scholarly journals Estimation of the Ambit of Breast Cancer with a Modified Resnet Analysis Using Machine Learning Approach

2021 ◽  
Vol 18 ◽  
pp. 183-190
Author(s):  
C. K. Narayanappa ◽  
G. R., Poornima ◽  
Basavaraj V. Hiremath
Keyword(s):  
Breast Cancer ◽  
Supervised Classification ◽  
Area Under The Curve ◽  
Breast Cancer Diagnosis ◽  
Learning Approach ◽  
Classification Models ◽  
Mortality And Morbidity ◽  
Machine Learning Approach ◽  
Sensitivity Specificity ◽  
Cancerous Cells

Breast Cancer has been one of the most common reasons for mortality and morbidity among the females around the world especially in developing countries. In this regard, Mammography is a popular screening technique for breast cancer diagnosis so as to label the existence of cancerous cells. The present work encompasses the design and development of a M-ResNet (Modified ResNet) approach so as to classify the breast cancer into benign and malignant conditions with the inclusions for supervised classification models with the training of both upper as well as the lower layers of the designed networks. The efficacy of the developed approach was evaluated using various performance evaluators such as those of sensitivity, specificity, accuracy and F1-Score. Bi-Rads score was used as a basis for the classification process wherein a score of 0-3 correlated to benign and it is non-cancerous nature of tissues whereas malignancy was denoted by a score of 4 and above. InBreast dataset, a publicly available online dataset with 112 breast images were used for the evaluation of the developed paradigm. The present paradigm portrayed an accuracy of 96.43% with Area Under the Curve (AUC) of 95.63%.

scholarly journals A machine learning approach to predict healthcare cost of breast cancer patients

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Pratyusha Rakshit ◽  
Onintze Zaballa ◽  
Aritz Pérez ◽  
Elisa Gómez-Inhiesto ◽  
Maria T. Acaiturri-Ayesta ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Machine Learning ◽  
Cancer Patients ◽  
Healthcare Cost ◽  
Percentage Error ◽  
Learning Approach ◽  
Early Prediction ◽  
Breast Cancer Patients ◽  
Machine Learning Approach ◽  
Clinical Records

AbstractThis paper presents a novel machine learning approach to perform an early prediction of the healthcare cost of breast cancer patients. The learning phase of our prediction method considers the following two steps: (1) in the first step, the patients are clustered taking into account the sequences of actions undergoing similar clinical activities and ensuring similar healthcare costs, and (2) a Markov chain is then learned for each group to describe the action-sequences of the patients in the cluster. A two step procedure is undertaken in the prediction phase: (1) first, the healthcare cost of a new patient’s treatment is estimated based on the average healthcare cost of its k-nearest neighbors in each group, and (2) finally, an aggregate measure of the healthcare cost estimated by each group is used as the final predicted cost. Experiments undertaken reveal a mean absolute percentage error as small as 6%, even when half of the clinical records of a patient is available, substantiating the early prediction capability of the proposed method. Comparative analysis substantiates the superiority of the proposed algorithm over the state-of-the-art techniques.

scholarly journals A Machine Learning Approach to Study Glycosidase Activities from Bifidobacterium

2021 ◽  
Vol 9 (5) ◽  
pp. 1034
Author(s):  
Carlos Sabater ◽  
Lorena Ruiz ◽  
Abelardo Margolles
Keyword(s):  
Machine Learning ◽  
Supervised Classification ◽  
Machine Learning Algorithms ◽  
Learning Approach ◽  
Human Milk Oligosaccharides ◽  
Future Studies ◽  
High Fiber ◽  
Machine Learning Approach ◽  
Prebiotic Oligosaccharides

This study aimed to recover metagenome-assembled genomes (MAGs) from human fecal samples to characterize the glycosidase profiles of Bifidobacterium species exposed to different prebiotic oligosaccharides (galacto-oligosaccharides, fructo-oligosaccharides and human milk oligosaccharides, HMOs) as well as high-fiber diets. A total of 1806 MAGs were recovered from 487 infant and adult metagenomes. Unsupervised and supervised classification of glycosidases codified in MAGs using machine-learning algorithms allowed establishing characteristic hydrolytic profiles for B. adolescentis, B. bifidum, B. breve, B. longum and B. pseudocatenulatum, yielding classification rates above 90%. Glycosidase families GH5 44, GH32, and GH110 were characteristic of B. bifidum. The presence or absence of GH1, GH2, GH5 and GH20 was characteristic of B. adolescentis, B. breve and B. pseudocatenulatum, while families GH1 and GH30 were relevant in MAGs from B. longum. These characteristic profiles allowed discriminating bifidobacteria regardless of prebiotic exposure. Correlation analysis of glycosidase activities suggests strong associations between glycosidase families comprising HMOs-degrading enzymes, which are often found in MAGs from the same species. Mathematical models here proposed may contribute to a better understanding of the carbohydrate metabolism of some common bifidobacteria species and could be extrapolated to other microorganisms of interest in future studies.

scholarly journals Identification of differentially methylated genes as diagnostic and prognostic biomarkers of breast cancer

2021 ◽  
Vol 19 (1) ◽  
Author(s):  
Xiao-hong Mao ◽  
Qiang Ye ◽  
Guo-bing Zhang ◽  
Jin-ying Jiang ◽  
Hong-ying Zhao ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Cancer Diagnosis ◽  
High Sensitivity ◽  
Breast Cancer Diagnosis ◽  
Diagnostic Value ◽  
Methylation Pattern ◽  
Clinical Value ◽  
Roc Curve Analysis ◽  
Differentially Methylated Genes ◽  
Sensitivity Specificity

Abstract Background Aberrant DNA methylation is significantly associated with breast cancer. Methods In this study, we aimed to determine novel methylation biomarkers using a bioinformatics analysis approach that could have clinical value for breast cancer diagnosis and prognosis. Firstly, differentially methylated DNA patterns were detected in breast cancer samples by comparing publicly available datasets (GSE72245 and GSE88883). Methylation levels in 7 selected methylation biomarkers were also estimated using the online tool UALCAN. Next, we evaluated the diagnostic value of these selected biomarkers in two independent cohorts, as well as in two mixed cohorts, through ROC curve analysis. Finally, prognostic value of the selected methylation biomarkers was evaluated breast cancer by the Kaplan-Meier plot analysis. Results In this study, a total of 23 significant differentially methylated sites, corresponding to 9 different genes, were identified in breast cancer datasets. Among the 9 identified genes, ADCY4, CPXM1, DNM3, GNG4, MAST1, mir129-2, PRDM14, and ZNF177 were hypermethylated. Importantly, individual value of each selected methylation gene was greater than 0.9, whereas predictive value for all genes combined was 0.9998. We also found the AUC for the combined signature of 7 genes (ADCY4, CPXM1, DNM3, GNG4, MAST1, PRDM14, ZNF177) was 0.9998 [95% CI 0.9994–1], and the AUC for the combined signature of 3 genes (MAST1, PRDM14, and ZNF177) was 0.9991 [95% CI 0.9976–1]. Results from additional validation analyses showed that MAST1, PRDM14, and ZNF177 had high sensitivity, specificity, and accuracy for breast cancer diagnosis. Lastly, patient survival analysis revealed that high expression of ADCY4, CPXM1, DNM3, PRDM14, PRKCB, and ZNF177 were significantly associated with better overall survival. Conclusions Methylation pattern of MAST1, PRDM14, and ZNF177 may represent new diagnostic biomarkers for breast cancer, while methylation of ADCY4, CPXM1, DNM3, PRDM14, PRKCB, and ZNF177 may hold prognostic potential for breast cancer.

Sign in / Sign up

Export Citation Format

Share Document