Diabetes Detection Models in Mexican Patients by Combining Machine Learning Algorithms and Feature Selection Techniques for Clinical and Paraclinical Attributes: A Comparative Evaluation

Abstract

The development of medical diagnostic models to support healthcare professionals has witnessed remarkable growth in recent years. Among the prevalent health conditions affecting the global population, diabetes stands out as a significant concern. In the domain of diabetes diagnosis, machine learning algorithms have been widely explored for generating disease detection models, leveraging diverse datasets primarily derived from clinical studies. The performance of these models heavily relies on the selection of the classifier algorithm and the quality of the dataset. Therefore, optimizing the input data by selecting relevant features becomes essential for accurate classification. This research presents a comprehensive investigation into diabetes detection models by integrating two feature selection techniques: the Akaike information criterion and genetic algorithms. These techniques are combined with six prominent classifier algorithms, including support vector machine, random forest, k-nearest neighbor, gradient boosting, extra trees, and naive Bayes. By leveraging clinical and paraclinical features, the generated models are evaluated and compared to existing approaches. The results demonstrate superior performance, surpassing accuracies of 94%. Furthermore, the use of feature selection techniques allows for working with a reduced dataset. The significance of feature selection is underscored in this study, showcasing its pivotal role in enhancing the performance of diabetes detection models. By judiciously selecting relevant features, this approach contributes to the advancement of medical diagnostic capabilities and empowers healthcare professionals in making informed decisions regarding diabetes diagnosis and treatment.

Optimizing Clinical Diabetes Diagnosis through Generative Adversarial Networks: Evaluation and Validation

“…In addition to comparing our results with the findings of García et al [ 5 ], we also benchmarked our study against the work of Tanaka et al [ 11 ]. In comparison to Tanaka et al’s findings, our study demonstrated notable advancements in diabetes classification.…”

“…Table 4 presents the performance metrics of the Random Forests classification model, as recommended by García et al [ 5 ]. These metrics offer a comprehensive understanding of the effectiveness and reliability of the classifier when trained on the expanded dataset, thereby substantiating the potential advantages of utilizing GAN-generated synthetic data for improving the model’s predictive capabilities.…”

“…Table 5 displays the evaluation outcomes of the Random Forests classification model presented by García et al [ 5 ]. These outcomes serve as a benchmark for comparison with the results attained in the current research, where the classification model was assessed using the augmented dataset generated by GANs.…”

“…These synthetic data were used to assess the performance of a Random Forest (RF) classification model trained on both real and synthetic data. The evaluation included a comparison of the model’s performance with the outcomes reported by García et al [ 5 ] for the original dataset.…”

A genetic algorithm-based feature selection approach for diabetes prediction

“…K-Nearest neigbour (KNN) outperforms decision trees when used as a stacking combiner. A novel diabetes detection technique was presented by Domínguez et al [14] employing design of experiment, GA, and multi-layer perceptron (MLP). GA was  ISSN: 2252-8938 Int J Artif Intell, Vol.…”