The AI-Delirium Guard is an AI-based predictive model developed to identify the risk of postoperative delirium in individuals aged 65 and older undergoing anesthesia for surgical procedures. It aims to enhance postoperative care for older adults by predicting delirium risk, thus improving patient outcomes and healthcare strategies.
In this project, I played a key role in the development of the predictive model, data preprocessing, and the design and implementation of the user interface. I collaborated with a team of data scientists, healthcare professionals, and software developers to ensure the success of the project.
The project involved a comprehensive workflow, including data collection, exploration, cleansing, preprocessing, feature selection, model development, training, validation, testing, and deployment. Machine learning algorithms such as Neural Network, Logistic Regression, Stochastic Gradient Descent, Random Forest, Decision Tree, and Ridge Classifier were utilized to develop the predictive model.
The AI-Delirium Guard features an intuitive dashboard design with user-friendly forms for data entry, visualization tools for results display, and interactive elements for user engagement. Accessibility features, documentation, support, data security, and privacy measures were also incorporated into the interface.
The AI-Delirium Guard serves as a valuable tool for anesthesiologists, surgeons, and nurses to assess and mitigate the risk of postoperative delirium in older adults. It also assists in planning postoperative care and interventions tailored to individual risk profiles.
AI-Delirium Guard represents a significant advancement in predictive healthcare, leveraging AI to enhance postoperative care for older adults. The project underscores the power of AI in healthcare, the importance of thorough data processing, and the need for user-centered design in medical software.
Repository Link: AI-Delirium Guard on GitHub