Evaluation of prognostic models developed using standardised image features from different PET automated segmentation methods

Parkinson, Craig; Foley, Kieran; Whybra, Philip; Hills, Robert Kerrin; Roberts, Ashley; Marshall, Christopher; Staffurth, John; Spezi, Emiliano

doi:10.1186/s13550-018-0379-3

Cited by 9 publications

(12 citation statements)

References 48 publications

(50 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This study re-uses highly curated patient data from previous studies 15,16 . The patients all had biopsy proven oesophageal cancer, either adenocarcinoma or squamous cell carcinoma (SCC), and underwent PET/CT as part of the routine diagnostic staging pathway.…”

Section: Methodsmentioning

confidence: 99%

“…Features were extracted from the segmented volumes using the Spaarc Pipeline for Automated Analysis and Radiomics Computing (SPAARC), an in-house software built on the MATLAB platform (MathWorks, Natick, MA, USA). SPAARC implements a wide range of solutions for processing imaging and radiotherapy data and utilises algorithms validated against a digital phantom as part of the IBSI international collaboration 13,16,19 . Full imaging feature names are found in supplementary materials.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Assessing radiomic feature robustness to interpolation in 18F-FDG PET imaging

Whybra

Parkinson

Foley

et al. 2019

Sci Rep

Self Cite

View full text Add to dashboard Cite

Radiomic studies link quantitative imaging features to patient outcomes in an effort to personalise treatment in oncology. To be clinically useful, a radiomic feature must be robust to image processing steps, which has made robustness testing a necessity for many technical aspects of feature extraction. We assessed the stability of radiomic features to interpolation processing and categorised features based on stable, systematic, or unstable responses. Here, 18 F-fluorodeoxyglucose ( 18 F-FDG) PET images for 441 oesophageal cancer patients (split: testing = 353, validation = 88) were resampled to 6 isotropic voxel sizes (1.5 mm, 1.8 mm, 2.0 mm, 2.2 mm, 2.5 mm, 2.7 mm) and 141 features were extracted from each volume of interest (VOI). Features were categorised into four groups with two statistical tests. Feature reliability was analysed using an intraclass correlation coefficient (ICC) and patient ranking consistency was assessed using a Spearman’s rank correlation coefficient ( ρ ). We categorised 93 features robust and 6 limited robustness (stable responses), 34 potentially correctable (systematic responses), and 8 not robust (unstable responses). We developed a correction technique for features with potential systematic variation that used surface fits to link voxel size and percentage change in feature value. Twenty-nine potentially correctable features were re-categorised to robust for the validation dataset, after applying corrections defined by surface fits generated on the testing dataset. Furthermore, we found the choice of interpolation algorithm alone (spline vs trilinear) resulted in large variation in values for a number of features but the response categorisations remained constant. This study attempted to quantify the diverse response of radiomics features commonly found in 18 F-FDG PET clinical modelling to isotropic voxel size interpolation.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Assessing radiomic feature robustness to interpolation in 18F-FDG PET imaging

Whybra

Parkinson

Foley

et al. 2019

Sci Rep

Self Cite

View full text Add to dashboard Cite

show abstract

“…Our results show that the performance of individual automated segmentation methodologies can be enhanced through the use of machine-learning techniques. Studies have shown how imaging parameters such as reconstruction settings [41] as well as tumour features impact TVD [31,[42][43][44]. The ATLAAS statistical model, however, was developed using tumour characteristics which have been demonstrated to be classifiers for accurate MTV delineation.…”

Section: Resultsmentioning

confidence: 99%

Machine-learned target volume delineation of 18F-FDG PET images after one cycle of induction chemotherapy

et al. 2019

Self Cite

View full text Add to dashboard Cite

Biological tumour volume (GTVPET) delineation on 18 F-FDG PET acquired during induction chemotherapy (ICT) is challenging due to the reduced metabolic uptake and volume of the GTVPET. Automatic segmentation algorithms applied to 18 F-FDG PET (PET-AS) imaging have been used for GTVPET delineation on 18 F-FDG PET imaging acquired before ICT. However, their role has not been investigated in 18 F-FDG PET imaging acquired after ICT. In this study we investigate PET-AS techniques, including ATLAAS a machine learned method, for accurate delineation of the GTVPET after ICT. Twenty patients were enrolled onto a prospective phase I study (FiGaRO). PET/CT imaging was acquired at baseline and 3 weeks following 1 cycle of induction chemotherapy. The GTVPET was manually delineated by a nuclear medicine physician and clinical oncologist. The resulting GTVPET was used as the reference contour. The ATLAAS original statistical model was expanded to include images of reduced metabolic activity and the ATLAAS algorithm was retrained on the new reference dataset. Estimated GTVPET contours were derived using sixteen PET-AS methods and compared to the GTVPET using the Dice Similarity Coefficient (DSC). The mean DSC for ATLAAS, 60% Peak Thresholding (PT60), Adaptive Thresholding (AT) and Watershed Thresholding (WT) was 0.72, 0.61, 0.63 and 0.60 respectively. The GTVPET generated by ATLAAS compared favourably with manually delineated volumes and in comparison, to other PET-AS methods, was more accurate for GTVPET delineation after ICT. ATLAAS would be a feasible method to reduce inter-observer variability in multi-centre trials.

show abstract

“…In addition to patient specific clinicopathological information, medical image analysis is gaining substantial interest in prognostic research. In oncology, the SUV of 18-fluorodeoxyglucose (FDG), which measures FDG activity in the tumor and correlates with viable tumor cell number and metabolism, obtained from FDG-PET scans is widely used for diagnosing, staging, monitoring response to therapy, as well as outcome prediction [16,17,18]. Correlation between higher SUVmax and worse survival has been reported in several studies [19,20,21].…”

Section: Discussionmentioning

confidence: 99%

“…This approach as a whole is named “radiomics” [24]. Several studies have successfully developed radiomic prognostic classifiers that can be associated with metastatic recurrence and survival in several types of cancers [3,4,16]. For example, Lambin et al demonstrated that the prognostic model based on pretreatment CT radiomic features yielded an AUC of 0.69 in predicting 3-year overall survival of esophageal cancer patients after chemoradiotherapy [25].…”

Section: Discussionmentioning

confidence: 99%

Deep Convolutional Neural Network-Based Positron Emission Tomography Analysis Predicts Esophageal Cancer Outcome

Yang¹,

Yeh²,

Yu³

et al. 2019

JCM

View full text Add to dashboard Cite

In esophageal cancer, few prediction tools can be confidently used in current clinical practice. We developed a deep convolutional neural network (CNN) with 798 positron emission tomography (PET) scans of esophageal squamous cell carcinoma and 309 PET scans of stage I lung cancer. In the first stage, we pretrained a 3D-CNN with all PET scans for a task to classify the scans into esophageal cancer or lung cancer. Overall, 548 of 798 PET scans of esophageal cancer patients were included in the second stage with an aim to classify patients who expired within or survived more than one year after diagnosis. The area under the receiver operating characteristic curve (AUC) was used to evaluate model performance. In the pretrain model, the deep CNN attained an AUC of 0.738 in identifying patients who expired within one year after diagnosis. In the survival analysis, patients who were predicted to be expired but were alive at one year after diagnosis had a 5-year survival rate of 32.6%, which was significantly worse than the 5-year survival rate of the patients who were predicted to survive and were alive at one year after diagnosis (50.5%, p < 0.001). These results suggest that the prediction model could identify tumors with more aggressive behavior. In the multivariable analysis, the prediction result remained an independent prognostic factor (hazard ratio: 2.830; 95% confidence interval: 2.252–3.555, p < 0.001). We conclude that a 3D-CNN can be trained with PET image datasets to predict esophageal cancer outcome with acceptable accuracy.

show abstract

Evaluation of prognostic models developed using standardised image features from different PET automated segmentation methods

Cited by 9 publications

References 48 publications

Assessing radiomic feature robustness to interpolation in 18F-FDG PET imaging

Assessing radiomic feature robustness to interpolation in 18F-FDG PET imaging

Machine-learned target volume delineation of 18F-FDG PET images after one cycle of induction chemotherapy

Deep Convolutional Neural Network-Based Positron Emission Tomography Analysis Predicts Esophageal Cancer Outcome

Contact Info

Product

Resources

About