{"paper":{"title":"From Full and Partial Intraoral Scans to Crown Proposal: A Classification-Guided Restoration Assistance Pipeline","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"A classification-guided pipeline produces patient-specific crown proposals from partial intraoral scans in 2.5-3.5 minutes.","cross_cats":["cs.CV","cs.LG"],"primary_cat":"eess.IV","authors_text":"Akio Tanaka, Amit Regmi, Dikshya Parajuli, Kennta Kashiwazaki, Kundan Siwakoti, Louis Digiorgio, Manabu Kanazawa, Masahiko Inada, Prince Panta, Rabin Kunwar, Romik Gosai, Rujal Acharya, Saugat Kafley, Shuvangi Adhikari, Yuriko Komagamine","submitted_at":"2026-05-14T06:05:32Z","abstract_excerpt":"Single-unit crown restoration is among the most common procedures in clinical dentistry, with CAD/CAM workflows now designing crowns directly from intraoral scans. Partial scans are often preferred over full-arch scans for single-unit cases due to fewer stitching errors, yet most segmentation networks trained on full arches fail on partial scans, while end-to-end generative crown methods often produce over-smoothed surfaces that lose occlusal detail. We propose an end-to-end pipeline that takes a raw intraoral scan and target FDI tooth number as input and outputs an initial, patient-specific c"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"The pipeline produces a preliminary crown shell in 2.5-3.5 minutes, offering a practical alternative to end-to-end generative approaches, with macro-average DSC 0.9249, Recall 0.8919, and Precision 0.9615 across 17 classes on 1,958 partial scans and sub-millimeter centroid errors (0.2666-0.2774 mm) for the prepared tooth and neighbors.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The retrieval step assumes that cosine similarity over DGCNN embeddings of neighboring and opposing teeth will surface geometrically suitable crown candidates for the specific patient anatomy; this premise is invoked in the context-aware retrieval and Blender fitting phase and is required for the output to be clinically usable rather than merely plausible.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"A classification-routed pipeline segments partial and full intraoral scans then retrieves and fits crown proposals from neighboring teeth embeddings, reporting macro DSC 0.9249 on 1958 partial scans.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"A classification-guided pipeline produces patient-specific crown proposals from partial intraoral scans in 2.5-3.5 minutes.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"4f2e67d5cfe9a9dc2ebc2473be969708f2f5352dc170aa04fc5e8053ab248edc"},"source":{"id":"2605.15241","kind":"arxiv","version":1},"verdict":{"id":"c9fd99cc-4ba5-42a4-a8ad-ac4208e1d276","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-19T16:10:27.460117Z","strongest_claim":"The pipeline produces a preliminary crown shell in 2.5-3.5 minutes, offering a practical alternative to end-to-end generative approaches, with macro-average DSC 0.9249, Recall 0.8919, and Precision 0.9615 across 17 classes on 1,958 partial scans and sub-millimeter centroid errors (0.2666-0.2774 mm) for the prepared tooth and neighbors.","one_line_summary":"A classification-routed pipeline segments partial and full intraoral scans then retrieves and fits crown proposals from neighboring teeth embeddings, reporting macro DSC 0.9249 on 1958 partial scans.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The retrieval step assumes that cosine similarity over DGCNN embeddings of neighboring and opposing teeth will surface geometrically suitable crown candidates for the specific patient anatomy; this premise is invoked in the context-aware retrieval and Blender fitting phase and is required for the output to be clinically usable rather than merely plausible.","pith_extraction_headline":"A classification-guided pipeline produces patient-specific crown proposals from partial intraoral scans in 2.5-3.5 minutes."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.15241/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"doi_title_agreement","ran_at":"2026-05-19T16:31:18.463551Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T16:27:11.584607Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"claim_evidence","ran_at":"2026-05-19T16:01:54.837032Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"ai_meta_artifact","ran_at":"2026-05-19T13:33:22.822746Z","status":"skipped","version":"1.0.0","findings_count":0}],"snapshot_sha256":"f2cb5b0ae3a502fa6f58f5a558ace3f9e9e789cb9959b1ea795826e5fc52395f"},"references":{"count":32,"sample":[{"doi":"","year":2025,"title":"The transformative role of artificial intelligence in dentistry: A comprehensive overview. Part 1: Fundamentals of AI, and its contemporary applications in dentistry,","work_id":"169b2f87-1e40-4b92-b402-2ebb583e696d","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"10.1007/s10278-024-01061-6","year":2022,"title":"Dilated dynamic graph CNN for efficient point cloud learning,","work_id":"b5210199-1106-4e3c-b7c9-95ccb139103b","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"A critical analysis of the limitation of deep learning based 3D dental mesh segmentation methods in segmenting partial scans,","work_id":"98ae591b-e18b-4d9e-a749-43ef41f1bd61","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2016,"title":"Effect of intraoral scanner and fixed partial denture situation on the scan accuracy of multiple implants: an in vitro study,","work_id":"b318c504-7eaa-4e3f-ad00-86ebc8016301","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2016,"title":"Effect of scanned area and operator on the accuracy of dentate arch scans with a single implant,","work_id":"2cbab15d-5be5-4965-a50b-9d33411d28e3","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":32,"snapshot_sha256":"98b03809e23040b935b87a7d0b6aa04f8d2d419c4a7150c76bcb6e9486f7f160","internal_anchors":2},"formal_canon":{"evidence_count":2,"snapshot_sha256":"a9d3ab1aa0725ea240565a67909c3378002d409ddb54f35f093d4bb5b42ed7ce"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}