Computer extracted texture features on T2w MRI to predict biochemical recurrence following radiation therapy for prostate cancer

Ginsburg, Shoshana B.; Rusu, Mirabela; Kurhanewicz, John; Madabhushi, Anant

doi:10.1117/12.2043937

Cited by 10 publications

(16 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Radiomic features employed in this study were extracted on a per‐voxel basis from T 2 WI and ADC maps within the annotated cancerous regions. The choice of radiomic features for characterizing PCa lesions was motivated by their use in previous studies . We also included additional features that have recently been shown to be capable of distinguishing among subtly different disease types on imaging (Table ).…”

Section: Methodsmentioning

confidence: 99%

“…The choice of radiomic features for characterizing PCa lesions was motivated by their use in previous studies. 12,15,18,19,29 We also included additional features that have recently been shown to be capable of distinguishing among subtly different disease types on imaging 19 (Table 3). Intensity-based tumor heterogeneity was characterized using Haralick, Gabor, and Laws features; gradient-based tumor heterogeneity was characterized using CoLlAGe and gradient features.…”

Section: Radiomic Feature Extractionmentioning

confidence: 99%

See 1 more Smart Citation

Radiomic features from pretreatment biparametric MRI predict prostate cancer biochemical recurrence: Preliminary findings

Shiradkar

Ghose

Jambor

et al. 2018

Magnetic Resonance Imaging

Self Cite

115

View full text Add to dashboard Cite

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Radiomic Feature Extractionmentioning

confidence: 99%

Radiomic features from pretreatment biparametric MRI predict prostate cancer biochemical recurrence: Preliminary findings

Shiradkar

Ghose

Jambor

et al. 2018

Magnetic Resonance Imaging

Self Cite

115

View full text Add to dashboard Cite

show abstract

“…There was inter-fold variability regarding the optimal number of features. For the prostate, the optimal number of features varied between three [30%], 10 Names and description of the highest ranking features for the two ROIs, when ranked in the whole dataset, can be found in Table 4. The majority of the top five prostate features originated from the filtered images and for the margin from the no filter image.…”

Section: For a Representative Patientmentioning

confidence: 99%

“…By assessing tissue micro-architecture and tumour aggressiveness, being the latest by definition related with BCR, this modality could provide insight on recurrence risk prediction. Local recurrence after RT is reported to occur predominantly at the site of the index lesion [8], and imaging features from the primary tumour were found to strongly associate with the probability of BCR following RT [9,10]. However, these studies have relatively small and inhomogeneous patient cohorts, with the study by Gnep et al [9] having a median follow-up time of only four years.…”

Section: Introductionmentioning

confidence: 99%

Biochemical recurrence prediction after radiotherapy for prostate cancer with T2w magnetic resonance imaging radiomic features

Fernandes

Dinh

Walraven

et al. 2018

Physics and Imaging in Radiation Oncology

View full text Add to dashboard Cite

Background and purpose: High-risk prostate cancer patients are frequently treated with external-beam radiotherapy (EBRT). Of all patients receiving EBRT, 15-35% will experience biochemical recurrence (BCR) within five years. Magnetic resonance imaging (MRI) is commonly acquired as part of the diagnostic procedure and imaging-derived features have shown promise in tumour characterisation and biochemical recurrence prediction. We investigated the value of imaging features extracted from pre-treatment T2w anatomical MRI to predict five year biochemical recurrence in high-risk patients treated with EBRT. Materials and methods: In a cohort of 120 high-risk patients, imaging features were extracted from the wholeprostate and a margin surrounding it. Intensity, shape and textural features were extracted from the original and filtered T2w-MRI scans. The minimum-redundancy maximum-relevance algorithm was used for feature selection. Random forest and logistic regression classifiers were used in our experiments. The performance of a logistic regression model using the patient's clinical features was also investigated. To assess the prediction accuracy we used stratified 10-fold cross validation and receiver operating characteristic analysis, quantified by the area under the curve (AUC). Results: A logistic regression model built using whole-prostate imaging features obtained an AUC of 0.63 in the prediction of BCR, outperforming a model solely based on clinical variables (AUC = 0.51). Combining imaging and clinical features did not outperform the accuracy of imaging alone. Conclusions: These results illustrate the potential of imaging features alone to distinguish patients with an increased risk of recurrence, even in a clinically homogeneous cohort.

show abstract

“…The best overall classification result exceeded 99% and corresponded to the application of the SVM classifier. Ginsburg et al [15] tried to predict the probability of developing biochemical recurrence risk (associated with raised risk of metastases and prostate cancer-related mortality) following the radiation therapy. In their work, they evaluated the efficiency of different textural features, extracted from the T2-weighted images.…”

Section: Introductionmentioning

confidence: 99%

Texture Classification Using Gradient Features with Artificial Neural Network

Al-Kilidar¹,

George²

2020

Journal of Southwest Jiaotong University

View full text Add to dashboard Cite

Texture Analysis is the technique usinga small number of measurable features to represent complex textures. It provides many important discriminating characteristics that are related with extracting features and coding images which are used in various patterns recognition applications and classification texture. This research studies the extraction of discriminating characteristics for various texture images from using Absolute Gradient Matrix (AGM) and a comparative study of conventional texture-analysis method and the proposed methods. Four different sets of features have been considered: the first set consists of simple features extracted from the traditional AGM, the second set consists of new features which are derived using AGM and intensity values (pAGM); the third feature set consists of two sub-sets of features extracted from two AGM calculated using two displacement values and intensity values (CtdAGM). The fourth method implies the extracted set of AGM through Artificial Neural Network (ANN) for classification purpose. The considered method was applied on 13 classes of textures belonging to three sets of Salzburg Texture Image Database (i.e., bark, marble and woven fabric), each set holds 16 images per class, thus 208 images were tested in total. The attained average accuracy of classification proved that the proposed methods were superior to the conventional texture-analysis method with respect to classification accuracy and computational complexity.

show abstract

Computer extracted texture features on T2w MRI to predict biochemical recurrence following radiation therapy for prostate cancer

Cited by 10 publications

References 29 publications

Radiomic features from pretreatment biparametric MRI predict prostate cancer biochemical recurrence: Preliminary findings

Radiomic features from pretreatment biparametric MRI predict prostate cancer biochemical recurrence: Preliminary findings

Biochemical recurrence prediction after radiotherapy for prostate cancer with T2w magnetic resonance imaging radiomic features

Texture Classification Using Gradient Features with Artificial Neural Network

Contact Info

Product

Resources

About