{"doi":"10.1016/j.ejrai.2025.100005","title":"Deep learning-based detection and segmentation of osseous metastatic prostate cancer lesions on computed tomography","abstract":"Introduction: Prostate adenocarcinoma frequently metastasizes to bone and is detected via computed tomography (CT) scans. Accurate detection and segmentation of these lesions are critical for diagnosis, prognosis, and monitoring. This study aims to automate lesion detection and segmentation using deep learning models. Methods and Materials: We evaluated deep learning models for lesion detection (EfficientNet, ResNet34, DenseNet) and segmentation (nnUNetv2, UNet, ResUNet, ResAttUNet). Performance metrics included F1 score, precision, recall, Area Under the Curve (AUC), and Dice Similarity Coefficient (DSC). Pairwise t-tests compared segmentation accuracy. Radiomic analyses compared lesions segmented by deep learning to manual segmentations. Results: EfficientNet achieved the highest detection performance, with an F1 score of 0.82, precision of 0.88, recall of 0.79, and AUC of 0.71. Among segmentation models, nnUNetv2 performed best, achieving a DSC of 0.74, precision of 0.73, and recall of 0.83. Pairwise t-tests showed that nnUNetv2 outperformed other models in segmentation accuracy (p < 0.01). Clinically, nnUNetv2 also demonstrated superior specificity for lesion detection (0.90) compared to the other models. All models performed similarly in distinguishing diffuse and focal lesions, predicting weight-bearing lesions, and identifying lesion locations, although nnUNetv2 had higher specificity. Sensitivity was highest for rib lesions and lowest for spine lesions across all models. Conclusions: EfficientNet and nnUNetv2 were the top-performing models for detection and segmentation, respectively. Radiomic features derived from deep learning-based segmentations were comparable to manual segmentations, supporting clinical applicability. Further analysis of lesion detection and spatial distribution underscores the models' potential for improving diagnostic workflows and patient outcomes.","journal":"European Journal of Radiology Artificial Intelligence","year":2025,"id":530995,"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":4,"citer_count":0,"citers_with_citation_signal":0,"citers_with_endowment":0,"datacite_reuse_total":0,"is_dataset":false,"is_dataset_confidence":0.9594,"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":1281437,"name":"Joseph M. Rich","orcid":"0000-0003-1400-8479","position":1,"is_corresponding":false},{"id":1411299,"name":"Shreyas Malewar","orcid":null,"position":2,"is_corresponding":false},{"id":1410917,"name":"Daksh Patel","orcid":"0000-0002-8693-1556","position":3,"is_corresponding":false},{"id":1411300,"name":"Matt Muellner","orcid":null,"position":4,"is_corresponding":false},{"id":271864,"name":"Darryl Hwang","orcid":"0000-0002-1111-9176","position":5,"is_corresponding":false},{"id":522005,"name":"Xiaomeng Lei","orcid":"0000-0001-8899-5177","position":6,"is_corresponding":false},{"id":522004,"name":"Steven Cen","orcid":"0000-0002-7859-8909","position":7,"is_corresponding":false},{"id":564636,"name":"Timothy J. Triche","orcid":"0000-0003-4869-8826","position":8,"is_corresponding":false},{"id":441991,"name":"Amir Goldkorn","orcid":"0000-0003-4664-9166","position":9,"is_corresponding":false},{"id":1411301,"name":"Passant Mohamed","orcid":null,"position":10,"is_corresponding":false},{"id":365071,"name":"Assad A. Oberai","orcid":"0000-0002-7668-7713","position":11,"is_corresponding":false},{"id":426633,"name":"Vinay Duddalwar","orcid":"0000-0002-4808-5715","position":12,"is_corresponding":false},{"id":1411298,"name":"S J Pawan","orcid":null,"position":0,"is_corresponding":true}],"reference_count":21,"raw_metadata":null,"created_at":"2026-07-19T02:51:10.077559Z","pmid":"41822381","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":[]}