Author(s): Rishi Kesaraju
Wound care is a crucial element of healthcare management, particularly for patients with chronic conditions like diabetic ulcers, pressure sores, or surgical wounds. However, the manual assessment of wound healing is resource-intensive, error-prone, and often impractical in remote or underserved regions. This research introduces an Artificial Intelligence (AI)-powered mobile application designed to address these challenges by automating the assessment of wound healing stages and providing real-time, personalized recovery progress. The primary objectives are to classify wounds based on their healing stage and predict recoverytimelines using sequential images, thereby improving accessibility and decision-making in wound management. The study utilizes Convolutional Neural Networks (CNNs) for image classification and Long Short-Term Memory (LSTM) networks for healing trajectory prediction. Using a curated dataset of 2,500 annotated wound images, the proposed models achieve a classification accuracy of 92% and a healing progression prediction error of less than 5%. The results indicate that the mobile application can provide scalable, efficient, and accessible wound care solutions. Future work includes expanding the dataset to enhance model generalization and integrating wearable sensor data for continuous monitoring. This study highlights the potential of AI-driven applications to revolutionize healthcare by bridging the gap in medical resources and providing patients with actionable insights in real-time.