Assessing the generalisability of radiomics features previously identified as predictive of radiation-induced sticky saliva and xerostomia

Berger, Thomas; Noble, D.; Yang, Zhuolin; Shelley, Leila E.A.; McMullan, Thomas; Bates, A.; Thomas, Simon; Carruthers, Linda; Beckett, George; Duffton, A.; Paterson, C.; Jena, R.; McLaren, Duncan; Burnet, N.G.; Nailon, William H.

doi:10.1016/j.phro.2022.12.001

Cited by 4 publications

(4 citation statements)

References 38 publications

(40 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Similar findings were reported in another study where GLCM features calculated on MRI were observed to be correlated with bladder wall changes during PCa radiotherapy [15]. In studies of radiation therapy for other cancers, short run emphasis from GLRLM and maximum CT intensity extracted on planning CT scans were found to significantly improve the prediction of xerostomia and sticky saliva in head and neck cancer patients [34,32]. Changes in first-order statistics (FOS) and GLCM features during oesophageal cancer radiotherapy were also reported to be correlated with G3 radiation pneumonitis (CTCAEv4).…”

Section: Discussionsupporting

confidence: 76%

See 1 more Smart Citation

Machine-learning with region-level radiomic and dosimetric features for predicting radiotherapy-induced rectal toxicities in prostate cancer patients

Yang¹,

Noble²,

Shelley³

et al. 2023

Radiotherapy and Oncology

Self Cite

View full text Add to dashboard Cite

Section: Discussionsupporting

confidence: 76%

“…The findings of the research are encouraging. However, their results are limited by the relatively small numbers of patients and by the fact that the IBSI processes, which promote reproducibility and standardisation across the field [32], were not followed [26].…”

Section: Discussionmentioning

confidence: 99%

Machine-learning with region-level radiomic and dosimetric features for predicting radiotherapy-induced rectal toxicities in prostate cancer patients

Yang¹,

Noble²,

Shelley³

et al. 2023

Radiotherapy and Oncology

Self Cite

View full text Add to dashboard Cite

“…This correlation enables the creation of a classification model capable of identifying at-risk patients [16]. Indeed, multiple studies have highlighted the utility of radiomics analysis in quantifying radiation therapy-induced damage in various organs, including the bladder, rectum, parotid, and lung [6,[17][18][19][20][21][22].…”

Section: Predictive Modeling For Radiation-induced Damagementioning

confidence: 99%

Multimodality radiomics prediction of radiotherapy-induced the early proctitis and cystitis in rectal cancer patients: a machine learning study

Abbaspour,

Barahman,

Abdollahi

et al. 2023

Biomed. Phys. Eng. Express

View full text Add to dashboard Cite

Purpose. This study aims to predict radiotherapy-induced rectal and bladder toxicity using computed tomography (CT) and magnetic resonance imaging (MRI) radiomics features in combination with clinical and dosimetric features in rectal cancer patients. Methods. A total of sixty-three patients with locally advanced rectal cancer who underwent three-dimensional conformal radiation therapy (3D-CRT) were included in this study. Radiomics features were extracted from the rectum and bladder walls in pretreatment CT and MR-T2W-weighted images. Feature selection was performed using various methods, including Least Absolute Shrinkage and Selection Operator (Lasso), Minimum Redundancy Maximum Relevance (MRMR), Chi-square (Chi2), Analysis of Variance (ANOVA), Recursive Feature Elimination (RFE), and SelectPercentile. Predictive modeling was carried out using machine learning algorithms, such as K-nearest neighbor (KNN), Support Vector Machine (SVM), Logistic Regression (LR), Decision Tree (DT), Random Forest (RF), Naive Bayes (NB), Gradient Boosting (XGB), and Linear Discriminant Analysis (LDA). The impact of the Laplacian of Gaussian (LoG) filter was investigated with sigma values ranging from 0.5 to 2. Model performance was evaluated in terms of the area under the receiver operating characteristic curve (AUC), accuracy, precision, sensitivity, and specificity. Results. A total of 479 radiomics features were extracted, and 59 features were selected. The pre-MRI T2W model exhibited the highest predictive performance with an AUC: 91.0/96.57%, accuracy: 90.38/96.92%, precision: 90.0/97.14%, sensitivity: 93.33/96.50%, and specificity: 88.09/97.14%. These results were achieved with both original image and LoG filter (sigma = 0.5–1.5) based on LDA/DT-RF classifiers for proctitis and cystitis, respectively. Furthermore, for the CT data, AUC: 90.71/96.0%, accuracy: 90.0/96.92%, precision: 88.14/97.14%, sensitivity: 93.0/96.0%, and specificity: 88.09/97.14% were acquired. The highest values were achieved using XGB/DT-XGB classifiers for proctitis and cystitis with LoG filter (sigma = 2)/LoG filter (sigma = 0.5–2), respectively. MRMR/RFE-Chi2 feature selection methods demonstrated the best performance for proctitis and cystitis in the pre-MRI T2W model. MRMR/MRMR-Lasso yielded the highest model performance for CT. Conclusion. Radiomics features extracted from pretreatment CT and MR images can effectively predict radiation-induced proctitis and cystitis. The study found that LDA, DT, RF, and XGB classifiers, combined with MRMR, RFE, Chi2, and Lasso feature selection algorithms, along with the LoG filter, offer strong predictive performance. With the inclusion of a larger training dataset, these models can be valuable tools for personalized radiotherapy decision-making.

show abstract

“…Radiomics analysis involved the parotid glands, cochlea, masticatory muscles and brain white matter as VOIs. The authors consider the results encouraging, needing additional validation in order to be able to be implemented as a tool for predicting the toxicities associated with irradiation (22)(23)(24).…”

Section: Radiomics and Treatment Related Toxicitiesmentioning

confidence: 99%

Radiomics Breakthrough Could Spark the Head and Neck Cancer Radiotherapy Revolution

Mireștean¹,

Iancu²,

Iancu³

2023

JMRO

View full text Add to dashboard Cite

Radiomics, the method by which digital images could be transformed into mineable data, opens new horizons for biomedical research and in particular in oncology, for diagnostic, predictive and prognostic purposes. The use of artificial intelligence (AI) algorithms in the radiomics algorithm makes radiomics and AI two inseparable, intricate domains. AI defined as machine capability of imitating human intelligence, has already been implemented on a large scale in oncology and radiotherapy. One of the two main branches (the virtual one) of machine learning depending on the application, artificial intelligence is involved both in the diagnostics processes as well as treatment planning, – dose delivery and radiotherapy quality assurance (QA). Head and neck cancer (HNC), although it is the 6th malignancy in incidence worldwide, is redoubtable due to the high rate of therapeutic failures, especially of loco-regional recurrence. Although intensity-modulated treatment techniques have brought benefits especially in limiting the toxicities associated with irradiation, AI and especially radiomics, due the possibility to extract data from high-resolution medical imaging in order to build predictive diagnostic and prognostic models, could upgrade the technological revolution in HNC radiotherapy at a higher level. Beyond the already intensively studied diagnostic applications, radiomics could be useful for predicting the response to radio-chemotherapy, anticipating treatment related toxicities and for pre-therapeutic evaluation of the need for adaptive radiotherapy (ART). Clinical-radiomic models have superior predictive power and the delta variation of radiomic features could be a biomarker still less evaluated. Due to characteristics of modern radiotherapy which includes as standard the image guided radiotherapy (IGRT) concept using the computer tomography (CT) simulator and Cone Beam CT (CBCT) to ensure the accuracy of the patient’s positioning during the treatment, radiomics in radiotherapy could be the spearhead of the translation radiomics in daily clinical routine and of the HNC RGRT concept development.

show abstract

Assessing the generalisability of radiomics features previously identified as predictive of radiation-induced sticky saliva and xerostomia

Cited by 4 publications

References 38 publications

Machine-learning with region-level radiomic and dosimetric features for predicting radiotherapy-induced rectal toxicities in prostate cancer patients

Machine-learning with region-level radiomic and dosimetric features for predicting radiotherapy-induced rectal toxicities in prostate cancer patients

Multimodality radiomics prediction of radiotherapy-induced the early proctitis and cystitis in rectal cancer patients: a machine learning study

Radiomics Breakthrough Could Spark the Head and Neck Cancer Radiotherapy Revolution

Contact Info

Product

Resources

About