Impact of Image Contrast Enhancement on Stability of Radiomics Feature Quantification on a 2D Mammogram Radiograph

Radzi, Siti Fairuz Mat; Karim, Muhammad Khalis Abdul; Saripan, M. Iqbal; Rahman, Mohd Amiruddin Abd; Osman, Nurul Huda; Dalah, Entesar Zawam; Noor, Noramaliza Mohd

doi:10.1109/access.2020.3008927

Cited by 31 publications

(18 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Hasil pengujian menyatakan CLAHE lebih bagus dibandingkan dengan AHE dan kelompok manual. Metode ini harus diterapkan untuk memprediksi hasil pada pasien dengan kanker payudara [14].…”

Section: Penelitian Meningkatkan Kontras Gambar Radiografi Menggunaka...unclassified

Identifikasi Objek pada Citra Thorax X-Ray Pasien COVID-19 dengan Metode Contrast Limited Adaptive Histogram Equalization (CLAHE)

Putra

Na`

Yuhandri

2022

jidt

View full text Add to dashboard Cite

Radiografi thorax X-Ray menghasilkan citra digital radiografi diarea dada seperti paru-paru, jantung, dan tulang rusuk. Citra ini dapat memvisualisasikan kondisi paru-paru pasien COVID-19. Pemeriksaan kondisi paru-paru pasien COVID-19 dengan X-Ray lebih mudah, murah dan banyak tersedia di rumah sakit dibandingkan teknik radiografi yang lainnya. Namun, hasil citra digital radiografi X-Ray memiliki kualitas yang kurang bagus sehingga perlu ditingkatkan. Kontras citra yang rendah menjadi faktor kesulitan identifikasi citra thorax pasien COVID-19. Meningkatkan kontras citra thorax pasien COVID-19 yang rendah dengan Contrast Limited Adaptive Histogram Equalization (CLAHE) sehingga mudah diamati secara konkret dan lebih jelas. Citra yang diolah dalam penelitian ini sebanyak 100 citra thorax pasien COVID-19 yang bersumber dari komputer bagian radiologi RS Bhayangkara Padang Indonesia. Selanjutnya citra diolah menggunakan pengolahan citra digital menggunakan software Matlab. Tahapan pengolahannya citra thorax dikonversi ke bentuk grayscale. Citra hasil dari grayscale dilanjutkan dengan pengolahan kontras dengan menggunakan metode CLAHE bertipe distribusi Uniform, Rayleigh dan Exponential. Dilanjutkan penghitungan nilai Peak Signal to Noise Ratio (PNSR) dan Mean Square Error (MSE) hasil citra dari pengolahan metode masing-masing tipe CLAHE. Hasil dari pengujian semua citra dapat ditingkatkan kualitas kontrasnya secara visual. Hasil rata-rata MSE CLAHE Uniform, Rayleigh dan Exponential secara berturut-turut sebesar 27.68, 25.86 dan 26.33. Nilai rata-rata PNSR CLAHE Uniform, Rayleigh dan Exponential > 30 dB yaitu sebesar 112.32 dB, 171.95 dB dan 151.90 dB yang berarti citra CLAHE mirip dengan citra aslinya. CLAHE Rayleigh memberikan hasil terbaik secara kualitas dan kuantitas dengan jumlah 85 citra atau nilai akurasi 85%, sedangkan CLAHE Exponential dan CLAHE Uniform sebesar 15% dan 0%.

show abstract

“…Hasil pengujian menyatakan CLAHE lebih bagus dibandingkan dengan AHE dan kelompok manual. Metode ini harus diterapkan untuk memprediksi hasil pada pasien dengan kanker payudara [14].…”

Section: Penelitian Meningkatkan Kontras Gambar Radiografi Menggunaka...unclassified

Identifikasi Objek pada Citra Thorax X-Ray Pasien COVID-19 dengan Metode Contrast Limited Adaptive Histogram Equalization (CLAHE)

Putra

Na`

Yuhandri

2022

jidt

View full text Add to dashboard Cite

show abstract

“…The mammogram images were enhanced by Contrast Limited Adaptive Histogram Equalization (CLAHE) to improve the quality of the image for better visual and computational analysis before the segmentation process [22,23]. The Active Contour Model (ACM) technique is a semiautomatic iterative region-growing image segmentation algorithm, and the iteration has been set to 200 for every mammogram image.…”

Section: Semiautomatic Segmentation For Region-of-interest (Roi)mentioning

confidence: 99%

Hyperparameter Tuning and Pipeline Optimization via Grid Search Method and Tree-Based AutoML in Breast Cancer Prediction

Radzi

Karim

Saripan

et al. 2021

JPM

Self Cite

View full text Add to dashboard Cite

Automated machine learning (AutoML) has been recognized as a powerful tool to build a system that automates the design and optimizes the model selection machine learning (ML) pipelines. In this study, we present a tree-based pipeline optimization tool (TPOT) as a method for determining ML models with significant performance and less complex breast cancer diagnostic pipelines. Some features of pre-processors and ML models are defined as expression trees and optimal gene programming (GP) pipelines, a stochastic search system. Features of radiomics have been presented as a guide for the ML pipeline selection from the breast cancer data set based on TPOT. Breast cancer data were used in a comparative analysis of the TPOT-generated ML pipelines with the selected ML classifiers, optimized by a grid search approach. The principal component analysis (PCA) random forest (RF) classification was proven to be the most reliable pipeline with the lowest complexity. The TPOT model selection technique exceeded the performance of grid search (GS) optimization. The RF classifier showed an outstanding outcome amongst the models in combination with only two pre-processors, with a precision of 0.83. The grid search optimized for support vector machine (SVM) classifiers generated a difference of 12% in comparison, while the other two classifiers, naïve Bayes (NB) and artificial neural network—multilayer perceptron (ANN-MLP), generated a difference of almost 39%. The method’s performance was based on sensitivity, specificity, accuracy, precision, and receiver operating curve (ROC) analysis.

show abstract

“…The result obtained from the experiment shows that TPOTbased selection produces an Auto ML pipeline that performed a conventional grid search for KNN selected model. The important is the consideration of features score in the model [5]. AutoML performance was given in Table 2.…”

Section: B Cad4covid and Automl Performancementioning

confidence: 99%

Detection of Novel Coronavirus from Chest X-Ray Radiograph Images via Automated Machine Learning and CAD4COVID

Izdihar

Karim

Aresli

et al. 2021

2021 International Congress of Advanced Technology and Engineering (ICOTEN)

View full text Add to dashboard Cite

Recently, Artificial Intelligence (AI) has been considered as a valuable tool to detect early infections and to monitor the condition of the infected patients. Machine learning and deep learning is a subset of AI that uses neural network algorithms. Hence, this study aimed to explore the sensitivity of CoV-19 detection by using CAD4COVID program (Delft Imaging, Netherland), and to evaluate the accuracy of the classifier performance using Automated Machine Learning (Auto ML) algorithm. 70 chest X-ray (CXR) images were assessed and of that, 39, 20 and 11 patients receive low range (0-35), medium range (36-65) and high range (66-100) probability score, respectively. The sensitivity of AutoML detection was 0.99, with an accuracy of 0.83. In summary, the AutoML with the best optimizer may comparable to CAD4COVID in detection of Cov-19 in term of its accuracy and sensitivity.

show abstract

Impact of Image Contrast Enhancement on Stability of Radiomics Feature Quantification on a 2D Mammogram Radiograph

Cited by 31 publications

References 31 publications

Identifikasi Objek pada Citra Thorax X-Ray Pasien COVID-19 dengan Metode Contrast Limited Adaptive Histogram Equalization (CLAHE)

Identifikasi Objek pada Citra Thorax X-Ray Pasien COVID-19 dengan Metode Contrast Limited Adaptive Histogram Equalization (CLAHE)

Hyperparameter Tuning and Pipeline Optimization via Grid Search Method and Tree-Based AutoML in Breast Cancer Prediction

Detection of Novel Coronavirus from Chest X-Ray Radiograph Images via Automated Machine Learning and CAD4COVID

Contact Info

Product

Resources

About