{"doi":"10.3934/ammc.2024015","title":"Empirical evidence of the task-adapted reconstruction framework for joint CT reconstruction and segmentation","abstract":"Powered by machine learning, computer-aided diagnostics support clinicians by streamlining their work. In cancer screening, for instance, this technique often involves an automated detection of potential cancerous lesions from X-ray Computed Tomography (CT) images reconstructed a priori and unaware of the detection task. Since different reconstruction algorithms enhance different image attributes, a natural question is whether adapting the algorithms to the downstream task is relevant to improving the performance of computer-aided diagnosis tools.This paper provides empirical evidence about the performance of the task-adapted framework for joint reconstruction-segmentation of lung nodules from CT scans. We describe the procedures that allow the joint training of reconstruction and segmentation operators. We also report that relevant segmentation metrics improve by up to twenty percentage points in the four angular-sparsity and X-ray dose settings studied. Finally, we give practical recommendations for the procedure's hyperparameter tuning.Our results suggest that the imaging pipeline could benefit from computing different images adapted to the clinician's eyes and the automatic segmentation tools, outlining future improvements in patient care and clinical tasks.","journal":"Applied Mathematics for Modern Challenges","year":2024,"id":477457,"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":2,"citer_count":0,"citers_with_citation_signal":0,"citers_with_endowment":0,"datacite_reuse_total":0,"is_dataset":false,"is_dataset_confidence":0.9554,"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":1315742,"name":"Ander Biguri","orcid":"0000-0002-2636-3032","position":1,"is_corresponding":false},{"id":1316118,"name":"Lorena Escudero Sánchez","orcid":null,"position":2,"is_corresponding":false},{"id":1062553,"name":"Cathal McCague","orcid":"0000-0002-9588-267X","position":3,"is_corresponding":false},{"id":1062557,"name":"Ozan Öktem","orcid":"0000-0002-1118-6483","position":4,"is_corresponding":false},{"id":638656,"name":"Carola‐Bibiane Schönlieb","orcid":"0000-0003-0099-6306","position":5,"is_corresponding":false},{"id":1315741,"name":"Emilien Valat","orcid":"0000-0003-0055-9659","position":0,"is_corresponding":true}],"reference_count":15,"raw_metadata":null,"created_at":"2026-07-19T02:06:37.812633Z","pmid":null,"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":[]}