Research on the Risk Warning Model and Prevention Strategies for Acute Kidney Injury Associated With Cyclosporine Based on Explainable Deep Neural Networks and Therapeutic Drug Monitoring

Research on the Risk Warning Model and Prevention Strategies for Acute Kidney Injury Associated With Cyclosporine Based on Explainable Deep Neural Networks and Therapeutic Drug Monitoring

In this study, the investigators will focus on hospitalized patients using cyclosporine and develop an acute kidney injury risk prediction model through in-depth analysis of electronic medical record data, employing interpretable deep learning methods. This model aims to provide timely decision-making support for clinicians regarding prevention and treatment. Compared to traditional machine learning models, deep neural network models can extract deeper features from complex medical data and perform more precise pattern recognition, thereby improving the accuracy and reliability of predictions. By developing a prediction tool based on interpretable deep learning models, the investigators will be able to better assess the association between the use of CNI-class immunosuppressants and acute kidney injury, explore targeted prevention strategies, and offer more accurate prediction and intervention guidance for clinicians. Additionally, this study has significant socioeconomic benefits and promising prospects for application and promotion.

Inclusion Criteria:

1. During hospitalization, tacrolimus or cyclosporine was used, and therapeutic drug monitoring was conducted according to standard procedures.
2. Aged 18 years or older at the time of admission.
3. Length of hospital stay > 48 hours.
4. At least 2 serum creatinine tests were conducted during hospitalization.

Exclusion Criteria:

1. Chronic kidney disease stage 5 was achieved before admission.
2. Incomplete clinical data.
3. Serum creatinine levels were consistently below 40 mmol/L during hospitalization.