Enhancement of multimodality texture-based prediction models via optimization of PET and MR image acquisition protocols: a proof of concept

Abstract

Texture-based radiomic models constructed from medical images have the potential to support cancer treatment management via personalized assessment of tumour aggressiveness. While the identification of stable texture features under varying imaging settings is crucial for the translation of radiomics analysis into routine clinical practice, we hypothesize in this work that a complementary optimization of image acquisition parameters prior to texture feature extraction could enhance the predictive performance of texture-based radiomic models. As a proof of concept, we evaluated the possibility of enhancing a model constructed for the early prediction of lung metastases in soft-tissue sarcomas by optimizing PET and MR image acquisition protocols via computerized simulations of image acquisitions with varying parameters. Simulated PET images from 30 STS patients were acquired by varying the extent of axial data combined per slice ('span'). Simulated T -weighted and T-weighted MR images were acquired by varying the repetition time and echo time in a spin-echo pulse sequence, respectively. We analyzed the impact of the variations of PET and MR image acquisition parameters on individual textures, and we investigated how these variations could enhance the global response and the predictive properties of a texture-based model. Our results suggest that it is feasible to identify an optimal set of image acquisition parameters to improve prediction performance. The model constructed with textures extracted from simulated images acquired with a standard clinical set of acquisition parameters reached an average AUC of [Formula: see text] in bootstrap testing experiments. In comparison, the model performance significantly increased using an optimal set of image acquisition parameters ([Formula: see text]), with an average AUC of [Formula: see text]. Ultimately, specific acquisition protocols optimized to generate superior radiomics measurements for a given clinical problem could be developed and standardized via dedicated computer simulations and thereafter validated using clinical scanners.

Radiomics analysis of apparent diffusion coefficient in cervical cancer: A preliminary study on histological grade evaluation

“…Radiomics analysis is easy to retrospectively perform on available imaging data; however, as there was heterogeneity in the MR acquisition parameters across multiple institutions, and even at a single institution, there might be changes in protocols over time. Variations in acquisition and image reconstruction parameters can introduce changes that are not due to underlying biologic effects, ie, the number of acquisitions, repetition time, echo time, and sampling bandwidth in T 2 imaging has been shown to affect texture analysis . To reduce bias and variance, we conducted this study prospectively and extracted all radiomics features from the same MRI unit using the same protocol at our institution.…”

“…Variations in acquisition and image reconstruction parameters can introduce changes that are not due to underlying biologic effects, ie, the number of acquisitions, repetition time, echo time, and sampling bandwidth in T 2 imaging has been shown to affect texture analysis. 31,32 To reduce bias and variance, we conducted this study prospectively and extracted all radiomics features from the same MRI unit using the same protocol at our institution. Another especially critical and challenging component of radiomics is tumor segmentation.…”

NCTN Assessment on Current Applications of Radiomics in Oncology

“…McGurk et al augmented PET images from 30 patients with soft tissue sarcoma by varying the extent of axial data combined per slice ("span"). 130 Simulated T1-weighted and T2-weighted MRI scans were acquired by varying the repetition time and echo time in a spin-echo pulse sequence, respectively. The impact of PET and MRI acquisition parameter variation on individual textures was investigated to assess the global response and the predictive properties of a texture-based model.…”

“…The results suggested that such a process is feasible for identifying an optimal set of image acquisition parameters to improve prediction performance. 130 Overall, these radiomics phantoms may be helpful in assessing the interscanner and intrascanner variabilities, and thus protocols for regular phantom quality assurance may be worth developing, similar to dosimetric study in current radiation therapy quality assurance programs, to monitor interscan and intervendor variability of imagederived features. In addition, phantom studies will also be useful in optimizing imaging protocols and image pre-/ postprocessing techniques that allow for reliable radiomics characterization.…”

<p>Application of Radiomics for Personalized Treatment of Cancer Patients</p>

“…In addition to the micro-heterogeneity associated with cell imaging and molecular markers of tumors based on medical imaging, there are recent categorizations and analyses of tumors and phenotypes, 12,13 localized regional modeling, 14 and applications such as prediction of future results. 15 Other, more quantitative, imaging models for specific tumor sites include head and neck, 16,17 lung, 18 breast, 19 liver, 20 cervix, 21 prostate, 22 limbs (sarcoma), 23 and the brain. 24 In the past five years, the field of radiology has garnered much more attention from experts in other fields, and the number of medical imaging publications has grown exponentially.…”