Prediction Model for Prevalence of Type-2 Diabetes Complications with ANN Approach Combining with K-Fold Cross Validation and K-Means Clustering

Munna, Tahsir Ahmed; Alam, Mirza Mohtashim; Allayear, Shaikh Muhammad; Sarker, Kaushik; Ara, Sheikh Joly Ferdaus

doi:10.1007/978-3-030-12388-8_71

Cited by 4 publications

(4 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We chose this algorithm because of its simplicity and ease of interpretation, which contributed to it being one of the most popular clustering algorithms in the medical domain 19. Before fitting the k -means algorithm, for categorical predictors we used one-hot encoding, that is, we converted each categorical predictor into a series of binary ones 20. We standardised the predictors before passing them on to k -means algorithm.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Predicting outcomes at the individual patient level: what is the best method?

et al. 2023

View full text Add to dashboard Cite

ObjectiveWhen developing prediction models, researchers commonly employ a single model which uses all the available data (end-to-endapproach). Alternatively, asimilarity-basedapproach has been previously proposed, in which patients with similar clinical characteristics are first grouped into clusters, then prediction models are developed within each cluster. The potential advantage of the similarity-based approach is that it may better address heterogeneity in patient characteristics. However, it remains unclear whether it improves the overall predictive performance. We illustrate the similarity-based approach using data from people with depression and empirically compare its performance with the end-to-end approach.MethodsWe used primary care data collected in general practices in the UK. Using 31 predefined baseline variables, we aimed to predict the severity of depressive symptoms, measured by Patient Health Questionnaire-9, 60 days after initiation of antidepressant treatment. Following the similarity-based approach, we usedk-means to cluster patients based on their baseline characteristics. We derived the optimal number of clusters using the Silhouette coefficient. We used ridge regression to build prediction models in both approaches. To compare the models’ performance, we calculated the mean absolute error (MAE) and the coefficient of determination (R2) using bootstrapping.ResultsWe analysed data from 16 384 patients. The end-to-end approach resulted in an MAE of 4.64 and R2of 0.20. The best-performing similarity-based model was for four clusters, with MAE of 4.65 and R2of 0.19.ConclusionsThe end-to-end and the similarity-based model yielded comparable performance. Due to its simplicity, the end-to-end approach can be favoured when using demographic and clinical data to build prediction models on pharmacological treatments for depression.

show abstract

Section: Methodsmentioning

confidence: 99%

“… 19 Before fitting the k -means algorithm, for categorical predictors we used one-hot encoding, that is, we converted each categorical predictor into a series of binary ones. 20 We standardised the predictors before passing them on to k -means algorithm.…”

Section: Methodsmentioning

confidence: 99%

Predicting outcomes at the individual patient level: what is the best method?

et al. 2023

View full text Add to dashboard Cite

show abstract

“…Several scientific researchers have worked with advanced methodologies for creating optimization and prediction by using machine learning algorithm, the results of studies are briefly discussed below. Md Tahsir Ahmed Munna, Mirza Mohtashim Alam used the Artificial neural network (ANN) and K-mean clustering methods, for improving statistical prediction accuracy of chronic diabetes mellitus based on extensive classification system, allow patients to know in advance about the prevalence of the disease [6]. Youshang Zhang, Qi Li utilized the regressive convulsive neural network (RCNN) and the reference support vector regressive method (SVR), considering the account the factors affecting electricity consumption, created a prediction model with a low error rate [7].…”

Section: Related Workmentioning

confidence: 99%

Research on predictive model based on classification with parameters of optimization

Gulzat¹,

Naizabayeva²,

Siladi³

et al. 2020

NNW

View full text Add to dashboard Cite

This paper effectively uses the data mining and optimization methods to investigate a classification based on decision trees algorithm, then optimizes by the method of grid search and cross-validation, which improves the prediction accuracy of the decision tree model for the PCs sales in practical application and solves insufficient training data, high computational cost, and low prediction accuracy. The main goal of the article is to predict PC sales using machine learning tools caused by various types of operating system factors in practical applications. This article proposes a combined predictive research model that fully reveals the benefits of optimization and neural networks, and also has a very accurate fit and forecasting accuracy. The proposed predictive model is implemented in the data science software platform RapidMiner. A decision tree model is executed, then the model's prediction capacity is evaluated and tested. Grid search optimizer is used to automatically build the final model using the best-optimized parameter for training the classifier. The paper combines grid the grid search and cross-validation to optimize the parameters of the decision tree to improve the classification prediction accuracy of the decision tree model. This article combines neural networks with optimization methods to establish a prediction model for laptop sales. This model gives full play to the advantages of optimization and neural networks and has very good fitting capabilities and prediction accuracy. Besides, the neural network for the prediction model has strong dynamic analysis capabilities. Once there are new observations, it can continue to be added to the modeling, which has high adaptability. The Neural Network algorithm has the highest accuracy of the predicted PC sales by evaluating the results of the five kinds of algorithms. The result for prediction accuracy shows the highest performance.

show abstract

“…Diagnosing COVID-19, chronic kidney disease, urinary infections, pulmonary hypertension, influenza, skin lesions, and acromegaly; predicting the risk of severe complications after bariatric surgery; 5 estimating the prevalence of long-term complications in patients with type 2 diabetes, 6 and determining the risk of major adverse events following intravenous lead extraction for cardiac rhythm management 7 are some examples of the application of ML in medicine. 8 , 9 Predicting fatal MI complications through prompt and critical preventive measures is crucial since competent experts are almost always unable to forecast all these problems.…”

Section: Introductionmentioning

confidence: 99%

Prediction of the Fatal Acute Complications of Myocardial Infarction via Machine Learning Algorithms

Ghafari,

Azar,

Ghafari

et al. 2024

JTHC

View full text Add to dashboard Cite

Background: Myocardial infarction (MI) is a major cause of death, particularly during the first year. The avoidance of potentially fatal outcomes requires expeditious preventative steps. Machine learning (ML) is a subfield of artificial intelligence science that detects the underlying patterns of available big data for modeling them. This study aimed to establish an ML model with numerous features to predict the fatal complications of MI during the first 72 hours of hospital admission. Methods: We applied an MI complications database that contains the demographic and clinical records of patients during the 3 days of admission based on 2 output classes: dead due to the known complications of MI and alive. We utilized the recursive feature elimination (RFE) method to apply feature selection. Thus, after applying this method, we reduced the number of features to 50. The performance of 4 common ML classifier algorithms, namely logistic regression, support vector machine, random forest, and extreme gradient boosting (XGBoost), was evaluated using 8 classification metrics (sensitivity, specificity, precision, false-positive rate, false-negative rate, accuracy, F1-score, and AUC). Results: In this study of 1699 patients with confirmed MI, 15.94% experienced fatal complications, and the rest remained alive. The XGBoost model achieved more desirable results based on the accuracy and F1-score metrics and distinguished patients with fatal complications from surviving ones (AUC=78.65%, sensitivity=94.35%, accuracy=91.47%, and F1-score=95.14%). Cardiogenic shock was the most significant feature influencing the prediction of the XGBoost algorithm. Conclusion: XGBoost algorithms can be a promising model for predicting fatal complications following MI.

show abstract

Prediction Model for Prevalence of Type-2 Diabetes Complications with ANN Approach Combining with K-Fold Cross Validation and K-Means Clustering

Cited by 4 publications

References 10 publications

Predicting outcomes at the individual patient level: what is the best method?

Predicting outcomes at the individual patient level: what is the best method?

Research on predictive model based on classification with parameters of optimization

Prediction of the Fatal Acute Complications of Myocardial Infarction via Machine Learning Algorithms

Contact Info

Product

Resources

About