Longitudinal clustering analysis and prediction of Parkinson's disease progression using radiomics and hybrid machine learning
We employed machine learning approaches to (I) determine distinct progression trajectories in Parkinson's disease (PD) (unsupervised clustering task), and (II) predict progression trajectories (supervised prediction task), from early (years 0 and 1) data, making use of clinical and imaging feat...
Saved in:
| Published in: | Quantitative imaging in medicine and surgery Vol. 12; no. 2; p. 906 |
|---|---|
| Main Authors: | , , , , |
| Format: | Journal Article |
| Language: | English |
| Published: |
China
01.02.2022
|
| Subjects: | |
| ISSN: | 2223-4292 |
| Online Access: | Get more information |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Abstract | We employed machine learning approaches to (I) determine distinct progression trajectories in Parkinson's disease (PD) (unsupervised clustering task), and (II) predict progression trajectories (supervised prediction task), from early (years 0 and 1) data, making use of clinical and imaging features.
We studied PD-subjects derived from longitudinal datasets (years 0, 1, 2 & 4; Parkinson's Progressive Marker Initiative). We extracted and analyzed 981 features, including motor, non-motor, and radiomics features extracted for each region-of-interest (ROIs: left/right caudate and putamen) using our standardized standardized environment for radiomics analysis (SERA) radiomics software. Segmentation of ROIs on dopamine transposer - single photon emission computed tomography (DAT SPECT) images were performed via magnetic resonance images (MRI). After performing cross-sectional clustering on 885 subjects (original dataset) to identify disease subtypes, we identified optimal longitudinal trajectories using hybrid machine learning systems (HMLS), including principal component analysis (PCA) + K-Means algorithms (KMA) followed by Bayesian information criterion (BIC), Calinski-Harabatz criterion (CHC), and elbow criterion (EC). Subsequently, prediction of the identified trajectories from early year data was performed using multiple HMLSs including 16 Dimension Reduction Algorithms (DRA) and 10 classification algorithms.
We identified 3 distinct progression trajectories. Hotelling's t squared test (HTST) showed that the identified trajectories were distinct. The trajectories included those with (I, II) disease escalation (2 trajectories, 27% and 38% of patients) and (III) stable disease (1 trajectory, 35% of patients). For trajectory prediction from early year data, HMLSs including the stochastic neighbor embedding algorithm (SNEA, as a DRA) as well as locally linear embedding algorithm (LLEA, as a DRA), linked with the new probabilistic neural network classifier (NPNNC, as a classifier), resulted in accuracies of 78.4% and 79.2% respectively, while other HMLSs such as SNEA + Lib_SVM (library for support vector machines) and t_SNE (t-distributed stochastic neighbor embedding) + NPNNC resulted in 76.5% and 76.1% respectively.
This study moves beyond cross-sectional PD subtyping to clustering of longitudinal disease trajectories. We conclude that combining medical information with SPECT-based radiomics features, and optimal utilization of HMLSs, can identify distinct disease trajectories in PD patients, and enable effective prediction of disease trajectories from early year data. |
|---|---|
| AbstractList | We employed machine learning approaches to (I) determine distinct progression trajectories in Parkinson's disease (PD) (unsupervised clustering task), and (II) predict progression trajectories (supervised prediction task), from early (years 0 and 1) data, making use of clinical and imaging features.
We studied PD-subjects derived from longitudinal datasets (years 0, 1, 2 & 4; Parkinson's Progressive Marker Initiative). We extracted and analyzed 981 features, including motor, non-motor, and radiomics features extracted for each region-of-interest (ROIs: left/right caudate and putamen) using our standardized standardized environment for radiomics analysis (SERA) radiomics software. Segmentation of ROIs on dopamine transposer - single photon emission computed tomography (DAT SPECT) images were performed via magnetic resonance images (MRI). After performing cross-sectional clustering on 885 subjects (original dataset) to identify disease subtypes, we identified optimal longitudinal trajectories using hybrid machine learning systems (HMLS), including principal component analysis (PCA) + K-Means algorithms (KMA) followed by Bayesian information criterion (BIC), Calinski-Harabatz criterion (CHC), and elbow criterion (EC). Subsequently, prediction of the identified trajectories from early year data was performed using multiple HMLSs including 16 Dimension Reduction Algorithms (DRA) and 10 classification algorithms.
We identified 3 distinct progression trajectories. Hotelling's t squared test (HTST) showed that the identified trajectories were distinct. The trajectories included those with (I, II) disease escalation (2 trajectories, 27% and 38% of patients) and (III) stable disease (1 trajectory, 35% of patients). For trajectory prediction from early year data, HMLSs including the stochastic neighbor embedding algorithm (SNEA, as a DRA) as well as locally linear embedding algorithm (LLEA, as a DRA), linked with the new probabilistic neural network classifier (NPNNC, as a classifier), resulted in accuracies of 78.4% and 79.2% respectively, while other HMLSs such as SNEA + Lib_SVM (library for support vector machines) and t_SNE (t-distributed stochastic neighbor embedding) + NPNNC resulted in 76.5% and 76.1% respectively.
This study moves beyond cross-sectional PD subtyping to clustering of longitudinal disease trajectories. We conclude that combining medical information with SPECT-based radiomics features, and optimal utilization of HMLSs, can identify distinct disease trajectories in PD patients, and enable effective prediction of disease trajectories from early year data. We employed machine learning approaches to (I) determine distinct progression trajectories in Parkinson's disease (PD) (unsupervised clustering task), and (II) predict progression trajectories (supervised prediction task), from early (years 0 and 1) data, making use of clinical and imaging features.BACKGROUNDWe employed machine learning approaches to (I) determine distinct progression trajectories in Parkinson's disease (PD) (unsupervised clustering task), and (II) predict progression trajectories (supervised prediction task), from early (years 0 and 1) data, making use of clinical and imaging features.We studied PD-subjects derived from longitudinal datasets (years 0, 1, 2 & 4; Parkinson's Progressive Marker Initiative). We extracted and analyzed 981 features, including motor, non-motor, and radiomics features extracted for each region-of-interest (ROIs: left/right caudate and putamen) using our standardized standardized environment for radiomics analysis (SERA) radiomics software. Segmentation of ROIs on dopamine transposer - single photon emission computed tomography (DAT SPECT) images were performed via magnetic resonance images (MRI). After performing cross-sectional clustering on 885 subjects (original dataset) to identify disease subtypes, we identified optimal longitudinal trajectories using hybrid machine learning systems (HMLS), including principal component analysis (PCA) + K-Means algorithms (KMA) followed by Bayesian information criterion (BIC), Calinski-Harabatz criterion (CHC), and elbow criterion (EC). Subsequently, prediction of the identified trajectories from early year data was performed using multiple HMLSs including 16 Dimension Reduction Algorithms (DRA) and 10 classification algorithms.METHODSWe studied PD-subjects derived from longitudinal datasets (years 0, 1, 2 & 4; Parkinson's Progressive Marker Initiative). We extracted and analyzed 981 features, including motor, non-motor, and radiomics features extracted for each region-of-interest (ROIs: left/right caudate and putamen) using our standardized standardized environment for radiomics analysis (SERA) radiomics software. Segmentation of ROIs on dopamine transposer - single photon emission computed tomography (DAT SPECT) images were performed via magnetic resonance images (MRI). After performing cross-sectional clustering on 885 subjects (original dataset) to identify disease subtypes, we identified optimal longitudinal trajectories using hybrid machine learning systems (HMLS), including principal component analysis (PCA) + K-Means algorithms (KMA) followed by Bayesian information criterion (BIC), Calinski-Harabatz criterion (CHC), and elbow criterion (EC). Subsequently, prediction of the identified trajectories from early year data was performed using multiple HMLSs including 16 Dimension Reduction Algorithms (DRA) and 10 classification algorithms.We identified 3 distinct progression trajectories. Hotelling's t squared test (HTST) showed that the identified trajectories were distinct. The trajectories included those with (I, II) disease escalation (2 trajectories, 27% and 38% of patients) and (III) stable disease (1 trajectory, 35% of patients). For trajectory prediction from early year data, HMLSs including the stochastic neighbor embedding algorithm (SNEA, as a DRA) as well as locally linear embedding algorithm (LLEA, as a DRA), linked with the new probabilistic neural network classifier (NPNNC, as a classifier), resulted in accuracies of 78.4% and 79.2% respectively, while other HMLSs such as SNEA + Lib_SVM (library for support vector machines) and t_SNE (t-distributed stochastic neighbor embedding) + NPNNC resulted in 76.5% and 76.1% respectively.RESULTSWe identified 3 distinct progression trajectories. Hotelling's t squared test (HTST) showed that the identified trajectories were distinct. The trajectories included those with (I, II) disease escalation (2 trajectories, 27% and 38% of patients) and (III) stable disease (1 trajectory, 35% of patients). For trajectory prediction from early year data, HMLSs including the stochastic neighbor embedding algorithm (SNEA, as a DRA) as well as locally linear embedding algorithm (LLEA, as a DRA), linked with the new probabilistic neural network classifier (NPNNC, as a classifier), resulted in accuracies of 78.4% and 79.2% respectively, while other HMLSs such as SNEA + Lib_SVM (library for support vector machines) and t_SNE (t-distributed stochastic neighbor embedding) + NPNNC resulted in 76.5% and 76.1% respectively.This study moves beyond cross-sectional PD subtyping to clustering of longitudinal disease trajectories. We conclude that combining medical information with SPECT-based radiomics features, and optimal utilization of HMLSs, can identify distinct disease trajectories in PD patients, and enable effective prediction of disease trajectories from early year data.CONCLUSIONSThis study moves beyond cross-sectional PD subtyping to clustering of longitudinal disease trajectories. We conclude that combining medical information with SPECT-based radiomics features, and optimal utilization of HMLSs, can identify distinct disease trajectories in PD patients, and enable effective prediction of disease trajectories from early year data. |
| Author | Rahmim, Arman Soltanian-Zadeh, Hamid Salmanpour, Mohammad R Shamsaei, Mojtaba Hajianfar, Ghasem |
| Author_xml | – sequence: 1 givenname: Mohammad R surname: Salmanpour fullname: Salmanpour, Mohammad R organization: Department of Physics & Astronomy, University of British Columbia, Vancouver BC, Canada – sequence: 2 givenname: Mojtaba surname: Shamsaei fullname: Shamsaei, Mojtaba organization: Department of Energy Engineering and Physics, Amirkabir University of Technology, Tehran, Iran – sequence: 3 givenname: Ghasem surname: Hajianfar fullname: Hajianfar, Ghasem organization: Rajaie Cardiovascular Medical and Research Center, Iran University of Medical Science, Tehran, Iran – sequence: 4 givenname: Hamid surname: Soltanian-Zadeh fullname: Soltanian-Zadeh, Hamid organization: Departments of Radiology and Research Administration, Henry Ford Health System, Detroit, USA – sequence: 5 givenname: Arman surname: Rahmim fullname: Rahmim, Arman organization: Department of Radiology, University of British Columbia, Vancouver BC, Canada |
| BackLink | https://www.ncbi.nlm.nih.gov/pubmed/35111593$$D View this record in MEDLINE/PubMed |
| BookMark | eNo1kD1PwzAQhj0U0QLdmJE3WAKxHedjRBVfUiUYYK4c-9waErv1JUMXfjuuWm6595UePTrdBZn44IGQa5bfc5aL6mHnesw4ywouJ2TGORcpNnxK5ojfeZqqZhXLz8lUSMaYbMSM_C6DX7thNM6rjupuxAGi82uqUt-jwxQM3UYwTg8ueBos_VDxx3kM_hapcQgKIRFhHQHxgIx4EERlXOidPho2-zY6Q3ulN84D7UBFn6grcmZVhzA_7Uvy9fz0uXjNlu8vb4vHZaZFXQyZtVA2AHWVC6NNmRd1o6zWtTQsN6WUBci2KpSpWisbW4O0XOlGSNAla7ll_JLcHb3pzt0IOKx6hxq6TnkII654ySUvSymahN6c0LHtway20fUq7lf_P-N_lVpydw |
| CitedBy_id | crossref_primary_10_1111_ene_16026 crossref_primary_10_1186_s40658_024_00651_1 crossref_primary_10_1007_s10278_025_01583_7 crossref_primary_10_3389_fneur_2025_1612222 crossref_primary_10_3389_fnins_2022_1012287 crossref_primary_10_1016_j_ejmp_2023_102647 crossref_primary_10_1016_j_health_2023_100181 crossref_primary_10_1186_s43055_025_01552_8 crossref_primary_10_1177_17085381241262575 crossref_primary_10_1016_j_crad_2025_106921 crossref_primary_10_3390_diagnostics13101696 crossref_primary_10_3390_diagnostics13101691 crossref_primary_10_1007_s40815_023_01665_0 crossref_primary_10_1016_j_compbiomed_2025_110156 crossref_primary_10_1007_s11042_023_15414_w crossref_primary_10_1016_j_cmpb_2023_107714 crossref_primary_10_1016_j_wneu_2023_07_103 crossref_primary_10_1002_ima_22868 crossref_primary_10_1016_j_brainres_2023_148675 crossref_primary_10_1038_s41531_025_01127_4 crossref_primary_10_2174_1574893618666230406085947 crossref_primary_10_1002_mds_29519 crossref_primary_10_1109_JBHI_2024_3482180 crossref_primary_10_1007_s00330_024_10886_2 crossref_primary_10_1016_j_phrs_2023_106984 |
| ContentType | Journal Article |
| Copyright | 2022 Quantitative Imaging in Medicine and Surgery. All rights reserved. |
| Copyright_xml | – notice: 2022 Quantitative Imaging in Medicine and Surgery. All rights reserved. |
| DBID | NPM 7X8 |
| DOI | 10.21037/qims-21-425 |
| DatabaseName | PubMed MEDLINE - Academic |
| DatabaseTitle | PubMed MEDLINE - Academic |
| DatabaseTitleList | PubMed MEDLINE - Academic |
| Database_xml | – sequence: 1 dbid: NPM name: PubMed url: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed sourceTypes: Index Database – sequence: 2 dbid: 7X8 name: MEDLINE - Academic url: https://search.proquest.com/medline sourceTypes: Aggregation Database |
| DeliveryMethod | no_fulltext_linktorsrc |
| Discipline | Medicine |
| ExternalDocumentID | 35111593 |
| Genre | Journal Article |
| GroupedDBID | 53G AAKDD ALMA_UNASSIGNED_HOLDINGS DIK HYE M~E NPM OK1 RPM 7X8 |
| ID | FETCH-LOGICAL-c384t-ffe69ee8703dcd60489afcc85d10d6554e5b74ad7bf59f8e5f2ac935ec61b2f12 |
| IEDL.DBID | 7X8 |
| ISICitedReferencesCount | 30 |
| ISICitedReferencesURI | http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000697249400001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| ISSN | 2223-4292 |
| IngestDate | Fri Sep 05 08:18:02 EDT 2025 Thu Jan 02 22:53:56 EST 2025 |
| IsDoiOpenAccess | false |
| IsOpenAccess | true |
| IsPeerReviewed | true |
| IsScholarly | true |
| Issue | 2 |
| Keywords | outcome prediction Parkinson’s disease (PD) hybrid machine learning methods longitudinal clustering |
| Language | English |
| License | 2022 Quantitative Imaging in Medicine and Surgery. All rights reserved. |
| LinkModel | DirectLink |
| MergedId | FETCHMERGED-LOGICAL-c384t-ffe69ee8703dcd60489afcc85d10d6554e5b74ad7bf59f8e5f2ac935ec61b2f12 |
| Notes | ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 23 |
| OpenAccessLink | https://qims.amegroups.com/article/viewFile/78913/pdf |
| PMID | 35111593 |
| PQID | 2625266539 |
| PQPubID | 23479 |
| ParticipantIDs | proquest_miscellaneous_2625266539 pubmed_primary_35111593 |
| PublicationCentury | 2000 |
| PublicationDate | 2022-Feb 20220201 |
| PublicationDateYYYYMMDD | 2022-02-01 |
| PublicationDate_xml | – month: 02 year: 2022 text: 2022-Feb |
| PublicationDecade | 2020 |
| PublicationPlace | China |
| PublicationPlace_xml | – name: China |
| PublicationTitle | Quantitative imaging in medicine and surgery |
| PublicationTitleAlternate | Quant Imaging Med Surg |
| PublicationYear | 2022 |
| SSID | ssj0000781710 |
| Score | 2.3564255 |
| Snippet | We employed machine learning approaches to (I) determine distinct progression trajectories in Parkinson's disease (PD) (unsupervised clustering task), and (II)... |
| SourceID | proquest pubmed |
| SourceType | Aggregation Database Index Database |
| StartPage | 906 |
| Title | Longitudinal clustering analysis and prediction of Parkinson's disease progression using radiomics and hybrid machine learning |
| URI | https://www.ncbi.nlm.nih.gov/pubmed/35111593 https://www.proquest.com/docview/2625266539 |
| Volume | 12 |
| WOSCitedRecordID | wos000697249400001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| hasFullText | |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV3LSgMxFA1qRdz4ftQXEQRXwXk0M5OViFhctKULle5KJo9asJ220wpu_HbvzaS6EgQ3w2wSQibJPZN77jmEXEnOrQikYElq4QeFa8mECmLGuTEyDyW3TnbxpZV2OlmvJ7r-wq30tMrlmegOal0ovCO_iQCoQzDhsbidTBm6RmF21VtorJJaDFAGKV1pL_u-Y0Ehm9QJEmAUZOjMVHHfI6frMx2OShbBCCP-O750caa5_d8R7pAtjzDpXbUkdsmKGe-RjbbPoe-Tz1aBHkULjX5YVL0tUCoBAhiVXp8EXjSdzLABfjVaWIq10a5M7LqkPqVDHbOrUvWgyJ4f0JnUQyxyrnp4_cBiMDpybE1DvT3F4IA8Nx-e7h-Zd2FgKs4ac2atSYQxsK9jrXQCO15Iq1TGdRjoBNCI4XnakDrNLRc2M9xGUomYG5WEeWTD6JCsjYuxOSa0kXAbaNGQEoCiCo0Uea6DQIbQLfQt6-RyObN9WOWYupBjUyzK_s_c1slR9Xn6k0qOo4-pUABl8ckfWp-SzQjrFxzt-ozULOxxc07W1ft8WM4u3PKBZ6fb_gJTSdRJ |
| linkProvider | ProQuest |
| 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=Longitudinal+clustering+analysis+and+prediction+of+Parkinson%27s+disease+progression+using+radiomics+and+hybrid+machine+learning&rft.jtitle=Quantitative+imaging+in+medicine+and+surgery&rft.au=Salmanpour%2C+Mohammad+R&rft.au=Shamsaei%2C+Mojtaba&rft.au=Hajianfar%2C+Ghasem&rft.au=Soltanian-Zadeh%2C+Hamid&rft.date=2022-02-01&rft.issn=2223-4292&rft.volume=12&rft.issue=2&rft.spage=906&rft_id=info:doi/10.21037%2Fqims-21-425&rft_id=info%3Apmid%2F35111593&rft_id=info%3Apmid%2F35111593&rft.externalDocID=35111593 |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=2223-4292&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=2223-4292&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=2223-4292&client=summon |