{"id":"https://openalex.org/W7163997470","doi":"https://doi.org/10.48550/arxiv.2606.08751","title":"Less Is More: Training-Free Acceleration Framework of 3D Diffusion Models for Low-Count PET Denoising via Global-Local Trajectory Reduction","display_name":"Less Is More: Training-Free Acceleration Framework of 3D Diffusion Models for Low-Count PET Denoising via Global-Local Trajectory Reduction","publication_year":2026,"publication_date":"2026-06-07","ids":{"openalex":"https://openalex.org/W7163997470","doi":"https://doi.org/10.48550/arxiv.2606.08751"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.08751","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.08751","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2606.08751","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5138230594","display_name":"Yuhan Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Yuhan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5112670597","display_name":"Scott Leonard","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Leonard, Scott M.","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138260580","display_name":"Marlee Crews","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Crews, Marlee","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138211756","display_name":"Muhannad Fadhel","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Fadhel, Muhannad","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5004033797","display_name":"Jinkui Hao","orcid":"https://orcid.org/0000-0002-7101-961X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hao, Jinkui","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138203975","display_name":"Tianqi Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Tianqi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138229798","display_name":"Ryan J. Avery","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Avery, Ryan J.","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5138265457","display_name":"Bo Zhou","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhou, Bo","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10522","display_name":"Medical Imaging Techniques and Applications","score":0.988099992275238,"subfield":{"id":"https://openalex.org/subfields/2741","display_name":"Radiology, Nuclear Medicine and Imaging"},"field":{"id":"https://openalex.org/fields/27","display_name":"Medicine"},"domain":{"id":"https://openalex.org/domains/4","display_name":"Health Sciences"}},"topics":[{"id":"https://openalex.org/T10522","display_name":"Medical Imaging Techniques and Applications","score":0.988099992275238,"subfield":{"id":"https://openalex.org/subfields/2741","display_name":"Radiology, Nuclear Medicine and Imaging"},"field":{"id":"https://openalex.org/fields/27","display_name":"Medicine"},"domain":{"id":"https://openalex.org/domains/4","display_name":"Health Sciences"}},{"id":"https://openalex.org/T10129","display_name":"Glioma Diagnosis and Treatment","score":0.0012000000569969416,"subfield":{"id":"https://openalex.org/subfields/2716","display_name":"Genetics"},"field":{"id":"https://openalex.org/fields/27","display_name":"Medicine"},"domain":{"id":"https://openalex.org/domains/4","display_name":"Health Sciences"}},{"id":"https://openalex.org/T12422","display_name":"Radiomics and Machine Learning in Medical Imaging","score":0.0012000000569969416,"subfield":{"id":"https://openalex.org/subfields/2741","display_name":"Radiology, Nuclear Medicine and Imaging"},"field":{"id":"https://openalex.org/fields/27","display_name":"Medicine"},"domain":{"id":"https://openalex.org/domains/4","display_name":"Health Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/noise-reduction","display_name":"Noise reduction","score":0.7756999731063843},{"id":"https://openalex.org/keywords/reduction","display_name":"Reduction (mathematics)","score":0.5092999935150146},{"id":"https://openalex.org/keywords/acceleration","display_name":"Acceleration","score":0.5024999976158142},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.49470001459121704},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.45969998836517334},{"id":"https://openalex.org/keywords/trajectory","display_name":"Trajectory","score":0.3878999948501587},{"id":"https://openalex.org/keywords/process","display_name":"Process (computing)","score":0.3813999891281128},{"id":"https://openalex.org/keywords/iterative-reconstruction","display_name":"Iterative reconstruction","score":0.3571999967098236},{"id":"https://openalex.org/keywords/noise","display_name":"Noise (video)","score":0.3560999929904938}],"concepts":[{"id":"https://openalex.org/C163294075","wikidata":"https://www.wikidata.org/wiki/Q581861","display_name":"Noise reduction","level":2,"score":0.7756999731063843},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6315000057220459},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.5092999935150146},{"id":"https://openalex.org/C117896860","wikidata":"https://www.wikidata.org/wiki/Q11376","display_name":"Acceleration","level":2,"score":0.5024999976158142},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.49470001459121704},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.47600001096725464},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.45969998836517334},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.427700012922287},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.39489999413490295},{"id":"https://openalex.org/C13662910","wikidata":"https://www.wikidata.org/wiki/Q193139","display_name":"Trajectory","level":2,"score":0.3878999948501587},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.3813999891281128},{"id":"https://openalex.org/C141379421","wikidata":"https://www.wikidata.org/wiki/Q6094427","display_name":"Iterative reconstruction","level":2,"score":0.3571999967098236},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.3560999929904938},{"id":"https://openalex.org/C68710425","wikidata":"https://www.wikidata.org/wiki/Q5275442","display_name":"Diffusion process","level":3,"score":0.34689998626708984},{"id":"https://openalex.org/C140779682","wikidata":"https://www.wikidata.org/wiki/Q210868","display_name":"Sampling (signal processing)","level":3,"score":0.34630000591278076},{"id":"https://openalex.org/C106131492","wikidata":"https://www.wikidata.org/wiki/Q3072260","display_name":"Filter (signal processing)","level":2,"score":0.3352999985218048},{"id":"https://openalex.org/C69357855","wikidata":"https://www.wikidata.org/wiki/Q163214","display_name":"Diffusion","level":2,"score":0.31360000371932983},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.3125999867916107},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.3098999857902527},{"id":"https://openalex.org/C143587482","wikidata":"https://www.wikidata.org/wiki/Q1543216","display_name":"Iterative and incremental development","level":2,"score":0.2922999858856201},{"id":"https://openalex.org/C149550507","wikidata":"https://www.wikidata.org/wiki/Q899360","display_name":"Diffusion MRI","level":3,"score":0.2906000018119812},{"id":"https://openalex.org/C52740198","wikidata":"https://www.wikidata.org/wiki/Q1539564","display_name":"Importance sampling","level":3,"score":0.28189998865127563},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.28029999136924744},{"id":"https://openalex.org/C159694833","wikidata":"https://www.wikidata.org/wiki/Q2321565","display_name":"Iterative method","level":2,"score":0.2759000062942505},{"id":"https://openalex.org/C55020928","wikidata":"https://www.wikidata.org/wiki/Q3813865","display_name":"Image quality","level":3,"score":0.26339998841285706},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.25929999351501465},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.2581000030040741}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.08751","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.08751","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2606.08751","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.08751","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[{"score":0.6972150206565857,"display_name":"Peace, Justice and strong institutions","id":"https://metadata.un.org/sdg/16"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Accurate":[0],"quantification":[1],"and":[2,13,28,107,156,192,200,231],"uptake":[3],"measurement":[4],"in":[5],"PET":[6,19,50,68,95,188],"are":[7],"critical":[8],"for":[9,47],"assessing":[10],"disease":[11],"progression":[12],"supporting":[14],"clinical":[15,33,77,229],"decision-making.":[16],"While":[17],"high-count":[18,53],"provides":[20],"reliable":[21],"image":[22,179],"quality,":[23,54],"the":[24,36,131,145,150,183,220],"associated":[25],"radiation":[26],"dose":[27],"prolonged":[29],"acquisition":[30],"remain":[31],"significant":[32],"concerns,":[34],"motivating":[35],"adoption":[37],"of":[38,144,152,206,210],"low-count":[39,49,146],"protocols.":[40],"Diffusion-model-based":[41],"methods":[42],"have":[43],"demonstrated":[44],"strong":[45],"potential":[46],"restoring":[48],"to":[51,65,110,219],"near":[52],"but":[55],"their":[56],"iterative":[57],"sampling":[58],"procedure":[59],"becomes":[60],"prohibitively":[61],"expensive":[62],"when":[63],"applied":[64],"high-resolution":[66],"3D":[67,94],"volumes,":[69],"introducing":[70],"substantial":[71],"inference":[72],"latency":[73],"that":[74,90,129,163],"limits":[75],"practical":[76],"deployment.":[78],"To":[79],"address":[80],"these":[81],"challenges,":[82],"we":[83,120],"propose":[84],"a":[85,123,141,158],"training-free":[86],"Global-Local":[87],"Skipping":[88],"Strategy":[89],"accelerates":[91],"diffusion":[92,112,133],"model-based":[93],"denoising":[96,125,138,154,171],"while":[97,177],"simultaneously":[98],"improving":[99],"reconstruction":[100,216],"quality.":[101,234],"The":[102],"proposed":[103,184],"method":[104],"is":[105],"plug-and-play":[106],"directly":[108],"applicable":[109],"pre-trained":[111],"models":[113],"without":[114],"retraining":[115],"or":[116,214],"architectural":[117],"modification.":[118],"Specifically,":[119],"introduce:":[121],"(i)":[122],"global":[124],"step":[126,139],"skipping":[127],"strategy":[128],"initializes":[130],"reverse":[132],"process":[134],"from":[135,190],"an":[136,208],"intermediate":[137],"using":[140],"noise-consistent":[142],"transformation":[143],"input,":[147],"substantially":[148],"reducing":[149,174],"number":[151],"required":[153],"steps;":[155],"(ii)":[157],"local":[159],"feature":[160],"reuse":[161],"shortcut":[162],"reuses":[164],"slowly-varying":[165],"high-level":[166],"U-Net":[167],"features":[168],"across":[169],"neighboring":[170],"steps,":[172],"further":[173,226],"per-step":[175],"computation":[176],"preserving":[178],"fidelity.":[180],"We":[181],"evaluate":[182],"approach":[185],"on":[186],"multiple":[187],"tracers":[189],"in-house":[191],"public":[193],"datasets,":[194],"including":[195],"18F-FDG":[196],"PET,":[197,199,202],"68Ga-DOTATATE":[198],"18F-PSMA":[201],"demonstrating":[203],"consistent":[204],"acceleration":[205],"over":[207],"order":[209],"magnitude":[211],"alongside":[212],"improved":[213],"comparable":[215],"performance":[217],"relative":[218],"full-step":[221],"baseline.":[222],"Blinded":[223],"reader":[224],"studies":[225],"confirm":[227],"enhanced":[228],"confidence":[230],"perceived":[232],"diagnostic":[233]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-06-10T00:00:00"}