Improved interpretation of 18F-florzolotau PET in progressive supranuclear palsy using a normalization-free deep-learning classifier
Summary: While 18F-florzolotau tau PET is an emerging biomarker for progressive supranuclear palsy (PSP), its interpretation has been hindered by a lack of consensus on visual reading and potential biases in conventional semi-quantitative analysis. As clinical manifestations and regions of elevated...
| Main Authors: | , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2023-08-01
|
| Series: | iScience |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2589004223015031 |
| _version_ | 1827878422648455168 |
|---|---|
| author | Jiaying Lu Christoph Clement Jimin Hong Min Wang Xinyi Li Lara Cavinato Tzu-Chen Yen Fangyang Jiao Ping Wu Jianjun Wu Jingjie Ge Yimin Sun Matthias Brendel Leonor Lopes Axel Rominger Jian Wang Fengtao Liu Chuantao Zuo Yihui Guan Qianhua Zhao Kuangyu Shi |
| author_facet | Jiaying Lu Christoph Clement Jimin Hong Min Wang Xinyi Li Lara Cavinato Tzu-Chen Yen Fangyang Jiao Ping Wu Jianjun Wu Jingjie Ge Yimin Sun Matthias Brendel Leonor Lopes Axel Rominger Jian Wang Fengtao Liu Chuantao Zuo Yihui Guan Qianhua Zhao Kuangyu Shi |
| author_sort | Jiaying Lu |
| collection | DOAJ |
| description | Summary: While 18F-florzolotau tau PET is an emerging biomarker for progressive supranuclear palsy (PSP), its interpretation has been hindered by a lack of consensus on visual reading and potential biases in conventional semi-quantitative analysis. As clinical manifestations and regions of elevated 18F-florzolotau binding are highly overlapping in PSP and the Parkinsonian type of multiple system atrophy (MSA-P), developing a reliable discriminative classifier for 18F-florzolotau PET is urgently needed. Herein, we developed a normalization-free deep-learning (NFDL) model for 18F-florzolotau PET, which achieved significantly higher accuracy for both PSP and MSA-P compared to semi-quantitative classifiers. Regions driving the NFDL classifier’s decision were consistent with disease-specific topographies. NFDL-guided radiomic features correlated with clinical severity of PSP. This suggests that the NFDL model has the potential for early and accurate differentiation of atypical parkinsonism and that it can be applied in various scenarios due to not requiring subjective interpretation, MR-dependent, and reference-based preprocessing. |
| first_indexed | 2024-03-12T17:51:22Z |
| format | Article |
| id | doaj.art-3c05407d807e465a8003123060023d00 |
| institution | Directory Open Access Journal |
| issn | 2589-0042 |
| language | English |
| last_indexed | 2024-03-12T17:51:22Z |
| publishDate | 2023-08-01 |
| publisher | Elsevier |
| record_format | Article |
| series | iScience |
| spelling | doaj.art-3c05407d807e465a8003123060023d002023-08-03T04:23:31ZengElsevieriScience2589-00422023-08-01268107426Improved interpretation of 18F-florzolotau PET in progressive supranuclear palsy using a normalization-free deep-learning classifierJiaying Lu0Christoph Clement1Jimin Hong2Min Wang3Xinyi Li4Lara Cavinato5Tzu-Chen Yen6Fangyang Jiao7Ping Wu8Jianjun Wu9Jingjie Ge10Yimin Sun11Matthias Brendel12Leonor Lopes13Axel Rominger14Jian Wang15Fengtao Liu16Chuantao Zuo17Yihui Guan18Qianhua Zhao19Kuangyu Shi20Department of Nuclear Medicine & PET Center & National Center for Neurological Disorders & National Clinical Research Center for Aging and Medicine, Huashan Hospital, Fudan University, Shanghai 200235, China; Department of Nuclear Medicine, Inselspital, Bern University Hospital, University of Bern, 3010 Bern, SwitzerlandDepartment of Nuclear Medicine, Inselspital, Bern University Hospital, University of Bern, 3010 Bern, SwitzerlandDepartment of Nuclear Medicine, Inselspital, Bern University Hospital, University of Bern, 3010 Bern, SwitzerlandInstitute of Biomedical Engineering, School of Life Science, Shanghai University, Shanghai 200444, China; Department of Informatics, Technical University of Munich, 80333 Munich, GermanyDepartment of Neurology & National Center for Neurological Disorders & National Clinical Research Center for Aging and Medicine, Huashan Hospital, Fudan University, Shanghai 200400, ChinaDepartment of Nuclear Medicine, Inselspital, Bern University Hospital, University of Bern, 3010 Bern, Switzerland; MOX - Modeling and Scientific Computing, Department of Mathematics, Politecnico di Milano, Piazza Leonardo da Vinci 32, 20133 Milan, ItalyAPRINOIA Therapeutics Co., Ltd, Suzhou 215122, ChinaDepartment of Nuclear Medicine & PET Center & National Center for Neurological Disorders & National Clinical Research Center for Aging and Medicine, Huashan Hospital, Fudan University, Shanghai 200235, ChinaDepartment of Nuclear Medicine & PET Center & National Center for Neurological Disorders & National Clinical Research Center for Aging and Medicine, Huashan Hospital, Fudan University, Shanghai 200235, ChinaDepartment of Neurology & National Center for Neurological Disorders & National Clinical Research Center for Aging and Medicine, Huashan Hospital, Fudan University, Shanghai 200400, ChinaDepartment of Nuclear Medicine & PET Center & National Center for Neurological Disorders & National Clinical Research Center for Aging and Medicine, Huashan Hospital, Fudan University, Shanghai 200235, ChinaDepartment of Neurology & National Center for Neurological Disorders & National Clinical Research Center for Aging and Medicine, Huashan Hospital, Fudan University, Shanghai 200400, ChinaDepartment of Nuclear Medicine, University of Munich, 80539 Munich, GermanyDepartment of Nuclear Medicine, Inselspital, Bern University Hospital, University of Bern, 3010 Bern, SwitzerlandDepartment of Nuclear Medicine, Inselspital, Bern University Hospital, University of Bern, 3010 Bern, SwitzerlandDepartment of Neurology & National Center for Neurological Disorders & National Clinical Research Center for Aging and Medicine, Huashan Hospital, Fudan University, Shanghai 200400, ChinaDepartment of Neurology & National Center for Neurological Disorders & National Clinical Research Center for Aging and Medicine, Huashan Hospital, Fudan University, Shanghai 200400, China; Corresponding authorDepartment of Nuclear Medicine & PET Center & National Center for Neurological Disorders & National Clinical Research Center for Aging and Medicine, Huashan Hospital, Fudan University, Shanghai 200235, China; Human Phenome Institute, Fudan University, Shanghai 200433, China; Corresponding authorDepartment of Nuclear Medicine & PET Center & National Center for Neurological Disorders & National Clinical Research Center for Aging and Medicine, Huashan Hospital, Fudan University, Shanghai 200235, ChinaDepartment of Neurology & National Center for Neurological Disorders & National Clinical Research Center for Aging and Medicine, Huashan Hospital, Fudan University, Shanghai 200400, ChinaDepartment of Nuclear Medicine, Inselspital, Bern University Hospital, University of Bern, 3010 Bern, Switzerland; Department of Informatics, Technical University of Munich, 80333 Munich, GermanySummary: While 18F-florzolotau tau PET is an emerging biomarker for progressive supranuclear palsy (PSP), its interpretation has been hindered by a lack of consensus on visual reading and potential biases in conventional semi-quantitative analysis. As clinical manifestations and regions of elevated 18F-florzolotau binding are highly overlapping in PSP and the Parkinsonian type of multiple system atrophy (MSA-P), developing a reliable discriminative classifier for 18F-florzolotau PET is urgently needed. Herein, we developed a normalization-free deep-learning (NFDL) model for 18F-florzolotau PET, which achieved significantly higher accuracy for both PSP and MSA-P compared to semi-quantitative classifiers. Regions driving the NFDL classifier’s decision were consistent with disease-specific topographies. NFDL-guided radiomic features correlated with clinical severity of PSP. This suggests that the NFDL model has the potential for early and accurate differentiation of atypical parkinsonism and that it can be applied in various scenarios due to not requiring subjective interpretation, MR-dependent, and reference-based preprocessing.http://www.sciencedirect.com/science/article/pii/S2589004223015031Health informaticsMedical imagingClinical neuroscience |
| spellingShingle | Jiaying Lu Christoph Clement Jimin Hong Min Wang Xinyi Li Lara Cavinato Tzu-Chen Yen Fangyang Jiao Ping Wu Jianjun Wu Jingjie Ge Yimin Sun Matthias Brendel Leonor Lopes Axel Rominger Jian Wang Fengtao Liu Chuantao Zuo Yihui Guan Qianhua Zhao Kuangyu Shi Improved interpretation of 18F-florzolotau PET in progressive supranuclear palsy using a normalization-free deep-learning classifier iScience Health informatics Medical imaging Clinical neuroscience |
| title | Improved interpretation of 18F-florzolotau PET in progressive supranuclear palsy using a normalization-free deep-learning classifier |
| title_full | Improved interpretation of 18F-florzolotau PET in progressive supranuclear palsy using a normalization-free deep-learning classifier |
| title_fullStr | Improved interpretation of 18F-florzolotau PET in progressive supranuclear palsy using a normalization-free deep-learning classifier |
| title_full_unstemmed | Improved interpretation of 18F-florzolotau PET in progressive supranuclear palsy using a normalization-free deep-learning classifier |
| title_short | Improved interpretation of 18F-florzolotau PET in progressive supranuclear palsy using a normalization-free deep-learning classifier |
| title_sort | improved interpretation of 18f florzolotau pet in progressive supranuclear palsy using a normalization free deep learning classifier |
| topic | Health informatics Medical imaging Clinical neuroscience |
| url | http://www.sciencedirect.com/science/article/pii/S2589004223015031 |
| work_keys_str_mv | AT jiayinglu improvedinterpretationof18fflorzolotaupetinprogressivesupranuclearpalsyusinganormalizationfreedeeplearningclassifier AT christophclement improvedinterpretationof18fflorzolotaupetinprogressivesupranuclearpalsyusinganormalizationfreedeeplearningclassifier AT jiminhong improvedinterpretationof18fflorzolotaupetinprogressivesupranuclearpalsyusinganormalizationfreedeeplearningclassifier AT minwang improvedinterpretationof18fflorzolotaupetinprogressivesupranuclearpalsyusinganormalizationfreedeeplearningclassifier AT xinyili improvedinterpretationof18fflorzolotaupetinprogressivesupranuclearpalsyusinganormalizationfreedeeplearningclassifier AT laracavinato improvedinterpretationof18fflorzolotaupetinprogressivesupranuclearpalsyusinganormalizationfreedeeplearningclassifier AT tzuchenyen improvedinterpretationof18fflorzolotaupetinprogressivesupranuclearpalsyusinganormalizationfreedeeplearningclassifier AT fangyangjiao improvedinterpretationof18fflorzolotaupetinprogressivesupranuclearpalsyusinganormalizationfreedeeplearningclassifier AT pingwu improvedinterpretationof18fflorzolotaupetinprogressivesupranuclearpalsyusinganormalizationfreedeeplearningclassifier AT jianjunwu improvedinterpretationof18fflorzolotaupetinprogressivesupranuclearpalsyusinganormalizationfreedeeplearningclassifier AT jingjiege improvedinterpretationof18fflorzolotaupetinprogressivesupranuclearpalsyusinganormalizationfreedeeplearningclassifier AT yiminsun improvedinterpretationof18fflorzolotaupetinprogressivesupranuclearpalsyusinganormalizationfreedeeplearningclassifier AT matthiasbrendel improvedinterpretationof18fflorzolotaupetinprogressivesupranuclearpalsyusinganormalizationfreedeeplearningclassifier AT leonorlopes improvedinterpretationof18fflorzolotaupetinprogressivesupranuclearpalsyusinganormalizationfreedeeplearningclassifier AT axelrominger improvedinterpretationof18fflorzolotaupetinprogressivesupranuclearpalsyusinganormalizationfreedeeplearningclassifier AT jianwang improvedinterpretationof18fflorzolotaupetinprogressivesupranuclearpalsyusinganormalizationfreedeeplearningclassifier AT fengtaoliu improvedinterpretationof18fflorzolotaupetinprogressivesupranuclearpalsyusinganormalizationfreedeeplearningclassifier AT chuantaozuo improvedinterpretationof18fflorzolotaupetinprogressivesupranuclearpalsyusinganormalizationfreedeeplearningclassifier AT yihuiguan improvedinterpretationof18fflorzolotaupetinprogressivesupranuclearpalsyusinganormalizationfreedeeplearningclassifier AT qianhuazhao improvedinterpretationof18fflorzolotaupetinprogressivesupranuclearpalsyusinganormalizationfreedeeplearningclassifier AT kuangyushi improvedinterpretationof18fflorzolotaupetinprogressivesupranuclearpalsyusinganormalizationfreedeeplearningclassifier |