Variational quantum classifier-based early identification and classification of chronic kidney disease using sparse...

Area of Science:

• Medical Informatics
• Quantum Computing
• Machine Learning

Background:

• Chronic kidney disease (CKD) is a leading cause of premature mortality, driven by hypertension and diabetes.
• Current diagnostic methods rely on glomerular filtration rate and inflammation biomarkers, often identifying disease at later stages.
• Prompt diagnosis is crucial to mitigate the severe impact and high mortality rate associated with CKD.

Purpose of the Study:

• To develop and evaluate a quantum machine learning (QML) based technique for the early diagnosis and prognosis of chronic kidney disease (CKD).
• To improve the accuracy and efficiency of CKD detection compared to traditional classification methods.

Main Methods:

• The research employed a four-phase approach: data pre-processing (Kalman filter, normalization), data augmentation (sparse autoencoders), feature selection (LASSO shrinkage), and classification (Variational Quantum classifiers).
• The proposed system was validated on the UCI dataset, containing 400 early-stage CKD patients with 25 attributes.

Main Results:

• The QML-based system demonstrated superior performance in classifying chronic kidney disease.
• Achieved a classification accuracy of 99.2%, surpassing traditional classifiers.
• Evaluation metrics included F1-score, precision, recall, and accuracy.

Conclusions:

• The proposed quantum machine learning model shows significant promise for accurate and early detection of chronic kidney disease.
• This QML approach offers a potential advancement in diagnosing and managing CKD, potentially reducing mortality rates.