Salam Mohammed, Sajad, Abih, Aya R., Mutar, Modhi Lafta, Hameed, Asaad Shakir, Hammad, Saja Jumaa, Jawad Alataby, Ahmed K.. (1405). Implementing a Convolutional Neural Network with Adam Optimizer for Pneumonia Detection Through Chest X-ray Imaging. , 18(3), 243-258. doi: 10.22059/jitm.2026.108393
Sajad Salam Mohammed; Aya R. Abih; Modhi Lafta Mutar; Asaad Shakir Hameed; Saja Jumaa Hammad; Ahmed K. Jawad Alataby. "Implementing a Convolutional Neural Network with Adam Optimizer for Pneumonia Detection Through Chest X-ray Imaging". , 18, 3, 1405, 243-258. doi: 10.22059/jitm.2026.108393
Salam Mohammed, Sajad, Abih, Aya R., Mutar, Modhi Lafta, Hameed, Asaad Shakir, Hammad, Saja Jumaa, Jawad Alataby, Ahmed K.. (1405). 'Implementing a Convolutional Neural Network with Adam Optimizer for Pneumonia Detection Through Chest X-ray Imaging', , 18(3), pp. 243-258. doi: 10.22059/jitm.2026.108393
Salam Mohammed, Sajad, Abih, Aya R., Mutar, Modhi Lafta, Hameed, Asaad Shakir, Hammad, Saja Jumaa, Jawad Alataby, Ahmed K.. Implementing a Convolutional Neural Network with Adam Optimizer for Pneumonia Detection Through Chest X-ray Imaging. , 1405; 18(3): 243-258. doi: 10.22059/jitm.2026.108393

Implementing a Convolutional Neural Network with Adam Optimizer for Pneumonia Detection Through Chest X-ray Imaging

Journal of Information Technology Management
دوره 18، شماره 3، 2026، صفحه 243-258    اصل مقاله (1.71 M)
نوع مقاله: Research Paper
شناسه دیجیتال (DOI): 10.22059/jitm.2026.108393
نویسندگان
Sajad Salam Mohammed1؛ Aya R. Abih2؛ Modhi Lafta Mutar3؛ Asaad Shakir Hameed* 4؛ Saja Jumaa Hammad5؛ Ahmed K. Jawad Alataby6
1Department of Electrical and Electronics Engineering, College of Engineering, University of Thi-Qar, Thi-Qar, Iraq
2University of Sumer, Thi-Qar, Iraq.
3Presidency of the University, Shatrah University, Thi-Qar, Iraq.
4Presidency of the University, Shatrah University, Thi-Qar, Iraq; Mazaya University College, Thi-Qar, Iraq.
5Immediate Ambulance Department, Al-Dour Technical Institute, Northern Technical University, Tikrit, Iraq
6General Directorate of Thi-Qar Education, Ministry of Education, Thi-Qar, Iraq.
چکیده
Pneumonia occurs frequently and can be life-threatening all over the world, mainly in locations where radiologists are hard to find. Even though chest X-ray imaging is widely used, correctly and promptly interpreting it is difficult due to human error and inadequate resources. It aims to overcome the problem of limited, reliable resources for advanced radiology by designing and evaluating a CNN model that can distinguish chest X-rays with and without pneumonia. To increase the usability of the model, the chest X-ray images were processed by being resized to a size of 224×224×3 and then enlarged through random rotation, flipping, and translation. The data was split so that 80 percent went to the training set and 20 percent to the testing set. Several layers were built into the custom CNN for overfitting prevention, such as convolution, pooling, batch normalization, ReLU, and dropout. For 20 epochs, I used Adam as the optimizer and set the mini-batch size to 32 and the learning rate to 1e-4. The model correctly distinguished between pneumonia and normal lung cases over 95% of the time. We found that the model was robust thanks to the confusion matrix and a close study of sample images after prediction. Microwave-based deep learning is introduced in this paper, which demonstrates strong accuracy in detecting pneumonia and makes it possible to use the tool as an automatic device for medical and remote healthcare screening.
کلیدواژه‌ها
Convolutional Neural Network (CNN)؛ Chest X-rays؛ Accuracy Alongside Precision Recall؛ ReLU؛ Confusion Matrix Evaluation
مراجع
آمار
تعداد مشاهده مقاله: 46
تعداد دریافت فایل اصل مقاله: 24