Radiomic analysis has exponentially increased the amount of quantitative data that can be extracted from a single image.These imaging biomarkers can aid in the generation of prediction models aimed to further personal...Radiomic analysis has exponentially increased the amount of quantitative data that can be extracted from a single image.These imaging biomarkers can aid in the generation of prediction models aimed to further personalized medicine.However,the generalizability of the model is dependent on the robustness of these features.The purpose of this study is to review the current literature regarding robustness of radiomic features on magnetic resonance imaging.Additionally,a phantom study is performed to systematically evaluate the behavior of radiomic features under various conditions(signal to noise ratio,region of interest delineation,voxel size change and normalization methods)using intraclass correlation coefficients.The features extracted in this phantom study include first order,shape,gray level cooccurrence matrix and gray level run length matrix.Many features are found to be non-robust to changing parameters.Feature robustness assessment prior to feature selection,especially in the case of combining multi-institutional data,may be warranted.Further investigation is needed in this area of research.展开更多
Background An early identification of the composition of arterial thrombus may have diagnostic, therapeutic, and prognostic implications. The variation of magnetic resonance （MR） signal intensity between white and r...Background An early identification of the composition of arterial thrombus may have diagnostic, therapeutic, and prognostic implications. The variation of magnetic resonance （MR） signal intensity between white and red thrombi, especially in the susceptibility sensitive MR sequence, remains unknown. Our research was to evaluate the feasibility of MRI in differentiating of white and red thrombi with a phantom study. Methods A total of 12 red and 12 white thrombi were prepared with the venous blood. Examination of the phantom was completed using a 3.0T MR unit, including fluid attenuated inversion recovery （FLAIR） T1, T2-weighted imaging （T2WI）, FLAIR T2, T2＊ gradient echo （T2＊GRE） imaging, and susceptibility weighted angiography sequences （SWAN）. MR signal intensity patterns of the thrombi were objectively classified as hyperintensity, isointensity and hypointensity, compared with the background agar. The volume of thrombus was calculated and correlated with its signal intensity. Results For white thrombi, 11/12 clots showed hyperintensity and 1/12 showed isointensity in FLAIR T1 images. In T2WI, 6/12 clots showed hyperintensity, 3/12 isointensity, and 3/12 hypointensity. In FLAIR T2, 8/12 clots showed hyperintensity and 4/12 showed isointensity. In T2＊GRE, 3/12 clots showed hyperintensity and the remaining 9/12 clots showed isointensity. In SWAN, 5/12 clots demonstrated hyperintensity and 7/12 isointensity. For the red thrombus, 12/12 clots demonstrated hyperintensity in FLAIR T1, T2WI, and FLAIR T2 sequences. In T2＊GRE and SWAN sequences, 3/12 clots displayed hypointensity and the remaining 9/12 clots showed slight hyperintensity. Thrombi with hypointensity displayed in T2＊GRE and SWAN sequences were significantly larger than those with hyperintensity. Conclusions Differentiation of white and red thrombi with conventional MR sequence is unreliable, because both kinds of thrombi do not possess unique signal intensity features in these sequences. Red thrombus may or may not show hypointensity in the susceptibility sensitive MR sequences, depending on its size and time course.展开更多
People with diabetes are 30 times more likely to undergo amputation than the general population.A total of 3.2million people in the UK have been diagnosed with diabetes and by 2025 this number is estimated to be 5mill...People with diabetes are 30 times more likely to undergo amputation than the general population.A total of 3.2million people in the UK have been diagnosed with diabetes and by 2025 this number is estimated to be 5million.One potential complication post amputation is phantom limb pain(PLP),a prevalent and often chronic condition which is difficult to treat.To inform a展开更多
基金This work is in part funded by Walk-for-Beauty Foundation and Carol M.Baldwin Breast Cancer Research Foundation.
文摘Radiomic analysis has exponentially increased the amount of quantitative data that can be extracted from a single image.These imaging biomarkers can aid in the generation of prediction models aimed to further personalized medicine.However,the generalizability of the model is dependent on the robustness of these features.The purpose of this study is to review the current literature regarding robustness of radiomic features on magnetic resonance imaging.Additionally,a phantom study is performed to systematically evaluate the behavior of radiomic features under various conditions(signal to noise ratio,region of interest delineation,voxel size change and normalization methods)using intraclass correlation coefficients.The features extracted in this phantom study include first order,shape,gray level cooccurrence matrix and gray level run length matrix.Many features are found to be non-robust to changing parameters.Feature robustness assessment prior to feature selection,especially in the case of combining multi-institutional data,may be warranted.Further investigation is needed in this area of research.
文摘Background An early identification of the composition of arterial thrombus may have diagnostic, therapeutic, and prognostic implications. The variation of magnetic resonance （MR） signal intensity between white and red thrombi, especially in the susceptibility sensitive MR sequence, remains unknown. Our research was to evaluate the feasibility of MRI in differentiating of white and red thrombi with a phantom study. Methods A total of 12 red and 12 white thrombi were prepared with the venous blood. Examination of the phantom was completed using a 3.0T MR unit, including fluid attenuated inversion recovery （FLAIR） T1, T2-weighted imaging （T2WI）, FLAIR T2, T2＊ gradient echo （T2＊GRE） imaging, and susceptibility weighted angiography sequences （SWAN）. MR signal intensity patterns of the thrombi were objectively classified as hyperintensity, isointensity and hypointensity, compared with the background agar. The volume of thrombus was calculated and correlated with its signal intensity. Results For white thrombi, 11/12 clots showed hyperintensity and 1/12 showed isointensity in FLAIR T1 images. In T2WI, 6/12 clots showed hyperintensity, 3/12 isointensity, and 3/12 hypointensity. In FLAIR T2, 8/12 clots showed hyperintensity and 4/12 showed isointensity. In T2＊GRE, 3/12 clots showed hyperintensity and the remaining 9/12 clots showed isointensity. In SWAN, 5/12 clots demonstrated hyperintensity and 7/12 isointensity. For the red thrombus, 12/12 clots demonstrated hyperintensity in FLAIR T1, T2WI, and FLAIR T2 sequences. In T2＊GRE and SWAN sequences, 3/12 clots displayed hypointensity and the remaining 9/12 clots showed slight hyperintensity. Thrombi with hypointensity displayed in T2＊GRE and SWAN sequences were significantly larger than those with hyperintensity. Conclusions Differentiation of white and red thrombi with conventional MR sequence is unreliable, because both kinds of thrombi do not possess unique signal intensity features in these sequences. Red thrombus may or may not show hypointensity in the susceptibility sensitive MR sequences, depending on its size and time course.
文摘People with diabetes are 30 times more likely to undergo amputation than the general population.A total of 3.2million people in the UK have been diagnosed with diabetes and by 2025 this number is estimated to be 5million.One potential complication post amputation is phantom limb pain(PLP),a prevalent and often chronic condition which is difficult to treat.To inform a
