DenseNet-CNN Architectural Model for Detection of Abnormality in Acute Pulmonary Edema

Cynthia Hayat

DOI: https://doi.org/10.23917/khif.v7i2.13455

Abstract

Acute pulmonary edema (EPA) is a condition of emergency respiratory distress that results from the sudden and rapid build-up of fluid into the lungs. Rapid screening of EPA patients is necessary so that radiologists can make the prognosis as early as possible. In addition, reliance on the expert's knowledge of reasoning also hinders the diagnostic process. This research proceeded by developing an architectural model for machine learning systems with a deep learning approach. With the concept of representative learning, the denseNet-CNN algorithm connects each layer to another utilizing a feed-forward. The data used is Image CXR-14 that is specifically labeled pulmonary edema pathology. Each CXR-14 image is 1024 × 1024 in size with a value of 8 bits grayscale. The architectural model development consists of several stages: the preparation stage, data resampling, data training, and data testing. Optimizer parameters used are Adam's optimizer, a learning rate of 0.0001, weight decay = 1e-5, and the loss used is binary cross-entropy. The resulting mean of AUROC analysis showed that the sensitivity value of the 10% dataset was 71.493%, and the specificity value of 10.011% was obtained at the second hold of the k-fold cross-validation method after holdout validation so that the resulting model was valid. The detection system developed from the denseNet-CNN model is to expectedly help radiologists identify abnormalities in CXR images quickly, precisely, and consistently. The development of the denseNet CNN model is in the form of a heatmap visualization by localizing the features one is watching. With localization in the heat map form, pathological abnormalities detection of PEA is easier to do and recognize.

Keywords

denseNet-CNN; acute edema; abnormality detection

Full Text:

PDF

References

S. H. Rampengan, “EDEMA PARU KARDIOGENIK AKUT,” J. BIOMEDIK, 2014, doi: 10.35790/jbm.6.3.2014.6320.

H. Nendrastuti and M. Soetomo, “Edema Paru Akut Kardiogenik Dan Non Kardiogenik,” Maj. Kedokt. Respirasi, 2010.

M. Irawaty, “Penatalaksanaan Edema Paru pada Kasus VSD dan Sepsis VAP Treatment of Lung Oedema in VSD and VAP Sepsis,” Anest. Crit. Care, 2010.

A. Jatu and Lusiana, “Peranan Epitel Alveoli pada Edema Paru Non-kardiogenik,” Ckd, 2015.

C. Hayat and B. Abian, “The modeling of artificial neural network of early diagnosis for malnutrition with backpropagation method,” 2018, doi: 10.1109/IAC.2018.8780505.

K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image recognition,” 2016, doi: 10.1109/CVPR.2016.90.

A. K. Jaiswal, P. Tiwari, S. Kumar, D. Gupta, A. Khanna, and J. J. P. C. Rodrigues, “Identifying pneumonia in chest X-rays: A deep learning approach,” Meas. J. Int. Meas. Confed., 2019, doi: 10.1016/j.measurement.2019.05.076.

G. Huang, S. Liu, L. Van Der Maaten, and K. Q. Weinberger, “CondenseNet: An Efficient DenseNet Using Learned Group Convolutions,” 2018, doi: 10.1109/CVPR.2018.00291.

G. Huang, Z. Liu, L. Van Der Maaten, and K. Q. Weinberger, “Densely connected convolutional networks,” 2017, doi: 10.1109/CVPR.2017.243.

T. NURHIKMAT, “IMPLEMENTASI DEEP LEARNING UNTUK IMAGE CLASSIFICATION MENGGUNAKAN ALGORITMA CONVOLUTIONAL NEURAL NETWORK (CNN) PADA CITRA WAYANG GOLEK,” UNIVERSITAS ISLAM INDONESIA. 2018.

S. Albawi, T. A. Mohammed, and S. Al-Zawi, “Understanding of a convolutional neural network,” 2018, doi: 10.1109/ICEngTechnol.2017.8308186.

X. Wang, Y. Peng, L. Lu, Z. Lu, M. Bagheri, and R.

M. Summers, “ChestX-ray8: Hospital-scale chest X-ray database and benchmarks on weakly-supervised classification and localization of common thorax diseases,” 2017, doi: 10.1109/CVPR.2017.369.

E. T. Nader, “Chest X-ray interpretation,” in Perioperative Assessment of the Maxillofacial Surgery Patient: Problem-based Patient Management, 2018.

G. Huang, S. Liu, L. Van Der Maaten, and K. Q. Weinberger, “CondenseNet: An efficient densenet using learned group convolutions,” arXiv. 2017.

S. Ramiz and M. Rajpurkar, “Pulmonary Embolism in Children,” Pediatric Clinics of North America. 2018, doi: 10.1016/j.pcl.2018.02.002.

Y. Bengio, “Learning deep architectures for AI,” Found. Trends Mach. Learn., 2009, doi: 10.1561/2200000006.

P. Rajpurkar et al., “CheXNet: Radiologist-level pneumonia detection on chest X-rays with deep learning,” arXiv. 2017.

N. Kalchbrenner, E. Grefenstette, and P. Blunsom, “A convolutional neural network for modelling sentences,” 2014, doi: 10.3115/v1/p14-1062.

Y. Zhang, J. Gao, and H. Zhou, “Breeds Classification with Deep Convolutional Neural Network,” 2020, doi: 10.1145/3383972.3383975.

C. Neural and N. Accelerator, “ISAAC : A Convolutional Neural Network Accelerator with I n- S itu A nalog A rithmetic in C rossbars,” Iscas, 2016.

R. M. PRASMATIO, B. Rahmat, and I. Yuniar, “Deteksi Dan Pengenalan Ikan Menggunakan Algoritma Convolutional Neural Network,” J. Inform. dan Sist. Inf., 2020.

Article Metrics

Abstract view(s): 61 time(s)
PDF: 36 time(s)

Refbacks

  • There are currently no refbacks.