽  
Mine Safety ◽  
Machine Learning Techniques ◽  
Health Administration ◽  
Tabular Data ◽  
Mining Operations ◽  
Learning Techniques

Mining is known to be one of the most hazardous occupations in the world. Many serious accidents have occurred worldwide over the years in mining. Although there have been efforts to create a safer work environment for miners, the number of accidents occurring at the mining sites is still significant. Machine learning techniques and predictive analytics are becoming one of the leading resources to create safer work environments in the manufacturing and construction industries. These techniques are leveraged to generate actionable insights to improve decision-making. A large amount of mining safety-related data are available, and machine learning algorithms can be used to analyze the data. The use of machine learning techniques can significantly benefit the mining industry. Decision tree, random forest, and artificial neural networks were implemented to analyze the outcomes of mining accidents. These machine learning models were also used to predict days away from work. An accidents dataset provided by the Mine Safety and Health Administration was used to train the models. The models were trained separately on tabular data and narratives. The use of a synthetic data augmentation technique using word embedding was also investigated to tackle the data imbalance problem. Performance of all the models was compared with the performance of the traditional logistic regression model. The results show that models trained on narratives performed better than the models trained on structured/tabular data in predicting the outcome of the accident. The higher predictive power of the models trained on narratives led to the conclusion that the narratives have additional information relevant to the outcome of injury compared to the tabular entries. The models trained on tabular data had a lower mean squared error compared to the models trained on narratives while predicting the days away from work. The results highlight the importance of predictors, like shift start time, accident time, and mining experience in predicting the days away from work. It was found that the F1 score of all the underrepresented classes except one improved after the use of the data augmentation technique. This approach gave greater insight into the factors influencing the outcome of the accident and days away from work.

Sign in / Sign up

Export Citation Format

Share Document