KMITL

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

Browse

Search Results

Now showing 1 - 5 of 5

Some of the metrics are blocked by your
consent settings
Performance of deep learning models for the classification and object detection of different oral white lesions using photographic images
(2025-12-01)
Khovidhunkit, Siribang on Piboonniyom
;
Phosri, Kunchidsong
;
Thanathornwong, Bhornsawan
;
Rungraungrayabkul, Dulyapong
;
Poomrittigul, Suvit
Computer vision adjunctive technology for oral lesion diagnoses has been developed to detect and identify Oral Potentially Malignant Disorders (OPMDs) and non-OPMDs. The early detection of OPMDs can reduce the risk of oral cancer development, improving the survival rate of the patients. This study aims to evaluate the computer vision technique in the white oral lesion domain within the scope of photographic images. Deep learning techniques for the classification of Convolution Neural Networks (CNNs) and transformer neural networks, and one-stage models of YOLOv7 and YOLOv8 were utilized to classify and detect five classes of OPMDs and non-OPMDs oral white lesions including oral leukoplakia, oral lichen planus, pseudomembranous candidiasis, oral ulcers covered with pseudomembrane and other white benign oral lesions. From the evaluation results of classification, the IFormerBase model achieves overperformance compared to CNN models with accuracy, precision, and F1 score of more than 80% on the test set. The best model for object detection is YOLOv7 with 84.5% mean Average Precision (mAP) at Intersection over Union (IoU) threshold of 0.3 and 74.5% at IoU of 0.5 on the test set. Object detection results reveal promising automatic oral lesion identification, which can be further developed to enhance the lesion screening system.
Some of the metrics are blocked by your
consent settings
Computational algorithm of high-intensity focused ultrasound beams in cancer tissue model for hyperthermia therapy
(2020-01-01)
Songkaitiwong, Kittiphot
;
Locharoenrat, Kitsakorn
We calculate the acoustic field of a high-intensity focused ultrasound field in a 2D-model of human breast carcinoma and induce temperature elevation for the generation of necrosis. The computational operation is based on the Pennes bioheat concept. This method provides precise heat transfer values based upon thermal conduction in soft tissue and thermal convection in the domain of the blood. An ultrasound beam at 1 MHz was laterally focused on a tumor of 15.0 mm x 28.5 mm at different focal depths without elevational focalization. The length of each focus point of the ultrasound beam was 6.8 - 45.4 mm on the vertical axis, whereas the full width at half maximum was 1.1-2.5 mm on the horizontal axis. Simulated results showed that a discrepancy of the acoustic pressure around the focus area rises with focal depth. Like the pressure profile, when the focal depth is close to the ultrasound source, the thermal homogeneity around the focus area is attained, whereas thermal uniformity around the focus area becomes worse with increasing focal depth. Using the data visualization arrangement, a temperature profile corresponding to the obtained pressure profile is converted to attain a 2D image of a model of human breast carcinoma to show that tumor ablation was achieved and the healthy surrounding tissues were safe.
Some of the metrics are blocked by your
consent settings
COMPUTATIONAL ALGORITHM of TWO PARALLEL ULTRASOUND BEAMS of 1D CANCER TISSUE MODEL for SAFE and EFFECTIVE HYPERTHERMIA TREATMENT
(2019-05-01)
Songkaitiwong, Kittiphot
;
Locharoenrat, Kitsakorn
The mathematical algorithm of two parallel ultrasound beams on a one-dimensional (1D) cancer tissue model for hyperthermia treatment was created using Matlab software. Physically, the model incorporated two beams; the first beam was permanently placed at the center of the tumor, whereas the other was set between the first beam and the tumor. The computational implementation of this technique relies on the Crank-Nicolson method. This technique is a finite different method that offers an exact heat transfer calculation based on the heat analysis of the heat node structure from a 1D biological tissue model. The Matlab software implementation was composed of two stages: tissue temperature profile calculation and optimization computation. To obtain the tissue temperature profile, the beam heat was varied from 45C to 75C (seven different levels of heat from the same source), while the second beam was allowed to move between the first beam and the tumor to locations at distances of 1 to 9mm (nine positions). The obtained tissue temperature profiles were subsequently analyzed to achieve the optimal time, beam position, and beam heat of the treatment. As a result of the optimization, the best position for the second beam was determined to be 5mm from the center of the tumor. Further, all tumor cells were observed to have died, whereas all normal tissues were safe. The optimal time, beam position, and beam heat of the treatment were finally collected to create and fit a mathematical function for further hyperthermia treatment.
Some of the metrics are blocked by your
consent settings
Free radical scavenging properties and induction of apoptotic effects of fa fraction obtained after proteolysis of bioactive peptides from microalgae synechococcus sp. VDW
(2019-01-01)
Suttisuwan, Rutairat
;
Phunpruch, Saranya
;
Saisavoey, Tanatorn
;
Sangtanoo, Papassara
;
Thongchul, Nuttha
This study aims to determine the antioxidant activity of bioactive peptides derived from Synechococcus sp. VDW cells cultured for 21 days. Synechococcus sp. VDW protein hydrolysates were prepared with trypsin and purified by ultrafiltration with molecular mass cut-off membranes of 10, 5 and 3 kDa. The M<3 kDa (FA) fraction had the highest 2,2'-azino- bis(3-ethylbenzothiazoline-6-sulphonic acid) (ABTS) and 2,2'-diphenyl-1-picrylhydrazyl (DPPH) radical scavenging activities, with IC50 values of (11.5±0.3) and (13.6±0.2) μg/ mL, respectively. The FA fraction was separated by reversed phase HPLC to yield four subfractions (F1-4). The F4 subfraction showed the highest maximum ABTS radical scavenging activity (3.55±0.61) % and it was selected for further analysis by electrospray ionisation quadrupole time-of-flight mass spectrometry (ESI-Q-TOF-MS/MS) based on de novo peptide sequencing. Five antioxidant peptides were identified, of which AILESYSAGKTK had the highest ABTS radical scavenging activity. Furthermore, the FA fraction showed high cytotoxic activities against human cancer-derived cell lines, especially the colon cancer cell line (SW620) with an IC50 value of (106.6±21.5) μg/mL, but not the untransformed Wi38 cell line. The FA fraction activated the apoptotic pathway in SW620 cells after treatment for 24, 48 and 72 h, with the highest activities of caspases-3, -8 and -9 being observed after treatment for 72 h. These findings suggested that microalgae Synechococcus sp. VDW may be used to develop natural anticancer drugs.
Some of the metrics are blocked by your
consent settings
Isolation of circulating tumor cells under hydrodynamic loading using microuidic technology
(2014-01-01)
Zhao, Cong
;
Lee, Yi Kuen
;
Xu, Rui
;
Liang, Chun
;
Liu, Dayu
Cancer is a leading cause of mortality worldwide causing human deaths. Circulating tumor cells (CTCs) are cells that have detached from a primary tumor and circulate in the bloodstream; they may constitute seeds for subsequent growth of additional tumors (metastasis) in different tissues. The detection of CTCs may have important prognostic and therapeutic implications but, because their number is very small, these cells are not easily detected. Circulating tumor cells are found in the order of 10-100 CTCs per mL of whole blood in patients with metastatic disease. Isolation of tumor cells circulating in the blood stream, by immobilizing them on surfaces functionalized with bio-active coating within microfluidic devices, presents an interdisciplinary challenge requiring expertise in different research areas: cell biology, surface chemistry, fluid mechanics and microsystem technology. We first review the fundamental of cell biology of CTCs and summarize the key microfluidic techniques for isolation of CTCs via cell-ligand interactions, magnetic interactions, filtration; detection and enumeration of CTCs; in vivo CTCs imaging.

KMITL

Browse

Filters

Settings

Sort By

Results per page

Search Results