International Research Journal of Engineering and Technology (IRJET) Volume: 09 Issue: 11 | Nov 2022
www.irjet.net
e-ISSN: 2395-0056 p-ISSN: 2395-0072
Drug Pill Recognition System Using Deep Learning Sudarshan Borude1, Sakshi Patil2, Roshni Bhoirkar3, Isheeta Shirsath4 1Mr.
Sudarshan S. Borude, Dept. of Information Technology, All India Shri Shivaji Memorial Society Institute Of Information Technology, Pune, Maharashtra, India 2Miss. Sakshi D. Patil, Dept. of Information Technology, All India Shri Shivaji Memorial Society Institute Of Information Technology, Pune, Maharashtra, India 3Miss. Roshni C. Bhoirkar, Dept. of Information Technology, All India Shri Shivaji Memorial Society Institute Of Information Technology, Pune, Maharashtra, India 4Miss. Isheeta M. Shirsath, Dept. of Information Technology, All India Shri Shivaji Memorial Society Institute Of Information Technology, Pune, Maharashtra, India ---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - — Loss of abilities like vision or memory are
the elderly. It uses computer vision techniques to assist elderly people with the task of pill identification with the goal of boosting confidence and autonomy.
typical characteristics of ageing, which is a natural process. These changes make it harder for senior adults to perform daily tasks, which can occasionally put them in danger. One of the most pertinent is connected to incorrect medication consumption or even forgetfulness. The health and life of the elderly are seriously endangered by mistakes of this nature. Furthermore, the technological solutions that are currently being considered for this issue are created for professionals or the general public without taking the needs of the elderly into consideration. Therefore, in order to address this lack of support, an image processing tool that serves as the foundation for a larger set of aging-friendly tools currently under development will be presented. The steps taken by this proposal include acquiring images and classifying pills according to their size, shape, and color. These features enable the system to describe and save information about pills on a local database during the learning phase. Later, during the recognition phase, the same features are chosen and compared to a database in order to give the user pertinent data regarding the pill being recognized.
Elderly people who are blind or visually impaired tend to take the wrong medications or forget to take them. The study also mentioned the topic. As a result, it is anticipated that patients who use drugs incorrectly and who are visually impaired will suffer significant medical losses. However, these patients may not have access to adequate support in this area. We take into account the requirement for a method of medication pill recognition for visually impaired chronic patients in order to solve this issue. The suggested system can help ensure that chronic patients with visual impairments take their medications safely. The review for visually impaired people's ability to identify drug pills is stated in the paper. Introduces the topic and provides an overview in section I. In section II, a literature review of a select few papers and datasets created for recognition are taken into account, and in section III, a proposed system for the system is presented.
Key Words: Health care, visually impaired, drug pill
2. LITERATURE REVIEW
recognition, image processing, CNN.
By various researchers, various approaches have been presented. Here, we've discussed a few of the techniques. Many related tools have been created and evaluated to provide related functionalities (such as drug pill recognition and medication reminders) to aid in safe medication use.
1. INTRODUCTION Aging also results in the loss of abilities like vision and memory, which puts elderly people in perilous situations. Medication is one that is frequently mentioned. Elderly people frequently forget to take their medications properly or fail to do so, which can have a serious negative impact on their health. However, when they become aware of the circumstance, their self-assurance is undermined, necessitating the need for support to change the situation. The disparity between taxpayers and beneficiaries will continue to widen annually given the situation in Europe. The fact that this support comes from the healthcare system is extremely unavailability. As a result, a different approach based on current technological developments, namely mobile devices, is suggested for this system. This solution is a part of a larger, still-under-development tool support for
© 2022, IRJET
|
Impact Factor value: 7.529
(1). D. Ushizima, A. Carneiro, M. Souza, and F. Medeiros. “Investigating pill recognition methods for a new national library of medicine image dataset" [1] Correct pill identification has grown to be essential to patient care and safety. This paper examines descriptors for pill detection and characterization using the recently released National Library of Medicine (NLM) pill image database.[1] The authors discuss their work on developing algorithms to
|
ISO 9001:2008 Certified Journal
|
Page 706