Efendi, Shendi Teuku Maulana (2025) Prediksi Penyakit Tuberkulosis Pada Citra Chest X-Ray Menggunakan Deep Learning Berbasis Arsitektur DenseNet-121. S1 thesis, Universitas PGRI Madiun.
![[img]](https://eprint.unipma.ac.id/style/images/fileicons/text.png) |
Text
cover-daftar_lampiran.pdf Download (2MB) |
|
Text
abstrak-en-id.pdf Download (545kB) |
|
Text
BAB I.pdf Download (1MB) |
|
Text
BAB II.pdf Download (7MB) |
|
Text
BAB III.pdf Restricted to Repository staff only Download (2MB) | Request a copy |
|
Text
BAB IV.pdf Restricted to Repository staff only Download (15MB) | Request a copy |
|
Text
BAB V.pdf Download (844kB) |
|
Text
daftar_pustaka.pdf Download (1MB) |
|
Text
lampiran.pdf Download (5MB) |
Abstract
Tuberculosis (TB) remains a communicable disease and a global health concern, especially in developing countries such as Indonesia. Manual diagnosis of TB through chest X-ray images is subjective, time-consuming, and requires experienced medical professionals. This study develops a TB detection system using a modified DenseNet-121 architecture to accept grayscale image input. The dataset was collected from Kaggle and Mendeley, consisting of 4,014 normal and 3,194 TB images. To address class imbalance, undersampling and class weighting techniques were applied. The model was trained using PyTorch with Binary Cross Entropy Loss, Adam optimizer, a learning rate of 0.001, batch size of 32, and 25 epochs. The dataset was split in a 70:15:15 ratio for training, validation, and testing. Model evaluation using the confusion matrix and AUC-ROC yielded an accuracy 97%, precision 99%, recall 94%, F1-score 97% and AUC-ROC 97%. The trained model was converted to ONNX format for deployment. The frontend was developed using Next.js with features including login, TB detection, and prediction history. The backend was built using FastAPI to handle model inference, authentication, and data management through the website.
| Item Type: | Thesis/Skripsi/Tugas Akhir (S1) |
|---|---|
| Kata Kunci: | Tuberculosis, Deep Learning, DenseNet-121, Chest X-ray, AUC-ROC, Confusion Matrix, Undersampling. NextJS, FastAPI |
| Subjects: | Q Science > Q Science (General) Q Science > QA Mathematics > QA75 Electronic computers. Computer science Q Science > QA Mathematics > QA76 Computer software |
| Divisions: | Fakultas Teknik > Teknik Informatika |
| Depositing User: | MAULANA TEUKU SHENDI |
| Date Deposited: | 30 Jul 2025 06:24 |
| Last Modified: | 30 Jul 2025 06:24 |
| URI: | http://eprint.unipma.ac.id/id/eprint/3298 |
Actions (login required)
 |
View Item |
