Loading…
Functional Data Analysis of Dynamic PET Data
One application of positron emission tomography (PET), a nuclear imaging technique, in neuroscience involves in vivo estimation of the density of various proteins (often, neuroreceptors) in the brain. PET scanning begins with the injection of a radiolabeled tracer that binds preferentially to the ta...
Saved in:
Published in: | Journal of the American Statistical Association 2019-04, Vol.114 (526), p.595-609 |
---|---|
Main Authors: | , , |
Format: | Article |
Language: | English |
Subjects: | |
Citations: | Items that this one cites Items that cite this one |
Online Access: | Get full text |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
cited_by | cdi_FETCH-LOGICAL-c529t-531587b50506a8e9716cd880450c1893b19340697171fbcd343b32ce1b77ee243 |
---|---|
cites | cdi_FETCH-LOGICAL-c529t-531587b50506a8e9716cd880450c1893b19340697171fbcd343b32ce1b77ee243 |
container_end_page | 609 |
container_issue | 526 |
container_start_page | 595 |
container_title | Journal of the American Statistical Association |
container_volume | 114 |
creator | Chen, Yakuan Goldsmith, Jeff Ogden, R. Todd |
description | One application of positron emission tomography (PET), a nuclear imaging technique, in neuroscience involves in vivo estimation of the density of various proteins (often, neuroreceptors) in the brain. PET scanning begins with the injection of a radiolabeled tracer that binds preferentially to the target protein; tracer molecules are then continuously delivered to the brain via the bloodstream. By detecting the radioactive decay of the tracer over time, dynamic PET data are constructed to reflect the concentration of the target protein in the brain at each time. The fundamental problem in the analysis of dynamic PET data involves estimating the impulse response function (IRF), which is necessary for describing the binding behavior of the injected radiotracer. Virtually all existing methods have three common aspects: summarizing the entire IRF with a single scalar measure; modeling each subject separately; and the imposition of parametric restrictions on the IRF. In contrast, we propose a functional data analytic approach that regards each subject's IRF as the basic analysis unit, models multiple subjects simultaneously, and estimates the IRF nonparametrically. We pose our model as a linear mixed effect model in which population level fixed effects and subject-specific random effects are expanded using a B-spline basis. Shrinkage and roughness penalties are incorporated in the model to enforce identifiability and smoothness of the estimated curves, respectively, while monotonicity and nonnegativity constraints impose biological information on estimates. We illustrate this approach by applying it to clinical PET data with subjects belonging to three diagnosic groups. We explore differences among groups by means of pointwise confidence intervals of the estimated mean curves based on bootstrap samples. Supplementary materials for this article are available online. |
doi_str_mv | 10.1080/01621459.2018.1497495 |
format | article |
fullrecord | <record><control><sourceid>proquest_pubme</sourceid><recordid>TN_cdi_pubmedcentral_primary_oai_pubmedcentral_nih_gov_6707736</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2239515491</sourcerecordid><originalsourceid>FETCH-LOGICAL-c529t-531587b50506a8e9716cd880450c1893b19340697171fbcd343b32ce1b77ee243</originalsourceid><addsrcrecordid>eNp9kU9P2zAYhy00REu3j8AUaRcOpPiN7di-TCAKDKkSHJi0m-W4DnOVxJ2dMPXbz6EFbTvgiy2_z_vznwehE8BzwAKfYygLoEzOCwxiDlRyKtkBmgIjPC84_fEBTUcmH6EJOo5xjdPgQhyhCQFKE0-m6Oxm6EzvfKebbKF7nV2m1Ta6mPk6W2w73TqTPVw_vhQ_osNaN9F-2s8z9P3m-vHqW768v727ulzmhhWyzxkBJnjFMMOlFlZyKM1KCEwZNiAkqUASisu0z6GuzIpQUpHCWKg4t7agZIa-7nI3Q9XalbFdH3SjNsG1OmyV1079W-ncT_Xkn1XJMeekTAGn-4Dgfw029qp10dim0Z31Q1RFITAjJZQkoV_-Q9d-COkTRopIBoxKSBTbUSb4GIOt3y4DWI0-1KsPNfpQex-p7_PfL3nrehWQgIsd4Lrah1b_9qFZqV5vGx_qoDvjYoLfPeMPiFKWvA</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2239515491</pqid></control><display><type>article</type><title>Functional Data Analysis of Dynamic PET Data</title><source>International Bibliography of the Social Sciences (IBSS)</source><source>Taylor and Francis Science and Technology Collection</source><creator>Chen, Yakuan ; Goldsmith, Jeff ; Ogden, R. Todd</creator><creatorcontrib>Chen, Yakuan ; Goldsmith, Jeff ; Ogden, R. Todd</creatorcontrib><description>One application of positron emission tomography (PET), a nuclear imaging technique, in neuroscience involves in vivo estimation of the density of various proteins (often, neuroreceptors) in the brain. PET scanning begins with the injection of a radiolabeled tracer that binds preferentially to the target protein; tracer molecules are then continuously delivered to the brain via the bloodstream. By detecting the radioactive decay of the tracer over time, dynamic PET data are constructed to reflect the concentration of the target protein in the brain at each time. The fundamental problem in the analysis of dynamic PET data involves estimating the impulse response function (IRF), which is necessary for describing the binding behavior of the injected radiotracer. Virtually all existing methods have three common aspects: summarizing the entire IRF with a single scalar measure; modeling each subject separately; and the imposition of parametric restrictions on the IRF. In contrast, we propose a functional data analytic approach that regards each subject's IRF as the basic analysis unit, models multiple subjects simultaneously, and estimates the IRF nonparametrically. We pose our model as a linear mixed effect model in which population level fixed effects and subject-specific random effects are expanded using a B-spline basis. Shrinkage and roughness penalties are incorporated in the model to enforce identifiability and smoothness of the estimated curves, respectively, while monotonicity and nonnegativity constraints impose biological information on estimates. We illustrate this approach by applying it to clinical PET data with subjects belonging to three diagnosic groups. We explore differences among groups by means of pointwise confidence intervals of the estimated mean curves based on bootstrap samples. Supplementary materials for this article are available online.</description><identifier>ISSN: 0162-1459</identifier><identifier>EISSN: 1537-274X</identifier><identifier>DOI: 10.1080/01621459.2018.1497495</identifier><identifier>PMID: 31447493</identifier><language>eng</language><publisher>United States: Taylor & Francis</publisher><subject>Bootstrap ; Brain ; Confidence intervals ; Constrained estimation ; Data analysis ; Density ; Emission analysis ; Emissions ; Function-on-scalar regression ; Imposition ; Impulse response ; In vivo methods and tests ; Nonparametric ; Nonparametric statistics ; Penalties ; Positron emission ; Positron emission tomography ; Proteins ; Radioactive decay ; Radioactive tracers ; Random effects ; Regression analysis ; Response functions ; Shrinkage ; Smoothness ; Splines ; Statistical methods ; Statistics ; Tomography</subject><ispartof>Journal of the American Statistical Association, 2019-04, Vol.114 (526), p.595-609</ispartof><rights>2018 American Statistical Association 2018</rights><rights>2018 American Statistical Association</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c529t-531587b50506a8e9716cd880450c1893b19340697171fbcd343b32ce1b77ee243</citedby><cites>FETCH-LOGICAL-c529t-531587b50506a8e9716cd880450c1893b19340697171fbcd343b32ce1b77ee243</cites><orcidid>0000-0002-0331-0819</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>230,314,780,784,885,27924,27925,33223</link.rule.ids><backlink>$$Uhttps://www.ncbi.nlm.nih.gov/pubmed/31447493$$D View this record in MEDLINE/PubMed$$Hfree_for_read</backlink></links><search><creatorcontrib>Chen, Yakuan</creatorcontrib><creatorcontrib>Goldsmith, Jeff</creatorcontrib><creatorcontrib>Ogden, R. Todd</creatorcontrib><title>Functional Data Analysis of Dynamic PET Data</title><title>Journal of the American Statistical Association</title><addtitle>J Am Stat Assoc</addtitle><description>One application of positron emission tomography (PET), a nuclear imaging technique, in neuroscience involves in vivo estimation of the density of various proteins (often, neuroreceptors) in the brain. PET scanning begins with the injection of a radiolabeled tracer that binds preferentially to the target protein; tracer molecules are then continuously delivered to the brain via the bloodstream. By detecting the radioactive decay of the tracer over time, dynamic PET data are constructed to reflect the concentration of the target protein in the brain at each time. The fundamental problem in the analysis of dynamic PET data involves estimating the impulse response function (IRF), which is necessary for describing the binding behavior of the injected radiotracer. Virtually all existing methods have three common aspects: summarizing the entire IRF with a single scalar measure; modeling each subject separately; and the imposition of parametric restrictions on the IRF. In contrast, we propose a functional data analytic approach that regards each subject's IRF as the basic analysis unit, models multiple subjects simultaneously, and estimates the IRF nonparametrically. We pose our model as a linear mixed effect model in which population level fixed effects and subject-specific random effects are expanded using a B-spline basis. Shrinkage and roughness penalties are incorporated in the model to enforce identifiability and smoothness of the estimated curves, respectively, while monotonicity and nonnegativity constraints impose biological information on estimates. We illustrate this approach by applying it to clinical PET data with subjects belonging to three diagnosic groups. We explore differences among groups by means of pointwise confidence intervals of the estimated mean curves based on bootstrap samples. Supplementary materials for this article are available online.</description><subject>Bootstrap</subject><subject>Brain</subject><subject>Confidence intervals</subject><subject>Constrained estimation</subject><subject>Data analysis</subject><subject>Density</subject><subject>Emission analysis</subject><subject>Emissions</subject><subject>Function-on-scalar regression</subject><subject>Imposition</subject><subject>Impulse response</subject><subject>In vivo methods and tests</subject><subject>Nonparametric</subject><subject>Nonparametric statistics</subject><subject>Penalties</subject><subject>Positron emission</subject><subject>Positron emission tomography</subject><subject>Proteins</subject><subject>Radioactive decay</subject><subject>Radioactive tracers</subject><subject>Random effects</subject><subject>Regression analysis</subject><subject>Response functions</subject><subject>Shrinkage</subject><subject>Smoothness</subject><subject>Splines</subject><subject>Statistical methods</subject><subject>Statistics</subject><subject>Tomography</subject><issn>0162-1459</issn><issn>1537-274X</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2019</creationdate><recordtype>article</recordtype><sourceid>8BJ</sourceid><recordid>eNp9kU9P2zAYhy00REu3j8AUaRcOpPiN7di-TCAKDKkSHJi0m-W4DnOVxJ2dMPXbz6EFbTvgiy2_z_vznwehE8BzwAKfYygLoEzOCwxiDlRyKtkBmgIjPC84_fEBTUcmH6EJOo5xjdPgQhyhCQFKE0-m6Oxm6EzvfKebbKF7nV2m1Ta6mPk6W2w73TqTPVw_vhQ_osNaN9F-2s8z9P3m-vHqW768v727ulzmhhWyzxkBJnjFMMOlFlZyKM1KCEwZNiAkqUASisu0z6GuzIpQUpHCWKg4t7agZIa-7nI3Q9XalbFdH3SjNsG1OmyV1079W-ncT_Xkn1XJMeekTAGn-4Dgfw029qp10dim0Z31Q1RFITAjJZQkoV_-Q9d-COkTRopIBoxKSBTbUSb4GIOt3y4DWI0-1KsPNfpQex-p7_PfL3nrehWQgIsd4Lrah1b_9qFZqV5vGx_qoDvjYoLfPeMPiFKWvA</recordid><startdate>20190403</startdate><enddate>20190403</enddate><creator>Chen, Yakuan</creator><creator>Goldsmith, Jeff</creator><creator>Ogden, R. Todd</creator><general>Taylor & Francis</general><general>Taylor & Francis Ltd</general><scope>NPM</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>8BJ</scope><scope>FQK</scope><scope>JBE</scope><scope>K9.</scope><scope>7X8</scope><scope>5PM</scope><orcidid>https://orcid.org/0000-0002-0331-0819</orcidid></search><sort><creationdate>20190403</creationdate><title>Functional Data Analysis of Dynamic PET Data</title><author>Chen, Yakuan ; Goldsmith, Jeff ; Ogden, R. Todd</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c529t-531587b50506a8e9716cd880450c1893b19340697171fbcd343b32ce1b77ee243</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2019</creationdate><topic>Bootstrap</topic><topic>Brain</topic><topic>Confidence intervals</topic><topic>Constrained estimation</topic><topic>Data analysis</topic><topic>Density</topic><topic>Emission analysis</topic><topic>Emissions</topic><topic>Function-on-scalar regression</topic><topic>Imposition</topic><topic>Impulse response</topic><topic>In vivo methods and tests</topic><topic>Nonparametric</topic><topic>Nonparametric statistics</topic><topic>Penalties</topic><topic>Positron emission</topic><topic>Positron emission tomography</topic><topic>Proteins</topic><topic>Radioactive decay</topic><topic>Radioactive tracers</topic><topic>Random effects</topic><topic>Regression analysis</topic><topic>Response functions</topic><topic>Shrinkage</topic><topic>Smoothness</topic><topic>Splines</topic><topic>Statistical methods</topic><topic>Statistics</topic><topic>Tomography</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Chen, Yakuan</creatorcontrib><creatorcontrib>Goldsmith, Jeff</creatorcontrib><creatorcontrib>Ogden, R. Todd</creatorcontrib><collection>PubMed</collection><collection>CrossRef</collection><collection>International Bibliography of the Social Sciences (IBSS)</collection><collection>International Bibliography of the Social Sciences</collection><collection>International Bibliography of the Social Sciences</collection><collection>ProQuest Health & Medical Complete (Alumni)</collection><collection>MEDLINE - Academic</collection><collection>PubMed Central (Full Participant titles)</collection><jtitle>Journal of the American Statistical Association</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Chen, Yakuan</au><au>Goldsmith, Jeff</au><au>Ogden, R. Todd</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Functional Data Analysis of Dynamic PET Data</atitle><jtitle>Journal of the American Statistical Association</jtitle><addtitle>J Am Stat Assoc</addtitle><date>2019-04-03</date><risdate>2019</risdate><volume>114</volume><issue>526</issue><spage>595</spage><epage>609</epage><pages>595-609</pages><issn>0162-1459</issn><eissn>1537-274X</eissn><abstract>One application of positron emission tomography (PET), a nuclear imaging technique, in neuroscience involves in vivo estimation of the density of various proteins (often, neuroreceptors) in the brain. PET scanning begins with the injection of a radiolabeled tracer that binds preferentially to the target protein; tracer molecules are then continuously delivered to the brain via the bloodstream. By detecting the radioactive decay of the tracer over time, dynamic PET data are constructed to reflect the concentration of the target protein in the brain at each time. The fundamental problem in the analysis of dynamic PET data involves estimating the impulse response function (IRF), which is necessary for describing the binding behavior of the injected radiotracer. Virtually all existing methods have three common aspects: summarizing the entire IRF with a single scalar measure; modeling each subject separately; and the imposition of parametric restrictions on the IRF. In contrast, we propose a functional data analytic approach that regards each subject's IRF as the basic analysis unit, models multiple subjects simultaneously, and estimates the IRF nonparametrically. We pose our model as a linear mixed effect model in which population level fixed effects and subject-specific random effects are expanded using a B-spline basis. Shrinkage and roughness penalties are incorporated in the model to enforce identifiability and smoothness of the estimated curves, respectively, while monotonicity and nonnegativity constraints impose biological information on estimates. We illustrate this approach by applying it to clinical PET data with subjects belonging to three diagnosic groups. We explore differences among groups by means of pointwise confidence intervals of the estimated mean curves based on bootstrap samples. Supplementary materials for this article are available online.</abstract><cop>United States</cop><pub>Taylor & Francis</pub><pmid>31447493</pmid><doi>10.1080/01621459.2018.1497495</doi><tpages>15</tpages><orcidid>https://orcid.org/0000-0002-0331-0819</orcidid><oa>free_for_read</oa></addata></record> |
fulltext | fulltext |
identifier | ISSN: 0162-1459 |
ispartof | Journal of the American Statistical Association, 2019-04, Vol.114 (526), p.595-609 |
issn | 0162-1459 1537-274X |
language | eng |
recordid | cdi_pubmedcentral_primary_oai_pubmedcentral_nih_gov_6707736 |
source | International Bibliography of the Social Sciences (IBSS); Taylor and Francis Science and Technology Collection |
subjects | Bootstrap Brain Confidence intervals Constrained estimation Data analysis Density Emission analysis Emissions Function-on-scalar regression Imposition Impulse response In vivo methods and tests Nonparametric Nonparametric statistics Penalties Positron emission Positron emission tomography Proteins Radioactive decay Radioactive tracers Random effects Regression analysis Response functions Shrinkage Smoothness Splines Statistical methods Statistics Tomography |
title | Functional Data Analysis of Dynamic PET Data |
url | http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-24T19%3A56%3A19IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_pubme&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Functional%20Data%20Analysis%20of%20Dynamic%20PET%20Data&rft.jtitle=Journal%20of%20the%20American%20Statistical%20Association&rft.au=Chen,%20Yakuan&rft.date=2019-04-03&rft.volume=114&rft.issue=526&rft.spage=595&rft.epage=609&rft.pages=595-609&rft.issn=0162-1459&rft.eissn=1537-274X&rft_id=info:doi/10.1080/01621459.2018.1497495&rft_dat=%3Cproquest_pubme%3E2239515491%3C/proquest_pubme%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-c529t-531587b50506a8e9716cd880450c1893b19340697171fbcd343b32ce1b77ee243%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_pqid=2239515491&rft_id=info:pmid/31447493&rfr_iscdi=true |