Exploring feature-based approaches in PET images for predicting cancer treatment outcomes

Naqa, Issam El; Grigsby, Perry W.; Apte, Aditya; Kidd, Elizabeth A.; Donnelly, Eric D.; Khullar, D.; Chaudhari, S; Yang, Deshan; Schmitt, Murilo Falleiros Lemos; Laforest, Richard; Thorstad, Wade L.; Deasy, Joseph O.

doi:10.1016/j.patcog.2008.08.011

Cited by 430 publications

(419 citation statements)

References 39 publications

Supporting

Mentioning

397

Contrasting

Unclassified

Order By: Relevance

“…The study focused on primary tumor volumes because of the limited resolution of PET images, which did not reproduce significant heterogeneity in small lymph nodes. On a similar note, El Naqa et al27 reported several logistic regression models of radiomic features, with good prediction power, for cervical cancer treatment outcomes. However, it was suggested that further testing and validation using large datasets is required.…”

Section: Discussionmentioning

confidence: 93%

“…We investigated the reproducibility of several subtypes of GLCM entropy feature as they were reported as one the highest reproducible and predictive radiomic features 21, 27, 30. We included: Entropy, Summation Entropy, Difference Entropy in addition to first‐order Entropy (Intensity Histogram Entropy) .…”

Section: Discussionmentioning

confidence: 99%

“…We calculated twenty common first‐order metrics such as the mean, standard deviation, maximum and minimum intensities, skewness, and kurtosis . We also studied other intensity–volume histogram (IVH)‐based features reported by El Naqa et al27 Examples of such features are V90 (volume percentage having at least intensity of 90%) and I90 (minimum intensity of 90% of the highest intensity volume).…”

Section: Methodsmentioning

confidence: 99%

“…Also, shape features (SF), which describe geometrical characteristics of tumors, have shown to provide a morphological characterization of PET uptake heterogeneity within a specified volume of interest 24, 25. Recent studies have emphasized on the higher discriminatory power of several radiomic features in comparison to SUV measurements regarding classification of tumor versus benign regions in lung, and head‐and‐neck patients,26 as well as for the prediction of cervical cancer treatment outcomes 27. Radiomic features were also reported as a significant tool to stage cervical cancer based on tumor heterogeneity information 28.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Reproducibility of F18‐FDG PET radiomic features for different cervical tumor segmentation methods, gray‐level discretization, and reconstruction algorithms

Altazi

Zhang

Fernandez

et al. 2017

J Applied Clin Med Phys

View full text Add to dashboard Cite

Site‐specific investigations of the role of radiomics in cancer diagnosis and therapy are emerging. We evaluated the reproducibility of radiomic features extracted from 18Flourine–fluorodeoxyglucose (18F‐FDG) PET images for three parameters: manual versus computer‐aided segmentation methods, gray‐level discretization, and PET image reconstruction algorithms. Our cohort consisted of pretreatment PET/CT scans from 88 cervical cancer patients. Two board‐certified radiation oncologists manually segmented the metabolic tumor volume (MTV1 and MTV2) for each patient. For comparison, we used a graphical‐based method to generate semiautomated segmented volumes (GBSV). To address any perturbations in radiomic feature values, we down‐sampled the tumor volumes into three gray‐levels: 32, 64, and 128 from the original gray‐level of 256. Finally, we analyzed the effect on radiomic features on PET images of eight patients due to four PET 3D‐reconstruction algorithms: maximum likelihood‐ordered subset expectation maximization (OSEM) iterative reconstruction (IR) method, fourier rebinning‐ML‐OSEM (FOREIR), FORE‐filtered back projection (FOREFBP), and 3D‐Reprojection (3DRP) analytical method. We extracted 79 features from all segmentation method, gray‐levels of down‐sampled volumes, and PET reconstruction algorithms. The features were extracted using gray‐level co‐occurrence matrices (GLCM), gray‐level size zone matrices (GLSZM), gray‐level run‐length matrices (GLRLM), neighborhood gray‐tone difference matrices (NGTDM), shape‐based features (SF), and intensity histogram features (IHF). We computed the Dice coefficient between each MTV and GBSV to measure segmentation accuracy. Coefficient values close to one indicate high agreement, and values close to zero indicate low agreement. We evaluated the effect on radiomic features by calculating the mean percentage differences (d¯) between feature values measured from each pair of parameter elements (i.e. segmentation methods: MTV1‐MTV2, MTV1‐GBSV, MTV2‐GBSV; gray‐levels: 64‐32, 64‐128, and 64‐256; reconstruction algorithms: OSEM‐FORE‐OSEM, OSEM‐FOREFBP, and OSEM‐3DRP). We used false|normald¯false| as a measure of radiomic feature reproducibility level, where any feature scored false|normald¯false| ±SD ≤ |25|% ± 35% was considered reproducible. We used Bland–Altman analysis to evaluate the mean, standard deviation (SD), and upper/lower reproducibility limits (U/LRL) for radiomic features in response to variation in each testing parameter. Furthermore, we proposed U/LRL as a method to classify the level of reproducibility: High— ±1% ≤ U/LRL ≤ ±30%; Intermediate— ±30% < U/LRL ≤ ±45%; Low— ±45 < U/LRL ≤ ±50%. We considered any feature below the low level as nonreproducible (NR). Finally, we calculated the interclass correlation coefficient (ICC) to evaluate the reliability of radiomic feature measurements for each parameter. The segmented volumes of 65 patients (81.3%) scored Dice coefficient >0.75 for all three volumes. The result outcomes revealed a tendency of higher radiomic fe...

show abstract

Section: Discussionmentioning

confidence: 93%

Section: Discussionmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Reproducibility of F18‐FDG PET radiomic features for different cervical tumor segmentation methods, gray‐level discretization, and reconstruction algorithms

Altazi

Zhang

Fernandez

et al. 2017

J Applied Clin Med Phys

View full text Add to dashboard Cite

show abstract

“…Such characterization provides additional and complementary PET image derived quantitative indices with potential value as already demonstrated in predicting therapy response or as prognostic factors in several cancers including lung [16], sarcoma [17], oesophageal [18,19] and rectal cancer [20]. A variety of methodologies has been proposed in order to assess intra-tumour uptake heterogeneity, including visual assessment [21], SUV coefficient of variation (SUVcov) [20], area under the curve of the cumulative histogram (CHAUC) [22] and textural features (TF) analysis [18].…”

mentioning

confidence: 99%