Supervised Machine Learning Empowered Multifactorial Genetic Inheritance Disorder Prediction.

Area of Science:

• Computational biology and bioinformatics
• Medical informatics
• Machine learning applications in healthcare

Background:

• Fatal diseases such as cancer, dementia, and diabetes pose significant global health risks.
• Early diagnosis is crucial for effective treatment and reducing mortality rates.
• Advancements in computer science and machine learning offer new avenues for disease prediction.

Purpose of the Study:

• To develop and compare machine learning classifiers for the early prediction of cancer, dementia, and diabetes.
• To utilize a multifactorial genetic inheritance disorder dataset for multiclass disease prediction.
• To evaluate the efficacy of Support Vector Machine (SVM) and K-nearest neighbor (KNN) algorithms.

Main Methods:

• Employed multiclass classification using Support Vector Machine (SVM) and K-nearest neighbor (KNN) algorithms.
• Utilized a multifactorial genetic inheritance disorder dataset for training and testing the models.
• Compared the predictive accuracy of SVM and KNN for the three target diseases.

Main Results:

• The proposed SVM model achieved 92.8% accuracy during training and 92.5% during testing.
• The proposed KNN model achieved 92.8% accuracy during training and 91.2% during testing.
• The SVM-based model demonstrated slightly superior performance compared to the KNN model for predicting these diseases.

Conclusions:

• The proposed machine learning models, particularly SVM, show significant potential for the early prediction and prognosis of cancer, dementia, and diabetes.
• Accurate early prediction can play a vital role in minimizing global death ratios.
• This approach offers a valuable tool for proactive healthcare management of complex genetic disorders.