{"doi":"10.1162/imag_a_00251","title":"Translating phenotypic prediction models from big to small anatomical MRI data using meta-matching","abstract":"Individualized phenotypic prediction based on structural magnetic resonance imaging (MRI) is an important goal in neuroscience. Prediction performance increases with larger samples, but small-scale datasets with fewer than 200 participants are often unavoidable. We have previously proposed a \"meta-matching\" framework to translate models trained from large datasets to improve the prediction of new unseen phenotypes in small collection efforts. Meta-matching exploits correlations between phenotypes, yielding large improvement over classical machine learning when applied to prediction models using resting-state functional connectivity as input features. Here, we adapt the two best performing meta-matching variants (\"meta-matching finetune\" and \"meta-matching stacking\") from our previous study to work with T1-weighted MRI data by changing the base neural network architecture to a 3D convolution neural network. We compare the two meta-matching variants with elastic net and classical transfer learning using the UK Biobank (N = 36,461), the Human Connectome Project Young Adults (HCP-YA) dataset (N = 1,017), and the HCP-Aging dataset (N = 656). We find that meta-matching outperforms elastic net and classical transfer learning by a large margin, both when translating models within the same dataset and when translating models across datasets with different MRI scanners, acquisition protocols, and demographics. For example, when translating a UK Biobank model to 100 HCP-YA participants, meta-matching finetune yielded a 136% improvement in variance explained over transfer learning, with an average absolute gain of 2.6% (minimum = -0.9%, maximum = 17.6%) across 35 phenotypes. Overall, our results highlight the versatility of the meta-matching framework.","journal":"Imaging Neuroscience","year":2024,"id":464267,"datarank":0.2952111645441708,"base_score":1.791759469228055,"endowment":1.791759469228055,"self_citation_contribution":0.26876392038420827,"citation_network_contribution":0.026447244159962524,"self_endowment_contribution":0.26876392038420827,"citer_contribution":0.026447244159962524,"corpus_percentile":null,"corpus_rank":null,"citation_count":5,"citer_count":3,"citers_with_citation_signal":2,"citers_with_endowment":2,"datacite_reuse_total":0,"is_dataset":false,"is_dataset_confidence":0.95,"is_data_producer":false,"deposit_databanks":null,"is_oa":true,"file_count":0,"downloads":0,"has_version_chain":false,"published_date":"2024-01-01","fair_score":null,"fair_percentile":null,"algorithm_id":"datarank_citation_only_1hop_v6","ranking_scope":"data_only","authors":[{"id":244798,"name":"Lijun An","orcid":"0000-0003-1030-4625","position":1,"is_corresponding":false},{"id":1174052,"name":"Chen Zhang","orcid":"0000-0002-1579-0047","position":2,"is_corresponding":false},{"id":258423,"name":"Ru Kong","orcid":"0000-0001-7842-0329","position":3,"is_corresponding":false},{"id":873718,"name":"Pansheng Chen","orcid":"0009-0009-8881-8643","position":4,"is_corresponding":false},{"id":18614,"name":"Danilo Bzdok","orcid":"0000-0003-3466-6620","position":5,"is_corresponding":false},{"id":106615,"name":"Simon B. Eickhoff","orcid":"0000-0001-6363-2759","position":6,"is_corresponding":false},{"id":16285,"name":"Avram J. Holmes","orcid":"0000-0001-6583-803X","position":7,"is_corresponding":false},{"id":30836,"name":"B.T. Thomas Yeo","orcid":"0000-0002-0119-3276","position":8,"is_corresponding":false},{"id":1030714,"name":"Naren Wulan","orcid":"0009-0005-3974-7528","position":0,"is_corresponding":true}],"reference_count":89,"raw_metadata":{"citation_network_status":"fetched"},"created_at":"2026-07-19T02:04:37.468651Z","pmid":"40800257","pmcid":null,"fwci":null,"citation_percentile":null,"influential_citations":0,"oa_status":null,"license":null,"views":0,"total_file_size_bytes":0,"version_count":0,"fair_f":null,"fair_a":null,"fair_i":null,"fair_r":null,"fair_zscore":null,"fair_rationale":null,"fair_model":null,"fair_agent_version":null,"fair_fulltext_source":null,"fair_has_llm":null,"fair_computed_at":null,"clinical_trials":[],"software_tools":[],"db_accessions":[],"linked_datasets":[],"topics":[]}