Jurnal Teknologi Informasi dan Pendidikan

Web-Based Dental Caries Detection Using a Convolutional Neural Network and OpenCV

Ahmad Dzaki Alfarouq, Ahmadul Hadi, Dony Novaliendry, Khairi Budayawan — Sat, 08 Nov 2025 00:00:00 +0000

Early detection of dental caries presents a significant challenge, particularly in regions with limited access to healthcare services. While many AI models focus on binary classification, real-world applications must handle irrelevant inputs to be robust. This study develops and evaluates a web-based system using a Convolutional Neural Network (CNN) for a three-class dental image classification task: 'Caries', 'No Caries', and 'Not a Tooth'. The method employs transfer learning with the MobileNetV3 Small architecture, trained on a custom augmented dataset of 5,811 images. The model was implemented into an accessible web application using the Flask framework and OpenCV library, supporting both image upload and real-time detection. On the test set, the model achieved an overall accuracy of 93%. It demonstrated exceptional performance in rejecting irrelevant images and high reliability in identifying caries. This study presents a practical and robust tool for initial dental screening, highlighting the importance of a dedicated 'non-target' class for building trustworthy real-world AI applications in tele-dentistry.

Interpretation of the Subsurface Structure of Batanta Island Using Gravity Method and 2D Modeling

Sat, 08 Nov 2025 00:00:00 +0000

Batanta Island in Raja Ampat, West Papua, is located in an active tectonic zone dominated by fault structures and igneous rock intrusions, thus possessing significant geological potential. This research aims to identify the subsurface structure of the island using the gravity method that has undergone drift, tidal, topographic, and Bouguer corrections, followed by regional-residual component separation. 2D modeling with GM-SYS software was used to construct geological cross-sections based on rock density distribution. Results show the presence of several main units, including massive lava, breccia, Quaternary deposits, and high-density intrusion bodies (±3.00 g/cm³) that are interpreted as diorite. An active fault zone and lithological boundary were also found, which serve as the main controllers of geological structure. These findings indicate the presence of mineralization potential and complex geological dynamics, and serve as an important initial contribution to further geological exploration on Batanta Island and its surroundings

Agile-Based Field Internship Information System for Academic Administration Digitalization

Made Pradnyana Ambara, I Made Agus Oka Gunawan, I Wayan Rizky Wijaya — Mon, 24 Nov 2025 00:00:00 +0000

This study presents the development of a web-based internship management information system (PKL) aimed at improving the efficiency, transparency, and accountability of internship administration and supervision processes in the Department of Information Technology, Politeknik Negeri Bali. The system was developed using the Agile methodology to allow iterative and adaptive development, supported by the CodeIgniter 4 framework for lightweight and modular implementation. Core system features include online registration, internship letter submission, document uploads, supervision logs, and lecturer monitoring tools, all integrated to facilitate collaboration among students, supervisors, and administrators. The system’s performance was evaluated using black-box testing, which confirmed valid results across all major functional requirements, ensuring technical reliability. Furthermore, a usability assessment using the System Usability Scale (SUS) involving 20 respondents produced a score of 84.50, categorized as excellent, indicating high user satisfaction, ease of use, and efficiency. These findings demonstrate that the system not only fulfills its functional objectives but also enhances the digital transformation of vocational academic services, offering a replicable and scalable model for other educational institutions aiming to modernize their internship management processes.

Modeling Bedoyo Majapahit Dance Motion Using HMM Emission Families

Anang Kukuh Adisusilo, Emmy Wahyuningtyas, Teguh Pribadi Ikhsan — Mon, 24 Nov 2025 00:00:00 +0000

This study investigates three types of emission families in Hidden Markov Models (HMMs) for reconstructing Bedoyo Majapahit dance motion captured using a markerless system. The recorded skeleton data, consisting of 3,341 frames and 33 joints per frame, were normalized and reduced into a 30-dimensional latent space using Principal Component Analysis (PCA). Three emission variants were evaluated: single-Gaussian HMM, Gaussian-mixture HMM (GMM-HMM), and Multinomial HMM. The evaluation employed a tri-metric scheme consisting of Mean Squared Error (MSE), Dynamic Time Warping (DTW), and Fréchet distance to measure reconstruction fidelity. The experimental results showed that GMM-HMM consistently outperformed the other two models, achieving the lowest reconstruction error and the closest alignment to the original temporal and geometric motion profiles. The Gaussian HMM demonstrated moderate performance but tended to underestimate motion amplitude, while the Multinomial HMM produced the weakest results due to the discretization of continuous pose data. These findings indicate that multimodal emission functions provide a more expressive representation for continuous dance motion. The study highlights the suitability of GMM-HMM for traditional dance preservation through computational modeling and contributes to the development of digital motion archiving for cultural heritage.