CO Breathalyzer with Voice Response is the device to measured the level of CO residual in a person's lung who consume tobacco. Measuring residual CO in human breath can identify the tobacco addiction level instead of measuring nicotine in blood.
เพื่อทดแทนการนำเข้าจากต่างประเทศ
คณะวิศวกรรมศาสตร์
This project has been developed to address medical challenges related to the process of counting and classifying blood cells from samples, a task that requires both time and high precision. To reduce the workload of medical personnel, the developers have created a platform and an artificial intelligence (AI) system capable of automatically classifying and counting cells from sample images. This system is designed to assist medical laboratory technicians by enabling them to work more efficiently and accurately, reducing the time required for analysis. Furthermore, it promotes the advancement of medical technology, ensuring effective usability from classrooms and laboratories to hospitals.
วิทยาเขตชุมพรเขตรอุดมศักดิ์
This project aims to design and develop an eye-tracking system to facilitate communication for paralyzed immobile patients. The system is designed to enable patients to convey their needs to caregivers or family members by detecting and tracking eye movements using the Tobii Eye Tracker 5 device. This approach serves as an alternative communication method, replacing the physical movement or speech of paralyzed patients. The system effectively detects and tracks eye movements at a distance of 55 to 85 centimeters and is designed for installation on a computer to ensure ease of use. The program interface consists of three main sections: (1) a set of emotions, (2) a set of needs, and (3) a set of additional needs. It supports input from a virtual keyboard in both Thai and English and allows users to specify additional needs through eye-tracking-enabled typing. Furthermore, the system can generate synthetic speech for text that is difficult to pronounce aloud, send notification messages via the Line application, and store usage data in a database presented in a dashboard format. System testing revealed that the optimal detection distance ranges from 65 to 75 centimeters, as this range yields an error rate of no more than 1 percent. The system accurately responds to eye movements for communication through sound within 3 seconds when interacting with various function buttons. This eye-tracking system effectively enables paralyzed immobile patients to communicate their emotions and needs, facilitating better understanding and interaction between patients and their caregivers or family members.
คณะแพทยศาสตร์
This study explores the application of deep convolutional neural networks (CNNs) for accurate pill identification, addressing the limitations of traditional human-based methods. Using a dataset of 1,250 images across 10 household remedy drugs, various CNN architectures, including YOLO models, were tested under different conditions. Results showed that natural lighting was optimal for imprinted pills, while a lightbox improved detection for plain pills. The YOLOv5-tiny model demonstrated the best detection accuracy, and efficientNet_b0 achieved the highest classification performance. While the model showed strong results, its generalization is limited by sample size and drug variability. Nonetheless, this approach holds promise for enhancing medication safety and reducing errors in outpatient care.