Agus Perdana Windarto
Department of Informatics, Master’s Program, STIKOM Tunas Bangsa, Indonesia

Published : 2 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 2 Documents
Search

Enhanced BiLSTM-CNN with Adaptive Regularization for Sequential Lung Cancer CT Scan Segmentation Juanda Hakim Lubis; Agus Perdana Windarto; Sundari Retno Andani
Jurnal Teknik Informatika (Jutif) Vol. 7 No. 4 (2026): JUTIF Volume 7, Number 4, August 2026
Publisher : Informatika, Universitas Jenderal Soedirman

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52436/1.jutif.2026.7.4.5999

Abstract

Lung cancer remains one of the leading causes of cancer-related mortality worldwide, making accurate and consistent CT scan segmentation essential for early diagnosis and treatment planning. This study proposes an enhanced Bidirectional Long Short-Term Memory-Convolutional Neural Network (BiLSTM-CNN) framework for sequential lung cancer CT scan segmentation. The proposed model integrates stacked BiLSTM layers, dropout regularization, adaptive learning-rate scheduling, and an Enhanced ResNet-18 backbone to improve temporal consistency and spatial feature representation across sequential CT slices. Unlike conventional CNN-based segmentation approaches that process slices independently, the proposed framework captures inter-slice contextual dependencies to produce more stable and anatomically consistent segmentation results. The model was evaluated using a publicly available lung cancer CT scan dataset with an 80:20 training–testing split. Experimental results demonstrate that the proposed method outperformed standard LSTM and conventional BiLSTM models, achieving superior segmentation performance with a Dice Similarity Coefficient (DSC) of 0.6960 and an Intersection over Union (IoU) of 0.5337. The findings indicate that the integration of bidirectional temporal modeling and enhanced feature extraction significantly improves segmentation accuracy and generalization capability. This study contributes to the development of lightweight and efficient AI-based medical imaging systems that support more reliable lung cancer diagnosis and clinical decision-making.
Hybrid DeepLabV3+ With Embedded U-Net Skip Connections for Precise Brain Tumor Segmentation in MRI Abdul Karim; Iwan Purnama; Kusmanto Kusmanto; Agus Perdana Windarto
Jurnal Teknik Informatika (Jutif) Vol. 7 No. 5 (2026): JUTIF Volume 7, Number 5, October 2026
Publisher : Informatika, Universitas Jenderal Soedirman

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52436/1.jutif.2026.7.5.5663

Abstract

Brain tumor segmentation from MRI is crucial for diagnosis, treatment planning, and patient monitoring, yet many automated segmentation models struggle to capture global contextual information while preserving fine spatial details, leading to inaccurate delineation of complex tumor boundaries and reduced clinical reliability. Manual segmentation by radiologists is also constrained by observer variability, prolonged analysis time, and reliance on individual expertise. To address these limitations, this study proposes Hybrid DeepLabV3+, an automated segmentation model that combines DeepLabV3+ multi-scale context modeling through Atrous Spatial Pyramid Pooling (ASPP) with spatial-detail preservation via U-Net skip connections. The objective is to develop and rigorously evaluate a more accurate and generalizable brain tumor segmentation model compared with U-Net, DeepLabV3+, and EfficientNet baselines. The model is trained on the BraTS dataset using T2-weighted MRI and evaluated using Dice score, intersection over union (IoU), precision, recall, and accuracy. Experimental results show that the proposed model achieves a Dice score of 0.785, IoU of 0.740, precision of 0.796, recall of 0.985, and accuracy of 0.991, surpassing the comparison models. These results indicate that Hybrid DeepLabV3+ can identify and map tumor regions with high fidelity, supporting its potential integration into clinical decision-support systems.