scholarly journals How to Predict the Long-term Course of Neurodegenerative Diseases?

2020 ◽  
Vol 8 ◽  
Author(s):  
Alberto Montolío Marco ◽  
José Cegoñino Banzo ◽  
Elena García Martín ◽  
Amaya Pérez del Palomar Aldea
Keyword(s):  
Machine Learning ◽  
Multiple Sclerosis ◽  
Neurodegenerative Diseases ◽  
Neurodegenerative Disease ◽  
Specific Treatment ◽  
Machine Learning Techniques ◽  
Learning Techniques

The aim of this work is to predict the disability state in neurodegenerative disease, such as multiple sclerosis (MS), using clinical batabases and machine learning techniques. This prediction could help clinicians select a more specific treatment for MS patients.

Long-term Cholesterol Risk Prediction using Machine Learning Techniques in ELSA Database

2021 ◽  
Author(s):  
Nikos Fazakis ◽  
Elias Dritsas ◽  
Otilia Kocsis ◽  
Nikos Fakotakis ◽  
Konstantinos Moustakas
Keyword(s):  
Machine Learning ◽  
Risk Prediction ◽  
Machine Learning Techniques ◽  
Learning Techniques

Machine Learning for Neurodegenerative Disorder Diagnosis — Survey of Practices and Launch of Benchmark Dataset

2018 ◽  
Vol 27 (03) ◽  
pp. 1850011 ◽  
Author(s):  
Athanasios Tagaris ◽  
Dimitrios Kollias ◽  
Andreas Stafylopatis ◽  
Georgios Tagaris ◽  
Stefanos Kollias
Keyword(s):  
Machine Learning ◽  
Neurodegenerative Disorder ◽  
Developed Countries ◽  
Machine Learning Techniques ◽  
Network Architectures ◽  
Imaging Data ◽  
Learning Techniques ◽  
Medical Examinations

Neurodegenerative disorders, such as Alzheimer’s and Parkinson’s, constitute a major factor in long-term disability and are becoming more and more a serious concern in developed countries. As there are, at present, no effective therapies, early diagnosis along with avoidance of misdiagnosis seem to be critical in ensuring a good quality of life for patients. In this sense, the adoption of computer-aided-diagnosis tools can offer significant assistance to clinicians. In the present paper, we provide in the first place a comprehensive recording of medical examinations relevant to those disorders. Then, a review is conducted concerning the use of Machine Learning techniques in supporting diagnosis of neurodegenerative diseases, with reference to at times used medical datasets. Special attention has been given to the field of Deep Learning. In addition to that, we communicate the launch of a newly created dataset for Parkinson’s disease, containing epidemiological, clinical and imaging data, which will be publicly available to researchers for benchmarking purposes. To assess the potential of the new dataset, an experimental study in Parkinson’s diagnosis is carried out, based on state-of-the-art Deep Neural Network architectures and yielding very promising accuracy results.

Monitoring Effects of Drought on Nitrogen and Phosphorus in Temperate Oak Forests Using Machine Learning Techniques

2018 ◽  
Author(s):  
Jan Kotlarz ◽  
Katarzyna Kubiak ◽  
Marcin Spiralski
Keyword(s):  
Machine Learning ◽  
Drought Stress ◽  
Negative Impact ◽  
Aerial Images ◽  
Machine Learning Techniques ◽  
Nutrient Concentrations ◽  
Nitrogen And Phosphorus ◽  
Learning Techniques ◽  
The Impact

Oak is a European tree species highly sensitive to drought. If declining symptoms appear they are often detectable at the crown (such as dieback) enabling monitoring using aerial images and remote sensing methods. Here, we analyzed the impact of short and long-term drought on oaks located in central Poland, between the years of 2014 and 2017. We used leaf nitrogen (N) and phosphorus (P) concentrations measured in the laboratory, aerial images collected in the range of 460-880 nm and machine learning techniques to estimate nutrient concentrations on the > 4000 oaks growing on gleysoil in the study area. We determined a negative impact on N and P concentrations during both types of drought stress (-23% and 19% for N concentration in leaves; -27% and -10% for P concentration in leaves) and an inconsiderable impact on N:P values (3% increase of N:P ration during short and 7% decrease of N:P ration during long-term drought stress). We found that the long-term drought impact was spatially diverse, possibly depending on the presence of drainage ditches and competing species.

Prediction of Neurodegenerative Diseases Based on Gait Signals Using Supervised Machine Learning Techniques

2018 ◽  
Vol 24 (3) ◽  
pp. 1974-1978 ◽  
Author(s):  
Satyabrata Aich ◽  
Ki-Won Choi ◽  
Pyari Mohan Pradhan ◽  
Jinse Park ◽  
Hee-Cheol Kim
Keyword(s):  
Machine Learning ◽  
Neurodegenerative Diseases ◽  
Supervised Machine Learning ◽  
Machine Learning Techniques ◽  
Learning Techniques

scholarly journals Modeling a Competence-based Industry Role Selection System for University Graduates Using Machine Learning

2020 ◽  
Vol 1 (1) ◽  
pp. 15-25
Author(s):  
Fullgence Mwachoo Mwakondo
Keyword(s):  
Machine Learning ◽  
Prior Information ◽  
Naive Bayes ◽  
Machine Learning Techniques ◽  
Selection System ◽  
Object Modeling ◽  
Learning Techniques ◽  
Blind Search ◽  
And Behavior

This paper presents a design of a system for industry role selection, representing both its structure and behavior. Knowing the right industry role that suits a graduate based on their competences on graduation has remained a critical matter for graduates when searching for jobs after graduation. Thousands of university students graduate each year and enter the market to search for jobs that are limited. Searching without prior information on the most appropriate industry role one is suitable for leads to blind search. Blind search not only puts graduates at risk of long-term unemployment and job mismatch but also overloads employers with many applications during job selection. Therefore, this paper addresses 2 objectives: 1) to model the system’s structure, and 2) to design the algorithm for the system’s behavior. Since object-oriented programming is currently the dominant programming paradigm, object modeling technique was selected to model both the system’s structure and the algorithm for the system’s behavior. To realize object modeling and represent the system’s artifacts in a highly simplified form, Unified Modeling Language (UML) was adopted as the standard modeling toolkit. More specifically, UML class diagram was used to represent the structural model of the system where the underlying objects of the model were exactly similar to those of the problem domain. Finally, use case diagram of the UML toolkit was used to represent the system’s behavior in selecting industry role for graduates. To ensure that the system improves performance of its behavior through experience in selecting industry roles for graduates, Machine Learning (ML) algorithm was designed. Two machine learning techniques, naïve Bayes and Support Vector Machines (SVM), were used as the algorithm’s criteria for selecting industry roles for graduates. Experiments to evaluate performance of the system were conducted using data collected from Software Engineering industry domain. The end product was design of an intelligent industry role selection system with relevant structure and behavior to easily work with both in the academia and industry. Findings reveal the system improves performance of its behavior in selecting industry roles for graduates much better under SVM (67%) than naïve Bayes (57%). On the same benchmark dataset, the system recorded better performance (85%) than reported performance (82%) in the benchmark system. These findings will benefit industry by getting evaluation tool for revealing graduate’s suitability for employment which they can use as prior information for decision making when filtering candidates for interview. Besides, this will provide researchers with a digital platform to study and bridge the gap between industry and academia. Lastly, this will attempt to reduce both low job satisfaction and long-term unemployment that is one of the causes of social and economic pain both in Kenya and around the world. This paper has revealed competence based industry role selection system with relevant structure and behavior can improve searching of jobs by providing a fairly accurate prior information. However, this paper recommends testing this approach with other alternative machine learning techniques as well as other alternative industry domains.

scholarly journals Machine Learning and Data Segmentation for Building Energy Use Prediction—A Comparative Study

Energies ◽  
2021 ◽  
Vol 14 (18) ◽  
pp. 5947
Author(s):  
William Mounter ◽  
Chris Ogwumike ◽  
Huda Dawood ◽  
Nashwan Dawood
Keyword(s):  
Machine Learning ◽  
Building Energy ◽  
Training Data ◽  
Machine Learning Techniques ◽  
Support Vector ◽  
Energy Usage ◽  
Research Attention ◽  
Learning Techniques ◽  
Term Energy

Advances in metering technologies and emerging energy forecast strategies provide opportunities and challenges for predicting both short and long-term building energy usage. Machine learning is an important energy prediction technique, and is significantly gaining research attention. The use of different machine learning techniques based on a rolling-horizon framework can help to reduce the prediction error over time. Due to the significant increases in error beyond short-term energy forecasts, most reported energy forecasts based on statistical and machine learning techniques are within the range of one week. The aim of this study was to investigate how facility managers can improve the accuracy of their building’s long-term energy forecasts. This paper presents an extensive study of machine learning and data processing techniques and how they can more accurately predict within different forecast ranges. The Clarendon building of Teesside University was selected as a case study to demonstrate the prediction of overall energy usage with different machine learning techniques such as polynomial regression (PR), support vector regression (SVR) and artificial neural networks (ANNs). This study further examined how preprocessing training data for prediction models can impact the overall accuracy, such as via segmenting the training data by building modes (active and dormant), or by days of the week (weekdays and weekends). The results presented in this paper illustrate a significant reduction in the mean absolute percentage error (MAPE) for segmented building (weekday and weekend) energy usage prediction when compared to unsegmented monthly predictions. A reduction in MAPE of 5.27%, 11.45%, and 12.03% was achieved with PR, SVR and ANN, respectively.

scholarly journals Ensemble learning predicts multiple sclerosis disease course in the SUMMIT study

2020 ◽  
Vol 3 (1) ◽  
Author(s):  
Yijun Zhao ◽  
◽  
Tong Wang ◽  
Riley Bove ◽  
Bruce Cree ◽  
...  
Keyword(s):  
Machine Learning ◽  
Multiple Sclerosis ◽  
San Francisco ◽  
Ensemble Learning ◽  
Disease Onset ◽  
Machine Learning Algorithms ◽  
Machine Learning Techniques ◽  
Learning Approaches ◽  
Disease Course ◽  
Learning Techniques

AbstractThe rate of disability accumulation varies across multiple sclerosis (MS) patients. Machine learning techniques may offer more powerful means to predict disease course in MS patients. In our study, 724 patients from the Comprehensive Longitudinal Investigation in MS at Brigham and Women’s Hospital (CLIMB study) and 400 patients from the EPIC dataset, University of California, San Francisco, were included in the analysis. The primary outcome was an increase in Expanded Disability Status Scale (EDSS) ≥ 1.5 (worsening) or not (non-worsening) at up to 5 years after the baseline visit. Classification models were built using the CLIMB dataset with patients’ clinical and MRI longitudinal observations in first 2 years, and further validated using the EPIC dataset. We compared the performance of three popular machine learning algorithms (SVM, Logistic Regression, and Random Forest) and three ensemble learning approaches (XGBoost, LightGBM, and a Meta-learner L). A “threshold” was established to trade-off the performance between the two classes. Predictive features were identified and compared among different models. Machine learning models achieved 0.79 and 0.83 AUC scores for the CLIMB and EPIC datasets, respectively, shortly after disease onset. Ensemble learning methods were more effective and robust compared to standalone algorithms. Two ensemble models, XGBoost and LightGBM were superior to the other four models evaluated in our study. Of variables evaluated, EDSS, Pyramidal Function, and Ambulatory Index were the top common predictors in forecasting the MS disease course. Machine learning techniques, in particular ensemble methods offer increased accuracy for the prediction of MS disease course.

Sign in / Sign up

Export Citation Format

Share Document