Abstract:This study proposed a novel technique for early diabetes prediction with high accuracy. Recently, Deep Learning (DL) has been proven to be expeditious in the diagnosis of diabetes. The supported model is constructed by implementing ten hidden layers and a multitude of epochs using the Deep Neural Network (DNN)-based multi-layer perceptron (MLP) algorithm. We proceeded to meticulously fine-tune the hyperparameters within the fully automated DL architecture to optimize data preprocessing, prediction, and classification using a novel dataset of Mansoura University Children's Hospital Diabetes (MUCHD), which allowed for a comprehensive evaluation of the system's performance. The system was validated and tested using a sample of 548 patients, each with 18 significant features. Various validation metrics were employed to ensure the reliability of the results using cross-validation approaches with various statistical measures of accuracy, F-score, precision, sensitivity, specificity, and Dice similarity coefficient. The high performance of the proposed system can help clinicians accurately diagnose diabetes, with a remarkable accuracy rate of 99.8%. According to our analysis, implementing this method results in a noteworthy increase of 0.39% in the overall system performance compared to the current state-of-the-art methods. Therefore, we recommend using this method to predict diabetes.

What problem does this paper attempt to address?

### Problems the Paper Aims to Solve This paper aims to propose a new technique for early high-precision prediction of childhood diabetes. Specifically, the research team utilized Deep Learning (DL) technology to construct a Multi-Layer Perceptron (MLP) model, implemented through a Deep Neural Network (DNN), and meticulously tuned the hyperparameters. #### Main Objectives 1. **Improve Diagnostic Accuracy**: Enhance the accuracy of diabetes prediction using deep learning technology. 2. **Dataset Application**: Use data from the Mansoura University Children's Hospital Diabetes (MUCHD) dataset for model training and validation. 3. **Performance Evaluation**: Ensure the reliability of results through cross-validation methods and various statistical metrics (such as accuracy, F-score, precision, sensitivity, specificity, and Dice similarity coefficient). 4. **Clinical Application Value**: The proposed system can help clinicians accurately diagnose diabetes, with a significant accuracy of 99.8%. #### Research Background Diabetes is a chronic disease characterized by high levels of glucose in the blood. If diabetes is not treated in time, it can lead to various complications such as vision impairment, cardiovascular diseases, dental diseases, stroke, and microvascular complications. Therefore, early detection of diabetes is crucial for timely treatment and prevention of disease progression. #### Methods and Innovations - **Model Structure**: Constructed a DNN model with 10 hidden layers. - **Feature Selection and Preprocessing**: Preprocessed the data, including handling missing values and feature scaling. - **Hyperparameter Optimization**: Meticulously tuned the hyperparameters of the model to optimize the data preprocessing, prediction, and classification process. - **Experimental Validation**: Validated using a sample dataset of 548 patients, each containing 18 important features. #### Summary The method proposed in this paper improves the overall system performance by 0.39% compared to the existing state-of-the-art methods, indicating its significant advantage in diabetes prediction. Therefore, it is recommended to use this method for diabetes prediction.

Pediatric diabetes prediction using deep learning

A Deep Learning Approach to Diabetes Diagnosis

Diabetes prediction model based on an enhanced deep neural network

A Proposed Technique Using Machine Learning for the Prediction of Diabetes Disease through a Mobile App

A novel hybrid deep learning model for early stage diabetes risk prediction

Hi-Le and HiTCLe: Ensemble Learning Approaches for Early Diabetes Detection Using Deep Learning and Explainable Artificial Intelligence

DiabetesNet: A Deep Learning Approach to Diabetes Diagnosis

A novel method for prediction of diabetes mellitus using deep convolutional neural network and long short-term memory

A Machine Learning Approach for Prediction of Diabetes Mellitus

Electronic Health Records-Based Data-Driven Diabetes Knowledge Unveiling and Risk Prognosis

Prediction of Diabetes using Machine Learning

Disease Prediction Using a Modified Multi-Layer Perceptron Algorithm in Diabetes

A novel ensemble machine learning framework for early stage diabetes mellitus prediction

Machine learning-based prediction of diabetic patients using blood routine data

An automated approach to predict diabetic patients using KNN imputation and effective data mining techniques

Diabetes prediction using machine learning and explainable AI techniques

A deep learning approach to diabetic blood glucose prediction

A novel solution of deep learning for enhanced support vector machine for predicting the onset of type 2 diabetes

Toward reliable diabetes prediction: Innovations in data engineering and machine learning applications

Predicting Diabetes Mellitus With Machine Learning Techniques.

Early Stage Diabetes Prediction via Extreme Learning Machine