Search Results

Now showing 1 - 4 of 4

Some of the metrics are blocked by your
consent settings
A deep learning model (FociRad) for automated detection of γ-H2AX foci and radiation dose estimation
(2022-12-01)
Wanotayan, Rujira
;
Chousangsuntorn, Khaisang
;
Petisiwaveth, Phasit
;
Anuttra, Thunchanok
;
Lertchanyaphan, Waritsara
DNA double-strand breaks (DSBs) are the most lethal form of damage to cells from irradiation. γ-H2AX (phosphorylated form of H2AX histone variant) has become one of the most reliable and sensitive biomarkers of DNA DSBs. However, the γ-H2AX foci assay still has limitations in the time consumed for manual scoring and possible variability between scorers. This study proposed a novel automated foci scoring method using a deep convolutional neural network based on a You-Only-Look-Once (YOLO) algorithm to quantify γ-H2AX foci in peripheral blood samples. FociRad, a two-stage deep learning approach, consisted of mononuclear cell (MNC) and γ-H2AX foci detections. Whole blood samples were irradiated with X-rays from a 6 MV linear accelerator at 1, 2, 4 or 6 Gy. Images were captured using confocal microscopy. Then, dose–response calibration curves were established and implemented with unseen dataset. The results of the FociRad model were comparable with manual scoring. MNC detection yielded 96.6% accuracy, 96.7% sensitivity and 96.5% specificity. γ-H2AX foci detection showed very good F1 scores (> 0.9). Implementation of calibration curve in the range of 0–4 Gy gave mean absolute difference of estimated doses less than 1 Gy compared to actual doses. In addition, the evaluation times of FociRad were very short (< 0.5 min per 100 images), while the time for manual scoring increased with the number of foci. In conclusion, FociRad was the first automated foci scoring method to use a YOLO algorithm with high detection performance and fast evaluation time, which opens the door for large-scale applications in radiation triage.
Some of the metrics are blocked by your
consent settings
Classification model of optical character recognition failures in unrecovered slider serial numbers in hard disk drive manufacturing and image capture processes
(2022-01-01)
Chousangsuntorn, Chousak
;
Tongloy, Teerawat
;
Chuwongin, Santhad
;
Boonsang, Siridech
In hard disk drive (HDD) manufacturing processes, there are unrecovered serial number images about 0.01% from the standard optical character recognition (OCR) reading and deep learning approach. We found several failures from two main causes, i.e. manufacturing process and image capture process during standard OCR reading. We proposed classification model used for recognizing the serial number reading failures based on object detection You-Only-Look- Once (YOLO) algorithm and EfficientNet-B0 classification network as well as histogram analysis. The 1000 images captured by digital camera were used for training (600 images) and validation (400 images) the ROI detection model. The other 2100 captured images were used for training and testing classification OCR failure from manufacturing process model. The model testing was performed in 900 images contained 9 causes (classes) of failures. The proposed model reaches F1 score = 0.94.
Some of the metrics are blocked by your
consent settings
Three-stage deep learning system for recognizing contaminated serial numbers in hard disk drive: A comparison study with two-stage deep learning model
(2022-01-01)
Chousangsuntorn, Chousak
;
Tongloy, Teerawat
;
Chuwongin, Santhad
;
Boonsang, Siridech
The previous two-stage deep learning model for detecting and classifying misidentified serial numbers on the defect hard disk drive (HDD) slider was proposed by authors. We found that the threshold level adjusted during preprocessing process could limit the robustness of the two-stage model in large-scale manufacturing. Thus, we proposed a three-stage deep learning model comprised of 1) region of interest (ROI) detection and cropping, 2) character detection and cropping, and 3) character classification. Object detection algorithm and classification network used in this model are based on YOLO v.4 and EfficientNet-B0. The 1000 images captured by the digital camera were used for training (600 images) and validation (400 images) of the ROI detection model. The other 1000 captured images were used for testing the performance of the proposed three-stage model, then we compared them with those obtained from the previous two-stage model. The proposed three-stage model reaches F1 score = 0.997 and recovery rate up to 95.9%, while the two-stage model yields only 0.948 and 73%, respectively.
Some of the metrics are blocked by your
consent settings
A deep learning system for recognizing and recovering contaminated slider serial numbers in hard disk manufacturing processes
(2021-09-01)
Chousangsuntorn, Chousak
;
Tongloy, Teerawat
;
Chuwongin, Santhad
;
Boonsang, Siridech
This paper outlines a system for detecting printing errors and misidentifications on hard disk drive sliders, which may contribute to shipping tracking problems and incorrect product delivery to end users. A deep-learning-based technique is proposed for determining the printed identity of a slider serial number from images captured by a digital camera. Our approach starts with image preprocessing methods that deal with differences in lighting and printing positions and then progresses to deep learning character detection based on the You-Only-Look-Once (YOLO) v4 algorithm and finally character classification. For character classification, four convolutional neural networks (CNN) were compared for accuracy and effectiveness: DarkNet-19, EfficientNet-B0, ResNet-50, and DenseNet-201. Experimenting on almost 15,000 photographs yielded accuracy greater than 99% on four CNN networks, proving the feasibility of the proposed technique. The EfficientNet-B0 network outperformed highly qualified human readers with the best recovery rate (98.4%) and fastest inference time (256.91 ms).

KMITL

Browse

Filters

Settings

Sort By

Results per page

Search Results