Phasukkit, Pattarapong
Loading...
Preferred name
Phasukkit, Pattarapong
Alternative Name
Phasukkit, P.
Main Affiliation
Email
pattarapong.ph@kmitl.ac.th
41 results
Now showing 1 - 10 of 41
- Some of the metrics are blocked by yourconsent settings
Item type:Publication, Multi-Entropy Feature Extraction With LSTM Networks for Acoustic Emission-Based Railway Crack Localization(2026-01-01) - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Recurrent neural network based underground object detection using a-scan ground penetrating radar(2021-05-19) - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Crack Localization Detection in Monolithic Zirconia Dental Crowns via 1D-Convolutional Neural Networks Algorithm-Based Acoustic Emission Analysis(2024-01-01); - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Cross-Sensor and Cross-Population Generalization of Deep Learning Models for Digital Mammography: A Controlled Four-Country Benchmark of Five Backbone Architectures with Statistical Significance Testing(2026-06-01); Background/Objectives: Deep learning models for digital mammography sensor data are increasingly deployed across hospitals using different X-ray detector technologies and patient populations. Whether models trained on one sensor platform and population maintain accuracy when transferred to another has not been tested for the latest generation of mammography-specific foundation models under one controlled protocol. Methods: We fine-tuned five backbone architectures (ResNet-50, DINOv2-B14, Rad-DINO, Mammo-CLIP B5, and Mammo-FM) on CBIS-DDSM (film-digitized, USA, n = 714 validation) with three seeds, ablated a density-aware focal loss across three auxiliary weights, and evaluated transfer to three external sensor cohorts: CMMD (full-field digital, China, n = 1032), DMID (mixed digital, India, n = 509), and MIAS (film-digitized, UK, n = 322). Significance used paired DeLong z-tests with Benjamini–Hochberg FDR correction; temperature scaling tested post hoc recalibration at all transfer targets. Results: Within this single-source three-seed evaluation, ResNet-50 outperformed all four foundation models on CBIS-DDSM (AUC 0.867 vs. 0.847, 0.846, 0.813, and 0.703; all gaps p_adj < 0.05). The density-aware focal loss degraded both AUC and calibration at every weight tested. At transfer, every model lost 0.165 to 0.320 AUC points relative to in-distribution performance, with sensitivity at 95% specificity collapsing from 0.31 to 0.47 in-distribution to 0.11 to 0.22 across the three external targets. A per-seed Stouffer meta-analysis confirms that Mammo-CLIP B5 and Mammo-FM significantly outperformed ResNet-50 on DMID and Mammo-CLIP on CMMD, after BH-FDR; MIAS comparisons remained directional only. In the extremely dense subgroup (BI-RADS D4), Mammo-FM reached AUC 0.870 versus ResNet-50 at 0.842, a directional observation whose 95% CIs overlap heavily at the n = 140 sample size and which we do not interpret as a statistically supported advantage. Conclusions: In this single training-source, three-seed protocol, mammography-specific pretraining did not deliver the in-distribution AUC premium reported in the originating papers, and no architecture reached a level at which transfer deployment without local validation would be defensible. We frame these as observations specific to the present protocol rather than as broader conclusions about foundation models for mammography classification. The findings argue for sensor-stratified and population-stratified external validation and for local recalibration as practical prerequisites before clinical use. Code and weights are released under MIT license. - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Feasibility of Prediction Model for Internal Tumor Target Volume from 4-D Computed Tomography of Lung cancer(2021-01-01); - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Vertical speed prediction for the efficient landing of aircraft using GRU(2020-11-04); Pradabpet, ChusitAir traffic management is currently the process of managing air traffic demand. For support the capacity to handle the amount of air traffic. Whether it is in the airspace sector or the airport area. There are many factors such as the ability of the air traffic providers, physical characteristics of airspace and airport terrain. The main goal of air traffic management is to manage air traffic control. Streamlined and demand-capacity balancing are safe and effective. Whether it is air traffic service providers or passengers. In this research, the experiment was carried out on specific aircraft model A-320. Traveling from Bangkok Suvarnabhumi Airport to Phuket International Airport. In which the landing airport has island terrain. Therefore, requires high security due to physical characteristics in the airspace which has many factors such as wind and wind direction and to provide safety and effective landing. Therefore, take the factors affecting the landing to predict the vertical speed. The epoch =2000, learning rate =0.01, accuracy more than 90% for training set and 90% for test set. - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Automated Crack Detection in Monolithic Zirconia Crowns Using Acoustic Emission and Deep Learning Techniques(2024-09-01); ; Monolithic zirconia (MZ) crowns are widely utilized in dental restorations, particularly for substantial tooth structure loss. Inspection, tactile, and radiographic examinations can be time-consuming and error-prone, which may delay diagnosis. Consequently, an objective, automatic, and reliable process is required for identifying dental crown defects. This study aimed to explore the potential of transforming acoustic emission (AE) signals to continuous wavelet transform (CWT), combined with Conventional Neural Network (CNN) to assist in crack detection. A new CNN image segmentation model, based on multi-class semantic segmentation using Inception-ResNet-v2, was developed. Real-time detection of AE signals under loads, which induce cracking, provided significant insights into crack formation in MZ crowns. Pencil lead breaking (PLB) was used to simulate crack propagation. The CWT and CNN models were used to automate the crack classification process. The Inception-ResNet-v2 architecture with transfer learning categorized the cracks in MZ crowns into five groups: labial, palatal, incisal, left, and right. After 2000 epochs, with a learning rate of 0.0001, the model achieved an accuracy of 99.4667%, demonstrating that deep learning significantly improved the localization of cracks in MZ crowns. This development can potentially aid dentists in clinical decision-making by facilitating the early detection and prevention of crack failures. - Some of the metrics are blocked by yourconsent settings
Item type:Publication, A Contactless Edge-AI Prototype for Simulated Apnea-like Respiratory Suppression and Motion Artifact Detection Using 60 GHz FMCW Radar(2026-07-01); Sleep-related respiratory disturbances are difficult to monitor continuously outside specialized laboratories because conventional polysomnography is resource-intensive and intrusive. This study presents a contactless edge-AI engineering prototype for detecting controlled voluntary respiratory-motion suppression and motion artifacts using a 60 GHz frequency-modulated continuous-wave radar. The system integrates a 60 GHz radar front end, lightweight local preprocessing, an INT8 one-dimensional convolutional neural network deployed on the Analog Devices MAX78000 CNN accelerator (Analog Devices Thailand, Chon Buri, Thailand), and an event-driven Raspberry Pi Zero 2W gateway for alert transmission. Evaluation was performed using a controlled healthy-volunteer dataset consisting of normal breathing, voluntary breath-holding-induced respiratory suppression, and deliberate motion artifact. The final valid test set contained 270 technically valid 30 s windows balanced across the three classes. The INT8 model achieved an overall accuracy of 92.6% (95% confidence interval: 88.8–95.2%), with a macro-averaged precision, recall, and F1-score of 92.6%, 92.6%, and 92.5%, respectively. Active CNN inference on the MAX78000 consumed 0.152 ± 0.011 mJ and was completed in 5.20 ± 0.11 ms, corresponding to approximately 280-fold lower active inference energy than Python 3.14.6/TensorFlow Lite 2.21.0-based execution on the Raspberry Pi Zero 2W. These results demonstrate the feasibility of privacy-aware, low-power respiratory-pattern classification at the edge. However, the study should be interpreted strictly as an engineering proof-of-concept based on controlled voluntary breathing and movement tasks in healthy volunteers. It is not a clinically validated apnea or obstructive sleep apnea detection system and did not include polysomnography, oxygen saturation measurement, airflow sensing, sleep staging, or diagnosed patient cohorts. - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Air turbulence forecasting of airbus type A320 in southeast asia using deep learning method(2020-11-04); Pradabpet, ChusitThis paper is present the Air Turbulence Forecasting of Airbus type A320 in Southeast Asia using the Deep Learning Method. It will collect flight data of aircraft in the region of Southeast Asia (Thailand and Vietnam). In which data was collected for 40 flights to forecast the occurrence of air turbulence in 3 status consist of non-turbulence, the rapid decrease and increase the altitude is out of pilot control. In this research using deep learning with supervise learning and create the mathematical model for air turbulence forecasting which could help to reduce the wastage that may affect to the passenger. The results, using 5 layers of deep learning (1 input layer, 3 hidden layers, and 1 output layer). The most suitable model consists of 9 features such as vertical speed, calibrated altitude, wind speed, wind direction (wind angle), temperature, latitude, longitude, true airspeed, and indicated airspeed. The output layer consists of 3 classes (class1=non-turbulence, class2=increase altitude, and class3=decrease altitude) and optimization the weight with gradient descent. The epoch number is 1500 and the learning rate is 0.1, which will get accuracy 88% for the train set and 86 % for the test set. - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Performance Comparison of Deep Learning Approach for Automatic CT Image Segmentation by Using Window Leveling(2021-01-01); ; Dankulchai, PittayaIn tumor radiotherapy process, radiologist need to make multipleorgans contouring on medical images such as CT scans for computing appropriate dose and making a suitable treatment plan for patients. This is a necessary step before treatment. This paper was written to be one of automatic image segmentation research by using deep learning. The experiment compared performance between preprocessing input datasets with custom window leveling normalization and following by organ types. We chose the bladder, the rectum and the femur as target organs in this paper. Datasets are directly obtained from Siriraj Hospital that contoured by radiologists. There are 10 datasets of each organs. We used U-Net as main structure to extract features on image then evaluated by dice similarity coefficient (DSC) and intersection over union (IoU). The experiment resulted that training with custom window leveling normalization is better performance. The bladder got DSC and IoU of 78.34% and 70.46%, femur were 39.71% and 28.03%, and rectum were 19.19% and 12.20%, respectively.