14 results
Search Results
Now showing 1 - 10 of 14
Item Open Access Deep learning based technique for classification of abdominal aortic aneurysm (AAA) in ct-scan images(Mahidol University, 2023) Miao, Zhiqi, 1998-; Worapan Kusakunniran; Thanongchai Siriapisith; Pairash Saiviroonporn-of-the-art deep convolutional neural networks (ResNet, DenseNet, and EfficientNet) are adopted for 2D-based AAA classification on the non-contrast and contrast CT datasets. Based on our final patient-based fusion data division of training, our AAA-orientedItem Open Access Brain hemorrhage segmentation in ct scan images using deep learning based approach(Mahidol University. Mahidol University Library and Knowledge Center, 2022) Zhang, Haibin, 1993-; Worapan Kusakunniran; Thanongchai Siriapisith; Pairash Saiviroonporn; therefore, rapid and accurate judgments of CT images of cerebral hemorrhage are vital to reduce mortality and sequelae. This research aims to propose an improved active learning method based on a convolutional neural network for small lesion cerebral... hemorrhage segmentation and improve performance by iterative retraining based on segmentation output on large datasets without ground truth. After using a dataset annotated by experts to obtain the initial network model, another unannotated datasetItem Metadata only Khmer WordNet construction(Mahidol University. Mahidol University Library and Knowledge Center, 2024) Udorm, Phon, 1996-; Charnyote Pluempitiwiriyawej; Suppawong Tuarob; Thanapon NorasetThis thesis described a semi-automatic translation-based expansion approach for Khmer WordNet construction. This approach expanded the Princeton English WordNet's synsets and took into account the existing machine-readable Cambodian-English... number of candidate links between Khmer words and English synsets can be derived. This approach applied a statistical approach sampling these candidate links and constructed a statistical model based on human verification of the sample of the candidateItem Open Access IOT security based on node-red for secure room monitoring(Mahidol University. Mahidol University Library and Knowledge Center, 2022) Tawatchai Chaipimansri; Thitinan Tantidham; Assadarat Khurat; Dolvara GunatilakaNowadays, the emergence of Internet of Things (IoT) for a secure room monitoring system with an external cloud server for a user notification on the Internet has been a significant role and data security concern. This research project is to develop a secure room monitoring system in such a server room compositing components as follows. (1) A Node-RED Edge System (NRES) equipped with Raspberry Pi (RPi) and a set of environmental sensors such as Pi camera, PIR, and DHT22 is an edge monitoring system. (2) Digital Ocean Cloud Server (DOCS) as a web portal interface service for receiving encrypted notification messages via Rivest–Shamir–Adleman (RSA) from the NRES and forwarding the messages to the system users. (3) Pinata is an InterPlanetary File System (IPFS) for sharing camera image files from an NRES to users. (4) LINE System is for displaying a notification message from DOCS when an unauthorized NRES is detected besides an anomaly environmental status. And (5) Email is used for users to access more notification descriptions. Security mechanisms in this research are deployed as follows. Besides strong authentication, authorization, and account access of each NRES member, the MAC address from each NRES member recorded on the DOCS is used for the verification to protect any rouge Node-RED system. Each MQTT payload between Node-RED system is encrypted by using RSA to mitigate a man-in-the-middle attack. Furthermore, the NRES sends an encrypted Pinata’s CID URL for a sharing camera image file to the user’s email. The DOCS provides the web UI for authenticating users in order to copy and paste the encrypted CID URL obtained from the email. Finally, LINE is applied to notify users of an anomaly event and to forward more event descriptions to user’s email address as multi-factor authentication. Implication of the Thematic Paper: According to the implementations and results of the study, suggestions for further research are as follows. 1. The payload encryption in Node-RED is considered Node-to-Node encryption rather than End-to-End encryption because the message will be decrypted by Node-RED before sending to clients. As a result, additional notification systems such as message forwarding to user’s email address can be implemented to enhance the security and privacy of data. 2. Since the connection speed of Cloud services is intermittent, the performance evaluation should be done in multiple sessions, for instance, once in the morning and the other in the evening.Item Open Access An automatic web server auditing tool based on CIS benchmark(Mahidol University. Mahidol University Library and Knowledge Center, 2022) Wasutum Kethom; Assadarat Khurat; Ittipon Rassameeroj; Dolvara Gunatilakabenchmark. Also, this tool generates two types of reports for the auditor: a summary report and a detailed report so that the auditor can get more information about the audit result. Besides, remediations based on CIS benchmark is also provided in the detail... generates two types of reports for the auditor: a summary report and a detailed report so that the auditor can get more information about the audit result. Besides, remediations based on CIS benchmark is also provided in the detail report.Item Metadata only Objective assessment and feedback generation in dental surgical simulation : a framework based on correlating procedure and outcome(Mahidol University. Mahidol University Library and Knowledge Center, 2023) Myat, Su Yin, 1978-; Haddawy, Peter F.; Siriwan Suebnukarn; Worapan Kusakunniran; Mores Prachyabruedspecific errors and their severity. Building upon this, we developed techniques to identify the portion of the procedure responsible for the errors. Specifically, for the errors in the outcome the responsible portions of the procedure are identified basedItem Metadata only Bioinformatics tool recommendation based on usage context evidences(Mahidol University. Mahidol University Library and Knowledge Center, 2024) Angkana Huang; Apirak Hoonlor; Ellison, Damon W.The abundance of bioinformatics tools has grown exponentially over the last three decades. Concurrently, many tools become outdated due to their discontinuation. Staying updated is highly difficult and is costly in terms of time and effort. The existing systems to ease tool discovery primarily attempts to index all the available tools according to their functionalities. Some of them provides the number of cites the tools received to indicate their popularity. The size of these lists have grown large calling for more targeted retrieval approaches. Since the tools are typically used in conjunction, a recent approach allows the users to browse the tools according to expert maintained pipelines. However, generating and updating these pipelines remains manual. Moreo- ever, the actual conjunct use of the tools are not provided. This thesis suggests an auto- mated pipeline derivation method through literature mining and cross-citation analysis. The analysis patterns and the tool usage patterns were recovered from the data. Recommendation models were built from the data and the derived patterns. Through evaluating the models, actual tool selection behaviors were also understood. During 2009-2016, the tool functionalities with their popularity considered was highly predictive of whether they have been chosen. Along the period, substituting the overall popularity with the local popularity within the recovered analysis patterns became increasingly predictive and was on par in 2016. Such results implies that the recovered patterns resemble the pipelines used in community. Lastly, the pipelines were queried into the recommendation models to obtain the community accepted best-practices. These analysis patterns and best-practices can be used to inform experts regarding the status-quo of the field and can be used as guidelines for newcomers entering the field.Item Open Access Human detection and social distancing measurement in a video(Mahidol University. Mahidol University Library and Knowledge Center, 2022) Kosin Saramas; Worapan Kusakunniran; Boonsit Yimwadsanavideo, 2) calibrating camera views, and 3) measuring distances between any two persons. The tracking technique is based on YOLO algorithm, a famous object detection algorithm, that identifies specific objects in the video. In this project, YOLOv3 is usedItem Open Access Shiro : a centralized system for detecting attacks from windows event logs(Mahidol University, 2021) Vatcharanun Moonkhaen; Vasaka Visoottiviseth; Damras Wongsawang; Thitinan TantidhamMany organizations or companies have experienced the damages of cyberattacks leveraging the MS Windows products' vulnerabilities, especially as numerous personal computers worldwide are still running the older Windows 7 version without installing... the detection signatures of each CVE from the specific Event logs and their details, so once SHIRO identifies the attack signatures in the records, it identifies the attack type and alerts the administrator. Our experiments, based on both the datasetsItem Open Access A comprehensive framework for migrating to zero trust architecture(Mahidol University. Mahidol University Library and Knowledge Center, 2023) Pacharee Phiayura; Songpon Teerakanok; Vasaka Visoottiviseththe desired security state before deploying ZTA. Later, the organization implemented ZT components and migrated users to ZTA. Finally, the organization monitored and optimized the security performance of ZTA. Thus, based on the meta-analysis results, a