The feasibility of K i parametric imaging in clinical dynamic 18 F‐FDG total‐body PET using a simulated‐data‐driven machine learning algorithm
Despite K parametric images providing high sensitivity and specificity in clinical diagnosis and therapeutic evaluation, their clinical application is limited by prolonged scan durations. Previous research has employed machine learning algorithms to generate reliable K images from F-FDG PET scans wi...
Uložené v:
| Vydané v: | Medical physics (Lancaster) Ročník 52; číslo 10; s. e70053 |
|---|---|
| Hlavní autori: | , , , , , , , , , , , , |
| Médium: | Journal Article |
| Jazyk: | English |
| Vydavateľské údaje: |
United States
01.10.2025
|
| Predmet: | |
| ISSN: | 0094-2405, 2473-4209 |
| On-line prístup: | Získať plný text |
| Tagy: |
Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
|
| Abstract | Despite K
parametric images providing high sensitivity and specificity in clinical diagnosis and therapeutic evaluation, their clinical application is limited by prolonged scan durations. Previous research has employed machine learning algorithms to generate reliable K
images from
F-FDG PET scans with a duration of 10 min, but these algorithms require extensive real-world data for effective training.
This study explored the feasibility of a simulated-data-driven approach for clinical K
parametric imaging with reduced scan duration.
A cluster analysis was conducted using K-Means on 60-minute dynamic total-body PET data from 25 subjects, followed by the construction of a noise-free simulated dataset using the Patlak equation. Subsequently, noise at varying levels was incorporated into the noise-free simulated dataset, resulting in four distinct training sets. Each dataset contains 1 200 000 simulated K
values as labels, along with corresponding tissue-to-blood concentration ratios categorized by noise level (free, low, middle, and high). Based on the same simulation methodology, we also generated four corresponding test datasets at different noise levels (each with identical size to the training dataset). Ultimately, XGBoost models, each trained with data at different noise levels, were employed to predict K
values from short-duration (50-60 min) PET scans. The accuracy of the K
values generated from short-duration scans by our proposed method was evaluated using both simulated data (the four test datasets at varying noise levels) and real-world data (dynamic total-body PET data from 25 subjects). The K
images generated by the conventional Patlak method with a t* of 20 min post-injection were chosen as the gold standard.
Evaluation based on both simulated and real-world data indicates that training algorithms with a noise-inclusive simulated dataset significantly improve the average accuracy of K
values. In evaluations with real-world data, K
images from 50-60-minute dynamic PET scans generated using our proposed approach demonstrated superior performance compared to the conventional Patlak method, achieving a Pearson's correlation coefficient of 0.94 (vs. 0.42 for Patlak), a lower normalized mean square error of 0.11 (vs. 5.33), and a higher peak signal-to-noise ratio of 64.32 (vs. 47.87).
The simulated-data-driven approach can generate reliable K
images from clinical
F-FDG dynamic total-body PET scans, thereby reducing the costs associated with collecting real-world data. |
|---|---|
| AbstractList | Despite K
parametric images providing high sensitivity and specificity in clinical diagnosis and therapeutic evaluation, their clinical application is limited by prolonged scan durations. Previous research has employed machine learning algorithms to generate reliable K
images from
F-FDG PET scans with a duration of 10 min, but these algorithms require extensive real-world data for effective training.
This study explored the feasibility of a simulated-data-driven approach for clinical K
parametric imaging with reduced scan duration.
A cluster analysis was conducted using K-Means on 60-minute dynamic total-body PET data from 25 subjects, followed by the construction of a noise-free simulated dataset using the Patlak equation. Subsequently, noise at varying levels was incorporated into the noise-free simulated dataset, resulting in four distinct training sets. Each dataset contains 1 200 000 simulated K
values as labels, along with corresponding tissue-to-blood concentration ratios categorized by noise level (free, low, middle, and high). Based on the same simulation methodology, we also generated four corresponding test datasets at different noise levels (each with identical size to the training dataset). Ultimately, XGBoost models, each trained with data at different noise levels, were employed to predict K
values from short-duration (50-60 min) PET scans. The accuracy of the K
values generated from short-duration scans by our proposed method was evaluated using both simulated data (the four test datasets at varying noise levels) and real-world data (dynamic total-body PET data from 25 subjects). The K
images generated by the conventional Patlak method with a t* of 20 min post-injection were chosen as the gold standard.
Evaluation based on both simulated and real-world data indicates that training algorithms with a noise-inclusive simulated dataset significantly improve the average accuracy of K
values. In evaluations with real-world data, K
images from 50-60-minute dynamic PET scans generated using our proposed approach demonstrated superior performance compared to the conventional Patlak method, achieving a Pearson's correlation coefficient of 0.94 (vs. 0.42 for Patlak), a lower normalized mean square error of 0.11 (vs. 5.33), and a higher peak signal-to-noise ratio of 64.32 (vs. 47.87).
The simulated-data-driven approach can generate reliable K
images from clinical
F-FDG dynamic total-body PET scans, thereby reducing the costs associated with collecting real-world data. |
| Author | Chen, Xiaojun Shi, Hongcheng Luo, Gongning Yang, Wentong Xu, Tianyi Li, Yanxiao Gu, Wenjian Zhou, Yun Liu, Weiping Zhu, Zhanshi Wang, Kuanquan Wang, Yihan Liu, Ze |
| Author_xml | – sequence: 1 givenname: Wenjian surname: Gu fullname: Gu, Wenjian organization: United Imaging Healthcare Group Co., Ltd Shanghai China, Faculty of Computing Harbin Institute of Technology Harbin China – sequence: 2 givenname: Weiping surname: Liu fullname: Liu, Weiping organization: Institute of Biomedical Manufacturing and Life Quality Engineering, School of Mechanical Engineering Shanghai Jiao Tong University Shanghai China, United Imaging Healthcare Group Co., Ltd Shanghai China – sequence: 3 givenname: Wentong surname: Yang fullname: Yang, Wentong organization: United Imaging Healthcare Group Co., Ltd Shanghai China, ShanghaiTech University Shanghai China – sequence: 4 givenname: Yihan surname: Wang fullname: Wang, Yihan organization: United Imaging Healthcare Group Co., Ltd Shanghai China, ShanghaiTech University Shanghai China – sequence: 5 givenname: Ze surname: Liu fullname: Liu, Ze organization: United Imaging Healthcare Group Co., Ltd Shanghai China, ShanghaiTech University Shanghai China – sequence: 6 givenname: Zhanshi surname: Zhu fullname: Zhu, Zhanshi organization: United Imaging Healthcare Group Co., Ltd Shanghai China, Faculty of Computing Harbin Institute of Technology Harbin China – sequence: 7 givenname: Yanxiao surname: Li fullname: Li, Yanxiao organization: United Imaging Healthcare Group Co., Ltd Shanghai China – sequence: 8 givenname: Tianyi surname: Xu fullname: Xu, Tianyi organization: United Imaging Healthcare Group Co., Ltd Shanghai China – sequence: 9 givenname: Hongcheng surname: Shi fullname: Shi, Hongcheng organization: Department of Nuclear Medicine, Zhongshan Hospital Fudan University Shanghai China – sequence: 10 givenname: Kuanquan surname: Wang fullname: Wang, Kuanquan organization: Faculty of Computing Harbin Institute of Technology Harbin China – sequence: 11 givenname: Gongning surname: Luo fullname: Luo, Gongning organization: Faculty of Computing Harbin Institute of Technology Harbin China – sequence: 12 givenname: Xiaojun surname: Chen fullname: Chen, Xiaojun organization: Faculty of Computing Harbin Institute of Technology Harbin China – sequence: 13 givenname: Yun surname: Zhou fullname: Zhou, Yun organization: United Imaging Healthcare Group Co., Ltd Shanghai China, ShanghaiTech University Shanghai China |
| BackLink | https://www.ncbi.nlm.nih.gov/pubmed/41076590$$D View this record in MEDLINE/PubMed |
| BookMark | eNo9kE1OwzAQRi1URH9A4gTISzYp4yR2kyUqbUFUgkX30dSZtEaxEzkpUnccgRUH5CSkLbD6RvrejDRvyHqucsTYtYCxAAjvbD2eAMjojA3CeBIFcQhpjw0A0jgIY5B9NmyaNwBQkYQL1o8FTJRMYcC-VlviBWFj1qY07Z5XBX_mhtfo0VLrjebG4sa4DTeO69I4o7Hk-d6h7TqR8Pn3x-f8YcHbqsWym9dVvuevsxXfNYct5I2xuxJbyrsyxxYP4c07OW5Rb40jXhJ6d4TLTeVNu7WX7LzAsqGr3xyx1Xy2mj4Gy5fF0_R-GWipIEBMRQ40wXWiVE4i0TGhpDQWhZBKaKE1hZjIREolcyh0ESmtpCoiIZTUFI3YzelsvVtbyrPad8_6ffbnpwNuT4D2VdN4Kv4RAdlBfWbr7Kg--gGKHnrA |
| Cites_doi | 10.1145/1772690.1772862 10.1037/1528-3542.6.2.269 10.1016/j.tics.2007.05.005 10.1088/1361-6560/aad97f 10.1007/s00330-022-08960-8 10.1155/2007/65641 10.1088/1361-6560/ad9ce4 10.1145/2939672.2939785 10.1088/0031-9155/60/22/8643 10.1109/TMI.2021.3112783 10.1007/s00259-022-06010-5 10.1007/s00259-022-06003-4 10.1002/mp.15113 10.1006/nimg.2001.1021 10.3389/fradi.2023.1153784 10.1109/78.650093 10.1269/jrr.11089 10.1007/s00330-018-5966-1 10.1088/0031-9155/58/20/7391 10.1016/S1053-8119(03)00017-X 10.1007/s00259-023-06456-1 10.1007/s00259-022-05731-x 10.1007/s00330-022-09237-w 10.1007/s00259-018-4153-6 10.1007/s00259-022-05904-8 10.2967/jnumed.120.261651 10.1038/s41467-022-34257-x 10.1109/42.476111 10.1186/2191-219X-3-77 10.1109/TRPMS.2023.3243576 10.1007/s00259-022-05867-w 10.1007/s00259-023-06429-4 10.1007/s00259-023-06299-w 10.1162/neco.1997.9.8.1735 10.2967/jnumed.113.133892 10.1007/s11307-011-0514-2 10.1007/s00259-020-05134-w 10.1007/s00259-024-07008-x 10.1038/jcbfm.1983.1 10.1259/bjr.20170508 |
| ContentType | Journal Article |
| Copyright | 2025 American Association of Physicists in Medicine. |
| Copyright_xml | – notice: 2025 American Association of Physicists in Medicine. |
| DBID | AAYXX CITATION CGR CUY CVF ECM EIF NPM |
| DOI | 10.1002/mp.70053 |
| DatabaseName | CrossRef Medline MEDLINE MEDLINE (Ovid) MEDLINE MEDLINE PubMed |
| DatabaseTitle | CrossRef MEDLINE Medline Complete MEDLINE with Full Text PubMed MEDLINE (Ovid) |
| DatabaseTitleList | MEDLINE |
| Database_xml | – sequence: 1 dbid: NPM name: PubMed url: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed sourceTypes: Index Database |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Medicine Physics |
| EISSN | 2473-4209 |
| ExternalDocumentID | 41076590 10_1002_mp_70053 |
| Genre | Journal Article |
| GrantInformation_xml | – fundername: National Natural Science Foundation of China grantid: 62001144 – fundername: National Natural Science Foundation of China grantid: 62272135 – fundername: National Natural Science Foundation of China grantid: 62372135 |
| GroupedDBID | --- --Z -DZ .GJ 0R~ 1OB 1OC 29M 2WC 33P 36B 3O- 4.4 53G 5GY 5RE 5VS AAHQN AAIPD AAMMB AAMNL AANLZ AAQQT AASGY AAXRX AAYCA AAYXX AAZKR ABCUV ABDPE ABEFU ABJNI ABLJU ABQWH ABUFD ABXGK ACAHQ ACBEA ACCZN ACGFO ACGFS ACGOF ACPOU ACXBN ACXQS ADBBV ADBTR ADKYN ADMLS ADOZA ADXAS ADZMN AEFGJ AEGXH AEIGN AENEX AEUYR AEYWJ AFBPY AFFPM AFWVQ AGHNM AGXDD AGYGG AHBTC AIACR AIAGR AIDQK AIDYY AIQQE AITYG AIURR ALMA_UNASSIGNED_HOLDINGS ALVPJ AMYDB ASPBG BFHJK C45 CITATION CS3 DCZOG DRFUL DRMAN DRSTM DU5 EBD EBS EJD EMB EMOBN F5P HDBZQ HGLYW I-F KBYEO LATKE LEEKS LH4 LOXES LUTES LYRES MEWTI O9- OVD P2P P2W PALCI PHY RJQFR RNS ROL SAMSI SUPJJ SV3 TEORI TN5 TWZ USG WOHZO WXSBR ZGI ZVN ZXP ZY4 ZZTAW ALUQN CGR CUY CVF ECM EIF NPM XJT |
| ID | FETCH-LOGICAL-c560-aa91d0e7ab866de18c4ea5e941f1561c1cce2a8585565d0fcf36c656f31165ce3 |
| ISSN | 0094-2405 |
| IngestDate | Mon Oct 13 01:41:30 EDT 2025 Thu Nov 20 00:55:27 EST 2025 |
| IsPeerReviewed | true |
| IsScholarly | true |
| Issue | 10 |
| Keywords | total‐body PET Patlak plot machine learning parametric image 18F‐FDG |
| Language | English |
| License | 2025 American Association of Physicists in Medicine. |
| LinkModel | OpenURL |
| MergedId | FETCHMERGED-LOGICAL-c560-aa91d0e7ab866de18c4ea5e941f1561c1cce2a8585565d0fcf36c656f31165ce3 |
| PMID | 41076590 |
| ParticipantIDs | pubmed_primary_41076590 crossref_primary_10_1002_mp_70053 |
| PublicationCentury | 2000 |
| PublicationDate | 2025-10-00 2025-Oct |
| PublicationDateYYYYMMDD | 2025-10-01 |
| PublicationDate_xml | – month: 10 year: 2025 text: 2025-10-00 |
| PublicationDecade | 2020 |
| PublicationPlace | United States |
| PublicationPlace_xml | – name: United States |
| PublicationTitle | Medical physics (Lancaster) |
| PublicationTitleAlternate | Med Phys |
| PublicationYear | 2025 |
| References | e_1_2_8_28_1 e_1_2_8_29_1 e_1_2_8_24_1 e_1_2_8_25_1 e_1_2_8_46_1 e_1_2_8_26_1 e_1_2_8_27_1 e_1_2_8_3_1 e_1_2_8_2_1 e_1_2_8_5_1 e_1_2_8_4_1 e_1_2_8_7_1 e_1_2_8_6_1 e_1_2_8_9_1 e_1_2_8_8_1 e_1_2_8_20_1 e_1_2_8_43_1 e_1_2_8_21_1 e_1_2_8_42_1 e_1_2_8_22_1 e_1_2_8_45_1 e_1_2_8_23_1 e_1_2_8_44_1 e_1_2_8_41_1 e_1_2_8_40_1 e_1_2_8_17_1 e_1_2_8_18_1 e_1_2_8_39_1 e_1_2_8_19_1 e_1_2_8_13_1 e_1_2_8_36_1 e_1_2_8_14_1 e_1_2_8_35_1 e_1_2_8_15_1 e_1_2_8_16_1 e_1_2_8_37_1 Gu W (e_1_2_8_47_1) 2023; 64 Pedregosa F (e_1_2_8_33_1) 2011; 12 Wu Q (e_1_2_8_32_1) 2022; 63 Zhou Y (e_1_2_8_38_1) 2019; 60 e_1_2_8_10_1 e_1_2_8_31_1 e_1_2_8_11_1 e_1_2_8_34_1 e_1_2_8_12_1 e_1_2_8_30_1 |
| References_xml | – ident: e_1_2_8_34_1 doi: 10.1145/1772690.1772862 – ident: e_1_2_8_22_1 doi: 10.1037/1528-3542.6.2.269 – ident: e_1_2_8_23_1 doi: 10.1016/j.tics.2007.05.005 – ident: e_1_2_8_10_1 doi: 10.1088/1361-6560/aad97f – ident: e_1_2_8_16_1 doi: 10.1007/s00330-022-08960-8 – ident: e_1_2_8_45_1 doi: 10.1155/2007/65641 – volume: 60 start-page: 1186 issue: 1 year: 2019 ident: e_1_2_8_38_1 article-title: A machine learning‐based parametric imaging algorithm for noninvasive quantification of dynamic [68Ga] DOTATATE PET‐CT publication-title: J Nucl Med – ident: e_1_2_8_40_1 doi: 10.1088/1361-6560/ad9ce4 – ident: e_1_2_8_39_1 doi: 10.1145/2939672.2939785 – ident: e_1_2_8_5_1 doi: 10.1088/0031-9155/60/22/8643 – ident: e_1_2_8_37_1 doi: 10.1109/TMI.2021.3112783 – ident: e_1_2_8_27_1 doi: 10.1007/s00259-022-06010-5 – volume: 63 start-page: 3184 issue: 2 year: 2022 ident: e_1_2_8_32_1 article-title: Impact of equilibration time (t*) on Patlak quantitation in dynamic total‐body imaging using the uEXPLORER PET scanner publication-title: J Nucl Med – ident: e_1_2_8_19_1 doi: 10.1007/s00259-022-06003-4 – ident: e_1_2_8_2_1 doi: 10.1002/mp.15113 – ident: e_1_2_8_44_1 doi: 10.1006/nimg.2001.1021 – ident: e_1_2_8_24_1 doi: 10.3389/fradi.2023.1153784 – ident: e_1_2_8_42_1 doi: 10.1109/78.650093 – ident: e_1_2_8_9_1 doi: 10.1269/jrr.11089 – ident: e_1_2_8_12_1 doi: 10.1007/s00330-018-5966-1 – ident: e_1_2_8_18_1 doi: 10.1088/0031-9155/58/20/7391 – ident: e_1_2_8_43_1 doi: 10.1016/S1053-8119(03)00017-X – ident: e_1_2_8_26_1 doi: 10.1007/s00259-023-06456-1 – ident: e_1_2_8_6_1 doi: 10.1007/s00259-022-05731-x – ident: e_1_2_8_15_1 doi: 10.1007/s00330-022-09237-w – ident: e_1_2_8_17_1 doi: 10.1007/s00259-018-4153-6 – ident: e_1_2_8_28_1 doi: 10.1007/s00259-022-05904-8 – ident: e_1_2_8_30_1 doi: 10.2967/jnumed.120.261651 – ident: e_1_2_8_13_1 doi: 10.1007/s00259-022-05904-8 – ident: e_1_2_8_25_1 doi: 10.1038/s41467-022-34257-x – ident: e_1_2_8_7_1 doi: 10.1109/42.476111 – ident: e_1_2_8_36_1 doi: 10.1186/2191-219X-3-77 – volume: 64 start-page: 1611 issue: 1 year: 2023 ident: e_1_2_8_47_1 article-title: A machine learning‐based parametric imaging algorithm with shortened scanning duration for noninvasive quantification of total‐body dynamic [18F] FDG PET‐CT publication-title: J Nucl Med – ident: e_1_2_8_21_1 doi: 10.1109/TRPMS.2023.3243576 – ident: e_1_2_8_20_1 doi: 10.1007/s00259-022-05867-w – ident: e_1_2_8_14_1 doi: 10.1007/s00259-023-06429-4 – ident: e_1_2_8_31_1 doi: 10.1007/s00259-023-06299-w – ident: e_1_2_8_41_1 doi: 10.1162/neco.1997.9.8.1735 – ident: e_1_2_8_3_1 doi: 10.2967/jnumed.113.133892 – ident: e_1_2_8_8_1 doi: 10.1007/s11307-011-0514-2 – ident: e_1_2_8_46_1 doi: 10.1007/s00259-020-05134-w – volume: 12 start-page: 2825 year: 2011 ident: e_1_2_8_33_1 article-title: Scikit‐learn: machine learning in Python publication-title: J Mach Learn Res – ident: e_1_2_8_35_1 doi: 10.1007/s00259-024-07008-x – ident: e_1_2_8_4_1 doi: 10.1038/jcbfm.1983.1 – ident: e_1_2_8_11_1 doi: 10.1259/bjr.20170508 – ident: e_1_2_8_29_1 doi: 10.1007/s00259-023-06429-4 |
| SSID | ssj0006350 |
| Score | 2.4781914 |
| Snippet | Despite K
parametric images providing high sensitivity and specificity in clinical diagnosis and therapeutic evaluation, their clinical application is limited... |
| SourceID | pubmed crossref |
| SourceType | Index Database |
| StartPage | e70053 |
| SubjectTerms | Algorithms Feasibility Studies Fluorodeoxyglucose F18 Humans Image Processing, Computer-Assisted - methods Machine Learning Positron-Emission Tomography - methods Whole Body Imaging - methods |
| Title | The feasibility of K i parametric imaging in clinical dynamic 18 F‐FDG total‐body PET using a simulated‐data‐driven machine learning algorithm |
| URI | https://www.ncbi.nlm.nih.gov/pubmed/41076590 |
| Volume | 52 |
| hasFullText | 1 |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| journalDatabaseRights | – providerCode: PRVWIB databaseName: Wiley Online Library - Journals customDbUrl: eissn: 2473-4209 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0006350 issn: 0094-2405 databaseCode: DRFUL dateStart: 19970101 isFulltext: true titleUrlDefault: https://onlinelibrary.wiley.com providerName: Wiley-Blackwell |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV1Lb9NAEF6FFhAXBOVVXhokbpbB7_UeEa2L1FBFKEA5Rc7uujWqnShNqnLjJ3DiL_E_-CXMPrxxKw7lwMVxJl4n8nyZnZn9doaQlzTgOM3GwqcSYxO0fqU_jVLho7qnOP3wMNPcnE9DenCQHx6y0WDwq9sLc3ZC2zY_P2fz_6pqlKGy1dbZf1C3uykK8ByVjkdUOx6vrPhKlpb2qhfQ973aUzW-G9U-i3t1Y1oTKQp6tzFSmM70Xph7hSNAFDt76Juie-4k05n45o12x95K5xhK77RuVAMwKdwlinO6frNQ1tRrNGVTdj0qcODJ0WxRL4-bvnPcLRqZbItOB6st2qVpHuIyFnsrTQyU7dcetIe1lape3EfOltlsOF6MPq4Tf7biL_WxvYPNe0SpY9A5W84StTZk1sSllkUJjf0kCljfvqdRH8fBX-cNU4e2mb-iyiqt58aOD3BpynRERlP0OZo084keeY1sRjRlaF43dz4UH4fOKUC_zuyGsj-6q4McRK-7b73gGV2IcbSvM75DbtsgBd4YcN0lA9lukZvvLQ1ji9wYGQ3dIz8RbdBDG8wq2Ica1mgDizaoW-jQBhZtEOZQ_P7-A3EGGmd4rhAGiDDQCIMSHMLwQ4Ut9aJRBRZV0KEKHKruk3GxO377zretPnyOLrdfliwUgaTlNM8yIcOcJ7JMJUvCKkQHn4dc9a1TS9gYf4ig4lWccYxEqlhVj-IyfkA22lkrHxGoUowZMWjhgtMky_g0yxgXIsvLWOK_hG2TF91TnsxNQZfJZS1uk4fm8bsrkjCgWcqCx1cY_YTcWuP1KdlYLlbyGbnOz5b16eK5BcYfJZ2dxw |
| linkProvider | Wiley-Blackwell |
| openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=The+feasibility+of+K+i+parametric+imaging+in+clinical+dynamic+18+F%E2%80%90FDG+total%E2%80%90body+PET+using+a+simulated%E2%80%90data%E2%80%90driven+machine+learning+algorithm&rft.jtitle=Medical+physics+%28Lancaster%29&rft.au=Gu%2C+Wenjian&rft.au=Liu%2C+Weiping&rft.au=Yang%2C+Wentong&rft.au=Wang%2C+Yihan&rft.date=2025-10-01&rft.issn=0094-2405&rft.eissn=2473-4209&rft.volume=52&rft.issue=10&rft_id=info:doi/10.1002%2Fmp.70053&rft.externalDBID=n%2Fa&rft.externalDocID=10_1002_mp_70053 |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0094-2405&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0094-2405&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0094-2405&client=summon |