Prediction Model for Prevalence of Type-2 Diabetes Mellitus Complications Using Machine Learning Approach

Younus, Muhammad Faraz; Munna, Tahsir Ahmed; Alam, Mirza Mohtashim; Allayear, Shaikh Muhammad; Ara, Sheikh Joly Ferdous

doi:10.1007/978-3-030-32587-9_7

Cited by 8 publications

(4 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [ 24 ], the authors focused on developing an ensemble model that utilized a majority voting technique to combine unweighted predictions from various machine learning models. On the other hand, the authors of [ 25 ] focused on discriminating techniques, where they combined the basic classifiers through a process that could adjust to the input observations and output requirements of each individual learning system.In the training set, the type of combination employed could be optimized by assigning weight to each classifier in order to improve the combined performance. In a study by the authors of [ 26 ], ensemble-based methods were recommended as the most effective approach for data stream classification problems.…”

Section: Related Workmentioning

confidence: 99%

Deep Learning-Based Classification and Feature Extraction for Predicting Pathogenesis of Foot Ulcers in Patients with Diabetes

Preiya

Kumar

2023

Diagnostics

View full text Add to dashboard Cite

The World Health Organization (WHO) has identified that diabetes mellitus (DM) is one of the most prevalent disease worldwide. Individuals with DM have a higher risk of mortality, and it is crucial to prioritize the treatment of foot ulcers, which is a significant complication associated with the disease, as they lead to the development of plantar ulcers, which results in the need to amputate part of the foot or leg. People with diabetes are at risk of experiencing various complications, such as heart disease, eye problems, kidney dysfunction, nerve damage, skin issues, foot ulcers, and dental diseases. Unawareness of the risk associated with diabetic foot ulcers (DFU) is a significant contributing factor to the mortality of diabetic patients. Evolving technological advancements such as deep learning techniques can be used to predict the symptoms of diabetic foot ulcers as early as possible, which helps to provide effective treatment to DM patients. This research introduces a methodology for analyzing images of foot ulcers in diabetic patients, focusing on feature extraction and classification. The dataset used in this study was collected from historical medical records and foot images of patients with diabetes, who commonly experience foot ulcers as a major complication. The dataset was pre-processed and segmented, and features were extracted using a deep recurrent neural network (DRNN). Image and numerical/text data were extracted separately, and the normal and abnormal diabetes ranges were identified. Foot images of patients with abnormal diabetes ranges were separated and classified using a pre-trained fast convolutional neural network (PFCNN) with U++net. The classification procedure involves the analysis of foot ulcers to predict their pathogenesis. To assess the effectiveness of the proposed technique, the study presented simulation results, including a confusion matrix and receiver operating characteristic curve. These results specifically focused on predicting two classes: normal and abnormal diabetes foot ulcerations. The analysis yielded various parameters, including accuracy, precision, recall curve, and area under the curve. The main goal of the study was to introduce an novel technique for assessing the risk of foot ulceration development in patients with diabetes, leveraging the analysis of foot ulcer images. The researchers collected a dataset of foot images and medical data from historical records of patients with diabetes and pre-processed and segmented the data. They then used a deep recurrent neural network to extract features from the segmented data and identified normal and abnormal diabetes ranges based on numerical and text data. Foot images of patients with abnormal diabetes ranges were classified using a pre-trained fast convolutional neural network with U++net to examine foot ulcers and forecast the development of the risk of diabetic foot ulcers (DFU). The study assessed the accuracy of the proposed technique as 99.32% by simulating results for feature extraction and the classification of diabetic foot ulcers. A comparison was made between this proposed technique and existing approaches.

show abstract

Section: Related Workmentioning

confidence: 99%

Deep Learning-Based Classification and Feature Extraction for Predicting Pathogenesis of Foot Ulcers in Patients with Diabetes

Preiya

Kumar

2023

Diagnostics

View full text Add to dashboard Cite

show abstract

“…They considered diabetes related attributes and their prediction by support vector machine, K nearest neighbor, decision tree, logistic regression, and discernment analyses find 75% accuracy by SVM [3]. Younus et al [2020], focused on the type-2 diabetes. Authors identified the prevalence of diabetes with type-2 diabetes mellitus.…”

Section: Related Workmentioning

confidence: 99%

“…Authors identified the prevalence of diabetes with type-2 diabetes mellitus. They used machine learning algorithm, decision tree, and random forest finally find complications in type-2 diabetes [4]. Shuja et al [2020], discussed about metabolic disorderly diabetes.…”

Section: Related Workmentioning

confidence: 99%

An Experimental Study of Diversity of Diabetes Disease Features by Bagging and Boosting Ensemble Method with Rule Based Machine Learning Classifier Algorithms

Yadav

Pal

2021

SN COMPUT. SCI.

View full text Add to dashboard Cite

Energy produces by blood sugar in human body. Diabetes mellitus is a condition in which human body cannot manage energy. Blood glucose, Insulin etc., and their functions become unmanaged in the glucose energy system in blood. This unbalanced system generates many dangerous diseases as like blood pressure, diabetes etc., in body. Many different recourses of energy are available in the nature. In this paper we present dataset with their pattern by box whisker plot, histograms and extract nine best features by Chi-Square and plot correlation matrix for correlation of each feature with their heat_map. We have applied rule-based classification algorithms as Decision table, OneR and JRIP on prepared dataset. In proposed work we have managed these three selected algorithms by Bagging and Boosting ensemble methods. These ensemble methods calculate Classification Accuracy, Precision, Recall and F1-Score for diabetes UCI dataset. The summary with the results, finally we concluded that the highest accuracy (98%) of Bagging Ensemble Method and the accuracy, precision, recall and f1-score of the Decision Table, OneR, Jrip and Boosting algorithms were carrying less.

show abstract

“…During text classification, such as medical data with machine learning or deep learning techniques, employing terms such as features using vector space representation can cause a high dimensionality of feature space and sparsity [1,2]. This kind of condition introduces some issues, including high computation cost in data analysis and reducing the classification accuracy performance [3].…”

Section: Introductionmentioning

confidence: 99%

An Improved Homogeneous Ensemble Technique for Early Accurate Detection of Type 2 Diabetes Mellitus (T2DM)

Faustin

Zou

2022

Computation

View full text Add to dashboard Cite

The objective of the present study is to improve the genetic algorithm (GA) supremacy in selecting the most suitable and relevant features within a highly dimensional dataset. This results in cost reduction and improving classification performance. During text classification, employing terms such as features using vector space representation can result in a high dimensionality of future space. This condition presents some issues, including high computation cost in data analysis and deteriorating classification accuracy performance. Several computational feature selection techniques can be applied in eliminating the least significant features within a dataset, including a genetic algorithm. The present study improved the performance of the classifier in classifying Pima Indian diabetes data. Despite the popularity of GA in the feature selection area, it does not provide the most optimal features due to one of its underlying issues: premature convergence due to insufficient population diversity in the future generations. GA was improved in its crossover operator using two steps: define a variable slice point on the size of the gene to be interchanged for every offspring generation and apply feature frequency scores in deciding the interchanging of genes. The above obtained results to the proposed technique will be better results than the results for standard GA. Our proposed algorithm attained an accuracy of 97.5%, precision of 98, recall of 97% and F1-score of 97%.

show abstract

Prediction Model for Prevalence of Type-2 Diabetes Mellitus Complications Using Machine Learning Approach

Cited by 8 publications

References 10 publications

Deep Learning-Based Classification and Feature Extraction for Predicting Pathogenesis of Foot Ulcers in Patients with Diabetes

Deep Learning-Based Classification and Feature Extraction for Predicting Pathogenesis of Foot Ulcers in Patients with Diabetes

An Experimental Study of Diversity of Diabetes Disease Features by Bagging and Boosting Ensemble Method with Rule Based Machine Learning Classifier Algorithms

An Improved Homogeneous Ensemble Technique for Early Accurate Detection of Type 2 Diabetes Mellitus (T2DM)

Contact Info

Product

Resources

About