KMITL

Permanent URI for this communityhttps://dspace.kmitl.ac.th/handle/123456789/1

Browse

Search Results

Now showing 1 - 6 of 6

Some of the metrics are blocked by your
consent settings
Performance validation of deep-learning-based approach in stool examination
(2025-12-01)
Corpuz, Kristal Dale Felimon
;
Kusolsuk, Teera
;
Wongphan, Benjamaporn
;
Chonsawat, Putza
;
Naing, Kaung Myat
Background: Human intestinal parasitic infections (IPI) pose a significant global health issue caused by parasitic helminths and protozoa, affecting around 3.5 billion people worldwide, with more than 200,000 deaths annually. Despite advancements in molecular methods with higher sensitivity and specificity, the Kato-Katz or formalin-ethyl acetate centrifugation technique (FECT) remains the gold standard and a routine diagnostic procedure suitable for its simplicity and cost-effectiveness. However, these techniques have limitations that must be addressed. Thus, this study evaluated the performance of a deep-learning-based approach for intestinal parasite identification and compared it with that of human experts. Methods: Human experts performed FECT and Merthiolate-iodine-formalin (MIF) techniques to serve as ground truth and reference for parasite species. Subsequently, a modified direct smear was conducted to gather images for the training (80%) and testing (20%) datasets. State-of-the-art models, including YOLOv4-tiny, YOLOv7-tiny, YOLOv8-m, ResNet-50, and DINOv2 (base, small, and large), were employed and were operated using in-house CIRA CORE platform. Overall performance was evaluated using confusion matrices, the metrics of which were calculated on the basis of the one-versus-rest and micro-averaging approaches. Moreover, the receiver operating characteristic (ROC) and precision-recall (PR) curves were determined for visual comparison. Lastly, Cohen’s Kappa and Bland–Altman analyses were used to statistically measure the significant differences and visualize the association levels between the human experts and the deep learning models’ classification performance in intestinal parasite identification. Results: Findings demonstrated the potential of a deep-learning-based approach, particularly of models DINOv2-large (accuracy: 98.93%; precision: 84.52%; sensitivity: 78.00%; specificity: 99.57%; F1 score: 81.13%; AUROC: 0.97) and YOLOv8-m (accuracy: 97.59%; precision: 62.02%; sensitivity: 46.78%; specificity: 99.13%; F1 score: 53.33%; AUROC: 0.755; AUPR: 0.556) for their high metric values in intestinal parasite identification. Class-wise prediction showed high precision, sensitivity, and F1 scores for helminthic eggs and larvae due to more distinct morphology. Moreover, all models obtained a > 0.90 k score, which indicates a strong level of agreement compared with the medical technologists. The Bland–Altman analysis also presented the best agreement between FECT performed by medical technologist A and YOLOv4-tiny, while the MIF technique performed by medical technologist B and DINOv2-small demonstrated the best bias-free agreement, with mean differences of 0.0199 and −0.0080, and standard deviation differences of 0.6012 and 0.5588, respectively. Conclusions: The results highlight the potential of integrating a deep-learning-based approach into parasite identification. The models showcased superiority in automated detection, suggesting a significant leap toward improving diagnostic procedures for IPI. This hybridization could enhance early detection and diagnosis, facilitating timely and targeted interventions to reduce the burden of IPI through more effective management and prevention strategies.
Some of the metrics are blocked by your
consent settings
Wireless Automated Exhibition System: Convenient Mobility at Low Cost
(2025-01-01)
Hirunwijitporn, Nirumol
;
Jongchanachavawat, Wirote
;
Bunyarittikit, Suphat
;
Mingmuang, Noppon
;
Roopkom, Ittipat
The application of wireless automatic exhibition system for event promotion facilitates the organization of exhibition-related activities by enabling automated control of exhibition displays. This system eliminates the need for allocating personnel to monitor the exhibition, thus reducing the cost of hiring exhibition attendants. From this research, it is possible to create a low-cost wireless automatic exhibition display system that is convenient for use in various events and exhibitions, especially when there are numerous booths involved. The system utilizes infrared sensors to detect visitors within a range of 90 cm, enabling accurate visitor counting. Once the desired number of visitors is reached, the exhibition will be displayed immediately. Additionally, the system can detect exiting visitors while the exhibition is ongoing. When the specified visitor count is met, the exhibition display will stop instantly. Due to these features, this wireless automatic exhibition system is affordable, easily portable, and suitable for various events and exhibitions. It is user-friendly and convenient for deployment.
Some of the metrics are blocked by your
consent settings
Optimum airflow to reduce particle contamination inside welding automation machine of hard disk drive production line
(2015-03-01)
Thongsri, Jatuporn
;
Pimsarn, Monsak
Welding automation machine (WAM), used for welding minute components to the head gimbal assembly (HGA) of a hard disk drive (HDD), needs to operate in a strictly clean environment. In today’s HDD factories, to prevent airborne particle contamination to the WAM, Fan Filter Units (FFUs) are installed on top of it to supply clean air and blow away outside airborne micro particles, keeping the microenvironment clean. Furthermore, the mass of the clean air should also carry away harmful particles generated inside the microenvironment. In this research, numerical simulation of airflow inside a WAM was performed in order to verify these cleaning functions of the airflow. A transition shear stress transport turbulence model was employed to simulate airflow from the FFUs through and out of the microenvironment. The simulation results showed that the airflow from the FFUs truly performs the two cleaning functions as intended. Moreover, they also revealed that the optimum air speed, the speed resulting in the lowest particle counts, is in the range of 0.35–0.55 m/s. Our findings can be useful for developers who may use FFUs to reduce particle counts in the environment of other types of industrial machinery.
Some of the metrics are blocked by your
consent settings
Automation system of continuous evaporating crystallization for cane sugar mills
(2014-02-04)
Puaughom, Charoenchai
;
Pankarnchanato, Phichet
;
Limsutthiphong, Pornchai
;
Julsereewong, Prasit
This article presents an automation system of continuous evaporating crystallization based on the DECHAROEN™ technology. The proposed system can be designed in two different ways to support all users' requirements in computerized automation controllers. The first solution is realized based on the use of distributed control system (DCS). The second solution is implemented by using combination of programmable logic controller (PLC) and supervisory control and data acquisition (SCADA). Hardware configurations using SIEMENS PCS7 and using SIEMENS S7-300/WinCC are included to demonstrate the designed concepts. Results of modernized and automated raw sugar boilings using continuous vacuum pans (CVPs) are compared to data of previous processes using batch pans for verifying the effectiveness of the proposed system. Moreover, the proposed automation system is widely accepted by 16 cane sugar manufacturers located in China and Thailand. © 2014 ISSN 2185-2766.
Some of the metrics are blocked by your
consent settings
Successful simulation of airflow in the microenvironment of an assembly automation machine and its implication
(2014-01-01)
Thongsri, Jatuporn
;
Khaokom, Adisorn
Today, the hard disk drive (HDD) industry is using assembly automation machine (AAM) to construct head stack assembly (HSA) from smaller parts. AAM needs to operate in a clean environment with very low particle counts. To achieve this end, fan filter Unit (FFU) is used to supply purified air into the environment by filtering out airborne particles from recirculating air. In this study, we investigated numerically the airflow induced by FFUs inside a microenvironment that houses an AAM in an HDD factory. The boundary conditions chosen for simulation were directly derived from the real ambient conditions in this HDD factory. We found that the FFUs not only filter out airborne particles from the air supplied into the microenvironment but also act as a particle blocker, pushing away the nearby particles in the air surrounding the openings of the microenvironment. The findings from this study can be applied to cases where other kinds of machinery need to be protected from airborne particles. © (2014) Trans Tech Publications, Switzerland.
Some of the metrics are blocked by your
consent settings
Education of automation infrastructure based on international standards - Foundation™ certified training program as an excellent illustration
(2010-01-01)
Pongswatd, Sawai
;
Julsereewong, Amphawan
;
Nontanakorn, Srinakorn
;
Sasajima, Hisashi
;
Kitai, Tsuyoshi
This paper presents a fieldbus education using real hands-on training courses set in a non-vender specific environment on Foundation™ technology, which is fully compliant with recognized international standards and enable an automation infrastructure for operational excellence. The Foundation™ Certified Training Program (FCTP) lunched by the Fieldbus Foundation to ensure uniform standards in the quality of the training center, course instructors, and program curriculum and materials is described as an excellent illustration. To actively promote the Fieldbus Foundation's technology, Foundation Fieldbus, in Thailand, King Mongkut's Institute of Technology Ladkrabang (KMITL) received strong supports from the Fieldbus Foundation Thai Association aims to complete FCTP certification process. © 2010 SICE.

KMITL

Browse

Filters

Settings

Sort By

Results per page

Search Results