{"doi":"10.1016/j.adro.2025.101826","title":"A Radiogenomic Deep Ensemble Learning Model for Identifying Radionecrosis Following Brain Metastases (BM) Stereotactic Radiosurgery in Patients With Non-small Cell Lung Cancer BM","abstract":"Purpose Stereotactic radiosurgery (SRS) is widely used for brain metastases (BM), but the risk of radionecrosis poses a challenge in post-SRS management. Given the lack of noninvasive imaging methods for distinguishing radionecrosis from recurrence, we aimed to design a deep ensemble learning model that integrates patient clinical features and genomic profiles to identify radionecrosis in patients with BM with post-SRS radiographic progression. Methods and Materials We studied 90 BMs from 62 patients with non-small cell lung cancer, with 27 biopsy-confirmed post-SRS local recurrences. Clinical features and molecular features were collected. A deep neural network (DNN) was trained for radionecrosis/recurrence prediction using the 3-month post-SRS T1+c magnetic resonance imaging. Preceding the binary prediction output, latent variables were extracted as 1024 deep features. An ensemble learning model was then developed, comprising 2 submodels that fused deep features with clinical (\" D+C\" ) or genomic (\" D+G\" ) features. We employed our positional encoding method to optimally fuse the low-dimensional clinical/genomic features with the high-dimensional image features. The postfusion feature in each submodel yielded a logit result after traversing fully connected layers. The ensemble's final output was the synthesized result of these 2 submodels' logits via logistic regression. Model training employed an 8:2 train/test split, and 10 model versions were developed for robustness evaluation. Performance metrics were compared against image-only DNN model and \" D+C\" and \" D+G\" submodels. Results The deep ensemble model showed satisfactory performance on the test set, with the area under the receiver operating characteristic curve (ROC AUC ) = 0.91 ± 0.04, sensitivity=0.87 ± 0.16, specificity=0.86 ± 0.08, and accuracy=0.87 ± 0.04. This significantly outperformed the image-only DNN result (ROC AUC = 0.71 ± 0.05, sensitivity=0.66 ± 0.32). Higher average performance was also observed compared to the \"D+C\" result (ROC AUC = 0.82 ± 0.03, sensitivity=0.67 ± 0.17) and \"D+G\" result (ROC AUC = 0.83 ± 0.02, sensitivity=0.76 ± 0.22). Conclusions The deep ensemble model achieved the best performance among the models evaluated in this study for distinguishing BM radionecrosis from recurrence using 3-month post-SRS T1+c MR images, clinical features, and genomic features. This highlights the potential of artificial intelligence in clinical decision-making for BM management, warranting further investigation into its clinical applications.","journal":"Advances in Radiation Oncology","year":2025,"id":569596,"datarank":0.0,"base_score":0.0,"endowment":0.0,"self_citation_contribution":0.0,"citation_network_contribution":0.0,"self_endowment_contribution":0.0,"citer_contribution":0.0,"corpus_percentile":null,"corpus_rank":null,"citation_count":0,"citer_count":0,"citers_with_citation_signal":0,"citers_with_endowment":0,"datacite_reuse_total":0,"is_dataset":false,"is_dataset_confidence":0.954,"is_data_producer":false,"deposit_databanks":null,"is_oa":true,"file_count":0,"downloads":0,"has_version_chain":false,"published_date":"2025-01-01","fair_score":null,"fair_percentile":null,"algorithm_id":"datarank_citation_only_1hop_v6","ranking_scope":"data_only","authors":[{"id":1057411,"name":"Eugene Vaios","orcid":"0000-0003-2122-6650","position":1,"is_corresponding":false},{"id":297411,"name":"Evan Calabrese","orcid":"0000-0002-1464-0354","position":2,"is_corresponding":false},{"id":1057408,"name":"Zhenyu Yang","orcid":"0000-0002-7137-2552","position":3,"is_corresponding":false},{"id":1475120,"name":"Scott W. Robertson","orcid":null,"position":4,"is_corresponding":false},{"id":1474780,"name":"John Ginn","orcid":"0000-0002-5694-9392","position":5,"is_corresponding":false},{"id":756026,"name":"Ke Lü","orcid":"0000-0001-8513-3352","position":6,"is_corresponding":false},{"id":345264,"name":"F Yin","orcid":"0000-0002-2025-4740","position":7,"is_corresponding":false},{"id":1386441,"name":"Zachary Reitman","orcid":null,"position":8,"is_corresponding":false},{"id":259415,"name":"John P. Kirkpatrick","orcid":"0000-0002-4019-0350","position":9,"is_corresponding":false},{"id":536197,"name":"Scott Floyd","orcid":"0000-0002-8067-2426","position":10,"is_corresponding":false},{"id":564965,"name":"Peter E. Fecci","orcid":"0000-0002-2912-8695","position":11,"is_corresponding":false},{"id":345257,"name":"Chunhao Wang","orcid":"0000-0002-6945-7119","position":12,"is_corresponding":false},{"id":1419865,"name":"Jingtong Zhao","orcid":"0009-0001-6674-3144","position":0,"is_corresponding":true}],"reference_count":55,"raw_metadata":null,"created_at":"2026-07-19T02:57:03.510013Z","pmid":"40686742","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":[]}