Impact of signal intensity normalization of MRI on the generalizability of radiomic-based prediction of molecular glioma subtypes
Objectives: Radiomic features have demonstrated encouraging results for non-invasive detection of molecular biomarkers, but the lack of guidelines for pre-processing MRI-data has led to poor generalizability. Here, we assessed the influence of different MRI-intensity normalization techniques on the...
Gespeichert in:
| Hauptverfasser: | , , , , , , , , |
|---|---|
| Dokumenttyp: | Article (Journal) |
| Sprache: | Englisch |
| Veröffentlicht: |
2023
|
| In: |
European radiology
Year: 2024, Jahrgang: 34, Heft: 4, Pages: 2782-2790 |
| ISSN: | 1432-1084 |
| DOI: | 10.1007/s00330-023-10034-2 |
| Online-Zugang: | Verlag, kostenfrei, Volltext: https://doi.org/10.1007/s00330-023-10034-2
|
| Verfasserangaben: | Martha Foltyn-Dumitru, Marianne Schell, Aditya Rastogi, Felix Sahm, Tobias Kessler, Wolfgang Wick, Martin Bendszus, Gianluca Brugnara, Philipp Vollmuth |
MARC
| LEADER | 00000caa a22000002c 4500 | ||
|---|---|---|---|
| 001 | 1868567567 | ||
| 003 | DE-627 | ||
| 005 | 20240626133036.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 231027s2023 xx |||||o 00| ||eng c | ||
| 024 | 7 | |a 10.1007/s00330-023-10034-2 |2 doi | |
| 035 | |a (DE-627)1868567567 | ||
| 035 | |a (DE-599)KXP1868567567 | ||
| 035 | |a (OCoLC)1425210025 | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rda | ||
| 041 | |a eng | ||
| 084 | |a 33 |2 sdnb | ||
| 100 | 1 | |a Foltyn-Dumitru, Martha |e VerfasserIn |0 (DE-588)1197171517 |0 (DE-627)1678980730 |4 aut | |
| 245 | 1 | 0 | |a Impact of signal intensity normalization of MRI on the generalizability of radiomic-based prediction of molecular glioma subtypes |c Martha Foltyn-Dumitru, Marianne Schell, Aditya Rastogi, Felix Sahm, Tobias Kessler, Wolfgang Wick, Martin Bendszus, Gianluca Brugnara, Philipp Vollmuth |
| 264 | 1 | |c 2023 | |
| 300 | |b Illustrationen | ||
| 300 | |a 9 | ||
| 336 | |a Text |b txt |2 rdacontent | ||
| 337 | |a Computermedien |b c |2 rdamedia | ||
| 338 | |a Online-Ressource |b cr |2 rdacarrier | ||
| 500 | |a Vorab online veröffentlicht: 06. September 2023 | ||
| 500 | |a Gesehen am 27.10.2023 | ||
| 520 | |a Objectives: Radiomic features have demonstrated encouraging results for non-invasive detection of molecular biomarkers, but the lack of guidelines for pre-processing MRI-data has led to poor generalizability. Here, we assessed the influence of different MRI-intensity normalization techniques on the performance of radiomics-based models for predicting molecular glioma subtypes. Methods: Preoperative MRI-data from n = 615 patients with newly diagnosed glioma and known isocitrate dehydrogenase (IDH) and 1p/19q status were pre-processed using four different methods: no normalization (naive), N4 bias field correction (N4), N4 followed by either WhiteStripe (N4/WS), or z-score normalization (N4/z-score). A total of 377 Image-Biomarker-Standardisation-Initiative-compliant radiomic features were extracted from each normalized data, and 9 different machine-learning algorithms were trained for multiclass prediction of molecular glioma subtypes (IDH-mutant 1p/19q codeleted vs. IDH-mutant 1p/19q non-codeleted vs. IDH wild type). External testing was performed in public glioma datasets from UCSF (n = 410) and TCGA (n = 160). Results: Support vector machine yielded the best performance with macro-average AUCs of 0.84 (naive), 0.84 (N4), 0.87 (N4/WS), and 0.87 (N4/z-score) in the internal test set. Both N4/WS and z-score outperformed the other approaches in the external UCSF and TCGA test sets with macro-average AUCs ranging from 0.85 to 0.87, replicating the performance of the internal test set, in contrast to macro-average AUCs ranging from 0.19 to 0.45 for naive and 0.26 to 0.52 for N4 alone. Conclusion: Intensity normalization of MRI data is essential for the generalizability of radiomic-based machine-learning models. Specifically, both N4/WS and N4/z-score approaches allow to preserve the high model performance, yielding generalizable performance when applying the developed radiomic-based machine-learning model in an external heterogeneous, multi-institutional setting. Clinical relevance statement: Intensity normalization such as N4/WS or N4/z-score can be used to develop reliable radiomics-based machine learning models from heterogeneous multicentre MRI datasets and provide non-invasive prediction of glioma subtypes. | ||
| 650 | 4 | |a Genotype | |
| 650 | 4 | |a Glioma | |
| 650 | 4 | |a Isocitrate dehydrogenase | |
| 650 | 4 | |a Magnetic resonance imaging | |
| 700 | 1 | |a Schell, Marianne |d 1986- |e VerfasserIn |0 (DE-588)1197163433 |0 (DE-627)167897076X |4 aut | |
| 700 | 1 | |a Rastogi, Aditya |e VerfasserIn |0 (DE-588)1303198118 |0 (DE-627)1860132634 |4 aut | |
| 700 | 1 | |a Sahm, Felix |d 1984- |e VerfasserIn |0 (DE-588)1022852132 |0 (DE-627)717318478 |0 (DE-576)366075020 |4 aut | |
| 700 | 1 | |a Keßler, Tobias |d 1987- |e VerfasserIn |0 (DE-588)1104151693 |0 (DE-627)86164591X |0 (DE-576)470840277 |4 aut | |
| 700 | 1 | |a Wick, Wolfgang |d 1970- |e VerfasserIn |0 (DE-588)120297736 |0 (DE-627)080586929 |0 (DE-576)186221320 |4 aut | |
| 700 | 1 | |a Bendszus, Martin |e VerfasserIn |0 (DE-588)1032676426 |0 (DE-627)738634131 |0 (DE-576)175567697 |4 aut | |
| 700 | 1 | |a Brugnara, Gianluca |e VerfasserIn |0 (DE-588)1197163123 |0 (DE-627)1678969850 |4 aut | |
| 700 | 1 | |a Vollmuth, Philipp |d 1987- |e VerfasserIn |0 (DE-588)1043270086 |0 (DE-627)771319177 |0 (DE-576)394600738 |4 aut | |
| 773 | 0 | 8 | |i Enthalten in |t European radiology |d Berlin : Springer, 1991 |g 34(2024), 4, Seite 2782-2790 |h Online-Ressource |w (DE-627)268757526 |w (DE-600)1472718-3 |w (DE-576)103868070 |x 1432-1084 |7 nnas |a Impact of signal intensity normalization of MRI on the generalizability of radiomic-based prediction of molecular glioma subtypes |
| 773 | 1 | 8 | |g volume:34 |g year:2024 |g number:4 |g pages:2782-2790 |g extent:9 |a Impact of signal intensity normalization of MRI on the generalizability of radiomic-based prediction of molecular glioma subtypes |
| 856 | 4 | 0 | |u https://doi.org/10.1007/s00330-023-10034-2 |x Verlag |x Resolving-System |z kostenfrei |3 Volltext |
| 951 | |a AR | ||
| 992 | |a 20231027 | ||
| 993 | |a Article | ||
| 994 | |a 2023 | ||
| 998 | |g 1043270086 |a Vollmuth, Philipp |m 1043270086:Vollmuth, Philipp |d 910000 |d 911100 |d 50000 |e 910000PV1043270086 |e 911100PV1043270086 |e 50000PV1043270086 |k 0/910000/ |k 1/910000/911100/ |k 0/50000/ |p 9 |y j | ||
| 998 | |g 1197163123 |a Brugnara, Gianluca |m 1197163123:Brugnara, Gianluca |d 910000 |d 911100 |e 910000PB1197163123 |e 911100PB1197163123 |k 0/910000/ |k 1/910000/911100/ |p 8 | ||
| 998 | |g 1032676426 |a Bendszus, Martin |m 1032676426:Bendszus, Martin |d 910000 |d 911100 |e 910000PB1032676426 |e 911100PB1032676426 |k 0/910000/ |k 1/910000/911100/ |p 7 | ||
| 998 | |g 120297736 |a Wick, Wolfgang |m 120297736:Wick, Wolfgang |d 910000 |d 911100 |e 910000PW120297736 |e 911100PW120297736 |k 0/910000/ |k 1/910000/911100/ |p 6 | ||
| 998 | |g 1104151693 |a Keßler, Tobias |m 1104151693:Keßler, Tobias |d 910000 |d 911100 |e 910000PK1104151693 |e 911100PK1104151693 |k 0/910000/ |k 1/910000/911100/ |p 5 | ||
| 998 | |g 1022852132 |a Sahm, Felix |m 1022852132:Sahm, Felix |d 910000 |d 912000 |d 50000 |e 910000PS1022852132 |e 912000PS1022852132 |e 50000PS1022852132 |k 0/910000/ |k 1/910000/912000/ |k 0/50000/ |p 4 | ||
| 998 | |g 1303198118 |a Rastogi, Aditya |m 1303198118:Rastogi, Aditya |d 910000 |d 911100 |e 910000PR1303198118 |e 911100PR1303198118 |k 0/910000/ |k 1/910000/911100/ |p 3 | ||
| 998 | |g 1197163433 |a Schell, Marianne |m 1197163433:Schell, Marianne |d 910000 |d 911100 |e 910000PS1197163433 |e 911100PS1197163433 |k 0/910000/ |k 1/910000/911100/ |p 2 | ||
| 998 | |g 1197171517 |a Foltyn-Dumitru, Martha |m 1197171517:Foltyn-Dumitru, Martha |d 910000 |d 911100 |e 910000PF1197171517 |e 911100PF1197171517 |k 0/910000/ |k 1/910000/911100/ |p 1 |x j | ||
| 999 | |a KXP-PPN1868567567 |e 439916557X | ||
| BIB | |a Y | ||
| SER | |a journal | ||
| JSO | |a {"relHost":[{"title":[{"subtitle":"official organ of the European Association of Radiology","title":"European radiology","title_sort":"European radiology"}],"disp":"Impact of signal intensity normalization of MRI on the generalizability of radiomic-based prediction of molecular glioma subtypesEuropean radiology","origin":[{"publisher":"Springer","publisherPlace":"Berlin ; Heidelberg","dateIssuedKey":"1991","dateIssuedDisp":"1991-"}],"part":{"extent":"9","volume":"34","issue":"4","pages":"2782-2790","year":"2024","text":"34(2024), 4, Seite 2782-2790"},"id":{"issn":["1432-1084","1613-3757"],"eki":["268757526"],"zdb":["1472718-3"]},"language":["eng"],"pubHistory":["1.1991 -"],"type":{"media":"Online-Ressource","bibl":"periodical"},"titleAlt":[{"title":"European radiology / Supplement"},{"title":"European radiology supplements"}],"note":["Gesehen am 06.02.2018","Ungezählte Beil.: Supplement"],"corporate":[{"display":"European Association of Radiology","role":"isb"}],"recId":"268757526","physDesc":[{"extent":"Online-Ressource"}]}],"recId":"1868567567","physDesc":[{"noteIll":"Illustrationen","extent":"9 S."}],"language":["eng"],"note":["Vorab online veröffentlicht: 06. September 2023","Gesehen am 27.10.2023"],"person":[{"given":"Martha","family":"Foltyn-Dumitru","display":"Foltyn-Dumitru, Martha","role":"aut"},{"family":"Schell","role":"aut","display":"Schell, Marianne","given":"Marianne"},{"given":"Aditya","family":"Rastogi","display":"Rastogi, Aditya","role":"aut"},{"family":"Sahm","display":"Sahm, Felix","role":"aut","given":"Felix"},{"given":"Tobias","role":"aut","display":"Keßler, Tobias","family":"Keßler"},{"display":"Wick, Wolfgang","role":"aut","family":"Wick","given":"Wolfgang"},{"family":"Bendszus","role":"aut","display":"Bendszus, Martin","given":"Martin"},{"given":"Gianluca","family":"Brugnara","display":"Brugnara, Gianluca","role":"aut"},{"family":"Vollmuth","role":"aut","display":"Vollmuth, Philipp","given":"Philipp"}],"id":{"doi":["10.1007/s00330-023-10034-2"],"eki":["1868567567"]},"origin":[{"dateIssuedKey":"2023","dateIssuedDisp":"2023"}],"name":{"displayForm":["Martha Foltyn-Dumitru, Marianne Schell, Aditya Rastogi, Felix Sahm, Tobias Kessler, Wolfgang Wick, Martin Bendszus, Gianluca Brugnara, Philipp Vollmuth"]},"title":[{"title":"Impact of signal intensity normalization of MRI on the generalizability of radiomic-based prediction of molecular glioma subtypes","title_sort":"Impact of signal intensity normalization of MRI on the generalizability of radiomic-based prediction of molecular glioma subtypes"}],"type":{"bibl":"article-journal","media":"Online-Ressource"}} | ||
| SRT | |a FOLTYNDUMIIMPACTOFSI2023 | ||