scholarly journals Dysregulated Alanine as a Potential Predictive Marker of Glioma—An Insight from Untargeted HRMAS-NMR and Machine Learning Data

Metabolites ◽  
2021 ◽  
Vol 11 (8) ◽  
pp. 507
Author(s):  
Safia Firdous ◽  
Rizwan Abid ◽  
Zubair Nawaz ◽  
Faisal Bukhari ◽  
Ammar Anwer ◽  
...  
Keyword(s):  
Machine Learning ◽  
Logistic Regression ◽  
Random Forest ◽  
Magic Angle Spinning ◽  
Diagnostic Methods ◽  
Wilcoxon Test ◽  
Resonance Spectroscopy ◽  
Magic Angle ◽  
Tree Classifier ◽  
Hrmas Nmr

Metabolic alterations play a crucial role in glioma development and progression and can be detected even before the appearance of the fatal phenotype. We have compared the circulating metabolic fingerprints of glioma patients versus healthy controls, for the first time, in a quest to identify a panel of small, dysregulated metabolites with potential to serve as a predictive and/or diagnostic marker in the clinical settings. High-resolution magic angle spinning nuclear magnetic resonance spectroscopy (HRMAS-NMR) was used for untargeted metabolomics and data acquisition followed by a machine learning (ML) approach for the analyses of large metabolic datasets. Cross-validation of ML predicted NMR spectral features was done by statistical methods (Wilcoxon-test) using JMP-pro16 software. Alanine was identified as the most critical metabolite with potential to detect glioma with precision of 1.0, recall of 0.96, and F1 measure of 0.98. The top 10 metabolites identified for glioma detection included alanine, glutamine, valine, methionine, N-acetylaspartate (NAA), γ-aminobutyric acid (GABA), serine, α-glucose, lactate, and arginine. We achieved 100% accuracy for the detection of glioma using ML algorithms, extra tree classifier, and random forest, and 98% accuracy with logistic regression. Classification of glioma in low and high grades was done with 86% accuracy using logistic regression model, and with 83% and 79% accuracy using extra tree classifier and random forest, respectively. The predictive accuracy of our ML model is superior to any of the previously reported algorithms, used in tissue- or liquid biopsy-based metabolic studies. The identified top metabolites can be targeted to develop early diagnostic methods as well as to plan personalized treatment strategies.

scholarly journals Machine Learning Assisted Intraoperative Assessment of Brain Tumor Margins Using HRMAS NMR Spectroscopy

2020 ◽  
Author(s):  
Doruk Cakmakci ◽  
Emin Onur Karakaslar ◽  
Elisa Ruhland ◽  
Marie-Pierre Chenard ◽  
Francois Proust ◽  
...  
Keyword(s):  
Machine Learning ◽  
Real Time ◽  
Magic Angle Spinning ◽  
Magic Angle ◽  
Full Spectrum ◽  
Real Time Analysis ◽  
Targeted Analysis ◽  
Hrmas Nmr ◽  
Angle Spinning ◽  
And Control

AbstractComplete resection of the tumor is important for survival in glioma patients. Even if the gross total resection was achieved, left-over micro-scale tissue in the excision cavity risks recurrence. High Resolution Magic Angle Spinning Nuclear Magnetic Resonance (HRMAS NMR) technique can distinguish healthy and malign tissue efficiently using peak intensities of biomarker metabolites. The method is fast, sensitive and can work with small and unprocessed samples, which makes it a good fit for real-time analysis during surgery. However, only a targeted analysis for the existence of known tumor biomarkers can be made and this requires a technician with chemistry background, and a pathologist with knowledge on tumor metabolism to be present during surgery. Here, we show that we can accurately perform this analysis in real-time and can analyze the full spectrum in an untargeted fashion using machine learning. We work on a new and large HRMAS NMR dataset of glioma and control samples (n = 568), which are also labeled with a quantitative pathology analysis. Our results show that a random forest based approach can distinguish samples with tumor cells and controls accurately and effectively with a mean AUC of 85.6% and AUPR of 93.4%. We also show that we can further distinguish benign and malignant samples with a mean AUC of 87.1% and AUPR of 96.1%. We analyze the feature (peak) importance for classification to interpret the results of the classifier. We validate that known malignancy biomarkers such as creatine and 2-hydroxyglutarate play an important role in distinguish tumor and normal cells and suggest new biomarker regions. The code is released at http://github.com/ciceklab/HRMAS_NC.

scholarly journals Detection of Local Prostate Metabolites by Hrmas Nmr Spectroscopy: A Comparative Study of Human and Rat Prostate Tissues

2010 ◽  
Vol 4 ◽  
pp. MRI.S6028 ◽  
Author(s):  
Katarina Stenman ◽  
Izabella Surowiec ◽  
Henrik Antti ◽  
Katrine Riklund ◽  
Pär Stattin ◽  
...  
Keyword(s):  
Prostate Cancer ◽  
Nmr Spectroscopy ◽  
Magnetic Resonance ◽  
Magic Angle Spinning ◽  
General Distribution ◽  
Resonance Spectroscopy ◽  
Magic Angle ◽  
Hrmas Nmr ◽  
Rat Prostate

The use of magnetic resonance spectroscopy (MRS) for the detection of in-vivo metabolic perturbations is increasing in popularity in Prostate Cancer (PCa) research on both humans and rodent models. However, there are distinct metabolic differences between species and prostate areas; a fact making general conclusions about PCa difficult. Here, we use High Resolution Magic Angle Spinning Nuclear Magnetic Resonance (HRMAS NMR) spectroscopy to provide tissue specific identification of metabolites and their relative ratios; information useful in providing insight into the biochemical pathways of the prostate. As our NMR-based approach reveals, human and rat prostate tissues have different metabolic signatures as reflected in numerous key metabolites, including citrate and choline compounds, but also aspartate, lysine, taurine, glutamate, glutamine, creatine and inositol. In general, distribution of these metabolites is not only highly dependent on the species (human versus rat), but also on the location (lobe/zone) in the prostate tissue and the sample pathology; an observation making HRMAS NMR of intact tissue samples a promising method for extracting differences and common features in various experimental prostate cancer models.

scholarly journals Swindling Shonky Anatomization of Credit Card Transactions using Machine Learning

2019 ◽  
Vol 8 (4) ◽  
pp. 1477-1483
Keyword(s):  
Machine Learning ◽  
Logistic Regression ◽  
Random Forest ◽  
Decision Tree ◽  
Credit Card ◽  
Naive Bayes ◽  
Gradient Boosting ◽  
Decision Tree Classifier ◽  
Tree Classifier ◽  
Feature Importance

With the fast moving technological advancement, the internet usage has been increased rapidly in all the fields. The money transactions for all the applications like online shopping, banking transactions, bill settlement in any industries, online ticket booking for travel and hotels, Fees payment for educational organization, Payment for treatment to hospitals, Payment for super market and variety of applications are using online credit card transactions. This leads to the fraud usage of other accounts and transaction that result in the loss of service and profit to the institution. With this background, this paper focuses on predicting the fraudulent credit card transaction. The Credit Card Transaction dataset from KAGGLE machine learning Repository is used for prediction analysis. The analysis of fraudulent credit card transaction is achieved in four ways. Firstly, the relationship between the variables of the dataset is identified and represented by the graphical notations. Secondly, the feature importance of the dataset is identified using Random Forest, Ada boost, Logistic Regression, Decision Tree, Extra Tree, Gradient Boosting and Naive Bayes classifiers. Thirdly, the extracted feature importance if the credit card transaction dataset is fitted to Random Forest classifier, Ada boost classifier, Logistic Regression classifier, Decision Tree classifier, Extra Tree classifier, Gradient Boosting classifier and Naive Bayes classifier. Fourth, the Performance Analysis is done by analyzing the performance metrics like Accuracy, FScore, AUC Score, Precision and Recall. The implementation is done by python in Anaconda Spyder Navigator Integrated Development Environment. Experimental Results shows that the Decision Tree classifier have achieved the effective prediction with the precision of 1.0, recall of 1.0, FScore of 1.0 , AUC Score of 89.09 and Accuracy of 99.92%.

scholarly journals Machine learning assisted intraoperative assessment of brain tumor margins using HRMAS NMR spectroscopy

2020 ◽  
Vol 16 (11) ◽  
pp. e1008184
Author(s):  
Doruk Cakmakci ◽  
Emin Onur Karakaslar ◽  
Elisa Ruhland ◽  
Marie-Pierre Chenard ◽  
Francois Proust ◽  
...  
Keyword(s):  
Machine Learning ◽  
Real Time ◽  
Magic Angle Spinning ◽  
Magic Angle ◽  
Full Spectrum ◽  
Real Time Analysis ◽  
Targeted Analysis ◽  
Hrmas Nmr ◽  
Angle Spinning ◽  
And Control

Complete resection of the tumor is important for survival in glioma patients. Even if the gross total resection was achieved, left-over micro-scale tissue in the excision cavity risks recurrence. High Resolution Magic Angle Spinning Nuclear Magnetic Resonance (HRMAS NMR) technique can distinguish healthy and malign tissue efficiently using peak intensities of biomarker metabolites. The method is fast, sensitive and can work with small and unprocessed samples, which makes it a good fit for real-time analysis during surgery. However, only a targeted analysis for the existence of known tumor biomarkers can be made and this requires a technician with chemistry background, and a pathologist with knowledge on tumor metabolism to be present during surgery. Here, we show that we can accurately perform this analysis in real-time and can analyze the full spectrum in an untargeted fashion using machine learning. We work on a new and large HRMAS NMR dataset of glioma and control samples (n = 565), which are also labeled with a quantitative pathology analysis. Our results show that a random forest based approach can distinguish samples with tumor cells and controls accurately and effectively with a median AUC of 85.6% and AUPR of 93.4%. We also show that we can further distinguish benign and malignant samples with a median AUC of 87.1% and AUPR of 96.1%. We analyze the feature (peak) importance for classification to interpret the results of the classifier. We validate that known malignancy biomarkers such as creatine and 2-hydroxyglutarate play an important role in distinguishing tumor and normal cells and suggest new biomarker regions. The code is released at http://github.com/ciceklab/HRMAS_NC.

scholarly journals Fake News Data Exploration and Analytics

Electronics ◽  
2021 ◽  
Vol 10 (19) ◽  
pp. 2326
Author(s):  
Mazhar Javed Awan ◽  
Awais Yasin ◽  
Haitham Nobanee ◽  
Ahmed Abid Ali ◽  
Zain Shahzad ◽  
...  
Keyword(s):  
Machine Learning ◽  
Logistic Regression ◽  
Random Forest ◽  
Random Forest Classifier ◽  
The Internet ◽  
Fake News ◽  
Learning Models ◽  
Decision Tree Classifier ◽  
Tree Classifier ◽  
Machine Learning Models

Before the internet, people acquired their news from the radio, television, and newspapers. With the internet, the news moved online, and suddenly, anyone could post information on websites such as Facebook and Twitter. The spread of fake news has also increased with social media. It has become one of the most significant issues of this century. People use the method of fake news to pollute the reputation of a well-reputed organization for their benefit. The most important reason for such a project is to frame a device to examine the language designs that describe fake and right news through machine learning. This paper proposes models of machine learning that can successfully detect fake news. These models identify which news is real or fake and specify the accuracy of said news, even in a complex environment. After data-preprocessing and exploration, we applied three machine learning models; random forest classifier, logistic regression, and term frequency-inverse document frequency (TF-IDF) vectorizer. The accuracy of the TFIDF vectorizer, logistic regression, random forest classifier, and decision tree classifier models was approximately 99.52%, 98.63%, 99.63%, and 99.68%, respectively. Machine learning models can be considered a great choice to find reality-based results and applied to other unstructured data for various sentiment analysis applications.

scholarly journals Chronic Kidney Disease Prediction using Machine Learning Algorithms

2021 ◽  
Vol 1 (3) ◽  
pp. 1-4
Author(s):  
Kallu Samatha ◽  
Muppidi Rohitha Reddy ◽  
Pattan Faizal Khan ◽  
Rayapati Akhil Chowdary ◽  
P.V.R.D Prasada Rao
Keyword(s):  
Machine Learning ◽  
Chronic Kidney Disease ◽  
Logistic Regression ◽  
Random Forest ◽  
Kidney Disease ◽  
Decision Tree ◽  
Kidney Diseases ◽  
Disease Prediction ◽  
Decision Tree Classifier ◽  
Tree Classifier

Kidney diseases are increasing day by day among people. It is becoming a major health issue around the world. Not maintaining proper food habits and drinking less amount of water are one of the major reasons that contribute this condition. With this, it has become necessary to build up a system to foresee Chronic Kidney Diseases precisely. Here, we have proposed an approach for real time kidney disease prediction. Our aim is to find the best and efficient machine learning (ML) application that can effectively recognize and predict the condition of chronic kidney disease. We have used the data from UCI machine learning repository. In this work, five important machine learning classification techniques were considered for predicting chronic kidney disease which are KNN, Logistic Regression, Random Forest Classifier, SVM and Decision Tree Classifier. In this process, the data has been divided into two sections. In one section train dataset got trained and another section got evaluated by test dataset. The analysis results show that Decision Tree Classifier and Logistic Regression algorithms achieved highest performance than the other classifiers, obtaining the accuracy of 98.75% followed by random Forest, which stands at 97.5%.

scholarly journals Pendekatan Data Science untuk Menemukan Churn Pelanggan pada Sector Perbankan dengan Machine Learning

2021 ◽  
Vol 1 (1) ◽  
pp. 8-13
Author(s):  
Amir Mahmud Husein ◽  
Mawaddah Harahap
Keyword(s):  
Machine Learning ◽  
Logistic Regression ◽  
Random Forest ◽  
Data Science ◽  
Random Forest Classifier ◽  
Random Tree ◽  
Chi Square ◽  
Tree Classifier

Peralihan pelanggan merupakan fenomena dimana pelanggan perusahaan berhenti membeli atau berinteraksi sehingga sangat penting bagi perusahaan khususnya perbankan untuk memprediksi kemungkinan churn pelanggan dan hasilnya dapat digunakan untuk membantu retensi pelanggan dan bagian dari strategi perusahaan. Makalah ini menyajikan analisis dan prediksi churn pelanggan dengan menggunakan lima model berbeda yaitu Kneighbors Classifier, Logistic Regression, Linear SVC, Random Tree Classifier dan Random Forest Classifier. Berdasarkan hasil pengujian pendekatan model Random Forest Classifier dan Kneighbors Classifier lebih baik dari pada model lain dengan akurasi sebesar 86% dan 84%. Rekayasa fitur dengan pendekatan Anova dan Chi Square memiliki pengaruh yang signifikan terhadap peningkatan kinerja model prediksi.

scholarly journals Chronic Kidney Disease Prediction using Machine Learning Algorithms

2021 ◽  
pp. 1-4
Author(s):  
Ms. Kallu Samatha ◽  
◽  
Ms. Muppidi Rohitha Reddy ◽  
Mr. Pattan Faizal Khan ◽  
Mr. Rayapati Akhil Chowdary ◽  
...  
Keyword(s):  
Machine Learning ◽  
Chronic Kidney Disease ◽  
Logistic Regression ◽  
Random Forest ◽  
Kidney Disease ◽  
Decision Tree ◽  
Kidney Diseases ◽  
Disease Prediction ◽  
Decision Tree Classifier ◽  
Tree Classifier

Kidney diseases are increasing day by day among people. It is becoming a major health issue around the world. Not maintaining proper food habits and drinking less amount of water are one of the major reasons that contribute this condition. With this, it has become necessary to build up a system to foresee Chronic Kidney Diseases precisely. Here, we have proposed an approach for real time kidney disease prediction. Our aim is to find the best and efficient machine learning (ML) application that can effectively recognize and predict the condition of chronic kidney disease. We have used the data from UCI machine learning repository. In this work, five important machine learning classification techniques were considered for predicting chronic kidney disease which are KNN, Logistic Regression, Random Forest Classifier, SVM and Decision Tree Classifier. In this process, the data has been divided into two sections. In one section train dataset got trained and another section got evaluated by test dataset. The analysis results show that Decision Tree Classifier and Logistic Regression algorithms achieved highest performance than the other classifiers, obtaining the accuracy of 98.75% followed by random Forest, which stands at 97.5%.

scholarly journals MBRS-69. METABOLITE PROFILING OF SHH MEDULLOBLASTOMA IDENTIFIES A SUBSET OF CHILDHOOD TUMOURS ENRICHED FOR HIGH-RISK MOLECULAR BIOMARKERS AND CLINICAL FEATURES

2020 ◽  
Vol 22 (Supplement_3) ◽  
pp. iii410-iii410
Author(s):  
Christopher Bennett ◽  
Sarah Kohe ◽  
Florence Burte ◽  
Heather Rose ◽  
Debbie Hicks ◽  
...  
Keyword(s):  
High Risk ◽  
Metabolite Profiling ◽  
Unmet Need ◽  
Magic Angle Spinning ◽  
Molecular Biomarkers ◽  
Mycn Amplification ◽  
Resonance Spectroscopy ◽  
Magic Angle ◽  
Metabolite Profiles ◽  
Cluster 2

Abstract SHH medulloblastoma patients have a variable prognosis. Infants (<3–5 years at diagnosis) are associated with a good prognosis, while disease-course in childhood is associated with specific prognostic biomarkers (MYCN amplification, TP53 mutation, LCA histology; all high-risk). There is an unmet need to identify prognostic subgroups of SHH tumours rapidly in the clinical setting, to aid in real-time risk stratification and disease management. Metabolite profiling is a powerful technique for characterising tumours. High resolution magic angle spinning NMR spectroscopy (HR-MAS) can be performed on frozen tissue samples and provides high quality metabolite information. We therefore assessed whether metabolite profiles could identify subsets of SHH tumours with prognostic potential. Metabolite concentrations of 22 SHH tumours were acquired by HR-MAS and analysed using unsupervised hierarchical clustering. Methylation profiling assigned the infant and childhood SHH subtypes, and clinical and molecular features were compared between clusters. Two clusters were observed. A significantly higher concentration of lipids was observed in Cluster 1 (t-test, p=0.012). Cluster 1 consisted entirely of childhood-SHH whilst Cluster 2 included both childhood-SHH and infant-SHH subtypes. Cluster 1 was enriched for high-risk markers - LCA histology (3/7 v. 0/5), MYCN amplification (2/7 v. 0/5), TP53 mutations (3/7 v. 1/5) and metastatic disease - whilst having a lower proportion of TERT mutations (0/7 v. 2/5) than Cluster 2. These pilot results suggest that (i) it is possible to identify childhood-SHH patients linked to high-risk clinical and molecular biomarkers using metabolite profiles and (ii) these may be detected non-invasively in vivo using magnetic-resonance spectroscopy.

Sign in / Sign up

Export Citation Format

Share Document