Authors:
S. Sathiyapoobalan, K. Naveenkumar, A.T. Ashmi Christus, Bhopendra Singh

Addresses:
Department of Instrumentation and Control Engineering, Sri Manakula Vinayagar Engineering College, Madagadipet, Puducherry, India. Department of Electronics and Communication Engineering, Dhaanish Ahmed College of Engineering, Chennai, Tamil Nadu, India. Department of Engineering, Amity University Dubai, Dubai, United Arab Emirates.

Abstract:

For the sake of both public health and environmental protection, efficient handling of medical waste is necessary.  There is a lack of automation in the waste segregation process in the current hospital waste management systems. Instead, health staff manually sort garbage into color-coded containers specifically designed for that purpose. When it comes to classifying medical waste, this research study presents an innovative method that utilizes the SSD MobileNet object detection model.  With the help of the suggested model, medical waste may be classified into five distinct categories: infectious waste, non-infectious waste, sharps, and glassware. This model displays a high level of accuracy and efficiency. With an average precision and recall of approximately 90 percent, the model can reliably identify and localize medical waste. The implementation of this approach in healthcare institutions may enable the automatic separation of trash, improve disposal efficiency, and enhance safety. This research makes a substantial contribution to the management of medical waste, addressing a critical issue that impacts public health and establishing a framework for automating waste segregation in the healthcare sector.

Keywords: Waste Management; Deep Learning; Object Detection; Hazardous Waste and Image Classification; Health and Environmental; Medical Waste; Healthcare Facilities; Healthcare Sector.

Received on: 03/10/2024, Revised on: 07/12/2024, Accepted on: 14/01/2025, Published on: 12/06/2025

DOI: 10.69888/FTSESS.2025.000468

FMDB Transactions on Sustainable Environmental Sciences, 2025 Vol. 2 No. 2, Pages: 111-119

• Views : 34
• Downloads : 5

Download PDF