Entity-Constrained CBCT Retrieval for Low-Resource Dental Record Completion

Nhi Ngoc-Yen Nguyen; Thai Nguyen; Kiet Huynh Cao Tuan; Huy-Hieu Pham

cs.CVAug 22, 2026

Entity-Constrained CBCT Retrieval for Low-Resource Dental Record Completion

Authors: Nhi Ngoc-Yen Nguyen, Thai Nguyen, Kiet Huynh Cao Tuan, Huy-Hieu Pham

Organizations: VinUni-Illinois Smart Health Center, VinUniversity, Hanoi, Vietnam · Vietnam National University Ho Chi Minh City, University of Science, Vietnam

Abstract

Completing dental records from cone-beam computed tomography (CBCT) is difficult when annotation is scarce and individual clinical fields are supported by different types of evidence. MMDental Task 3 requires seven-field record completion from only 50 labeled CBCT cases and scores the correctness of structured FDI positions and ICD codes; consequently, a visually plausible retrieved record can still be harmful when it introduces an unsupported entity. We propose Entity-Constrained CBCT-Guided Retrieval (ECCR), a parameter-free framework that separates evidence availability from evidence authority. A corpus-derived prior first supplies the complete record. A frozen 3D encoder retrieves image-conditioned Diagnosis evidence, which is appended only if it does not expand the prior FDI or ICD entity set, so the asserted entity set is invariant by construction. On public validation, ECCR reaches a weighted score of 0.3134, improving on both full-record multimodal retrieval (0.2237) and a static text-only prior (0.2915); the guard blocks 63.3% of retrieved candidates, each of which would otherwise have injected an FDI position or ICD code absent from the prior. On the final test evaluation, ECCR obtains 11.37 of a 97.4-point attainable maximum, securing second place overall. The result indicates that, in an extreme low-resource setting, controlling what multimodal evidence is allowed to modify can be more reliable than transferring an entire retrieved record.

Figures & tables

Field	Filled (%)	Unique	Observed structure
Handle	96	145	Near visit-specific
Oral check	93	146	Near visit-specific
Doctor advices	80	59	Strongly templated
Diagnosis	76	99	Diverse, recurrent groups
Treatment plan	59	88	Diverse, incomplete
Main appeal	34	50	Sparse

Table 1: Field heterogeneity motivates entity-constrained retrieval update. Statistics cover 161 labeled visits; Filled (%) denotes non-empty rows and Unique counts distinct non-empty strings. Oral check corresponds to the Oral exam aspect of the official metric.

Figure 1: Retrieved evidence is controlled rather than copied. A frozen 3D encoder retrieves a neighbour, while prior $p$ supplies all seven fields. Retrieval may propose a Diagnosis update only ( $A_{f}=0$ otherwise), and admission requires $\mathcal{E}(r)\subseteq\mathcal{E}(p)$ ; rejection retains the prior. With $k=1$ , entity consistency is the operative gate.

Component	Configuration
GPU	1 $\times$ NVIDIA GeForce RTX 4090 (24 GB)
Encoder	Med3DVLM VisionTower [ 8 ] (frozen)
Optimization	No training; parameter-free inference

Table 2: System configuration. The proposed pipeline is parameter-free and requires only frozen encoder inference over the cached bank and query volumes.

System	Retrieval update	Score
M3D-CLIP retrieval	Full: replaces the complete record	0.2237
BLEU-medoid	None: text prior only	0.2915
Ours	Diagnosis only; entity-consistent	0.3134

Table 3: Public-validation comparison of evidence authority (0–1 scale). Full authority replaces the complete record, no authority emits the text prior, and ECCR permits only an entity-compatible Diagnosis update.

Table 5

Rank	Team	Diagnosis	Oral exam	Treatment	Completeness	Total
1	abc-abc	5.56	1.49	4.71	5.95	17.70
2	Ours	4.89	0.00	2.72	3.76	11.37
3	sjtu426lab-task3	5.29	0.00	3.12	1.68	10.08
4	jlshen	4.50	0.00	2.07	3.50	10.06
Aspect weight		0.45	0.20	0.20	0.15	1.00

Table 6: Final Task 3 leaderboard under the 100-point protocol (attainable maximum

97.4

). ECCR ranks second.

Aspect (Points / Credited)	Sub-metric	$n$	F1	Points
Diagnosis 8.73 / 4.89	Condition detection	47	0.2345	4.69
	Tooth–condition match	32	0.0159	0.16
	ICD codes	40	0.1630	1.63
	Whole-mouth	20	0.4500	2.25
Oral exam 0.00 / 0.00	Findings detection	34	0.0000	0.00
	Tooth-level findings	21	0.0000	0.00

Table 7: Sub-metric decomposition of ECCR’s final score. Points are on the 100-point scale before the diversity factor; Credited applies

\phi=0.56

. Completeness sub-metrics combine a completion rate and ROUGE-L and are reported at aspect level only.

Figure 2: Similarity alone does not determine safe admission. Validation queries are ordered by top-1 cosine similarity. Blue candidates satisfy the entity condition; orange candidates contain an FDI or ICD conflict. Blocked candidates occur across the similarity range, showing why an entity-aware rule cannot be replaced by a single similarity threshold.

Requirements	Answer
A meaningful title	Yes
The number of authors ( $\leq$ 6)	4
Author affiliations and ORCID	Yes
Corresponding author email is presented	Yes
Validation scores are presented in the abstract	Yes
Introduction includes at least three parts: background, related work, and motivation	Yes

Table 8: Checklist Table. Please fill out this checklist table in the answer column. (Delete this Table in the camera-ready submission)

Explore similar work

EviDent-CBCT: Evidence-Bottlenecked Report Generation from Dental CBCT under Non-Exhaustive Report Supervision
Oct 1, 2026Ruiyang Hao, Zhi Qin Tan, Yulan He +2Radiology Report Generation Medical Report Generation
King’s College London
Clinical Reasoning Under a Partially Observed Objective in Cone Beam CT Report Generation
Sep 2, 2026Ajo Babu George, Govind Arun, Sidharth N Krishna +1Radiology Report Generation Radiology Report Evaluation
DiceMed, Cuttack, Odisha, India · University of Maryland, College Park, Maryland, USA · Indira Gandhi National Open University, New Delhi, India +1
Field-of-View Extension in Dental Cone-Beam CT via Implicit Neural Representations and Diffusion Model-Based Refinement
Sep 23, 2026Susanne Schaub, Florentin Bieder, Matheus L. Oliveira +5Cone-Beam CT Image Reconstruction
Department of Biomedical Engineering, University of Basel, Allschwil, Switzerland · Piracicaba Dental School, University of Campinas, Piracicaba, Brazil · State Key Laboratory of Oral & Maxillofacial Reconstruction and Regeneration, Key Laboratory of Oral Biomedicine Ministry of Education, Hubei Key Laboratory of Stomatology, School & Hospital of Stomatology, Wuhan University, Wuhan, China +2

Oct 1, 2026, cs.CVJ/K move · Enter open · S save

EviDent-CBCT: Evidence-Bottlenecked Report Generation from Dental CBCT under Non-Exhaustive Report Supervision

Ruiyang Hao, Zhi Qin Tan, Yulan He, Owen Addison+1

King’s College London

Radiology Report Generation Medical Report Generation Evidence-Grounded Generation Cone-Beam CT Weakly Supervised Learning Concept Bottleneck Models

Open paper AskPDF arXiv

Dento-maxillofacial cone-beam CT (CBCT) reports may contain dozens of tooth-specific, anatomical, and spatial findings from a single 3D scan. Learning to generate such reports from limited clinical data is challenging because routine reports may not exhaustively document image findings, and a non-mention may reflect either absence or non-reporting. We present EviDent-CBCT, an evidence-bottlenecked framework designed for this incomplete supervision. An anatomy-aware network maps each CBCT scan to a discrete record of tooth-level, global, and tooth-IAC evidence. A dental-logic consistency projection reconciles incompatible evidence before a deterministic renderer and an image-blind local language model generate the report using only this record. For tooth-level evidence, reliability-aware training uses eligible non-mentions as reduced-weight negatives, while unreported global and tooth-IAC labels remain unknown. A metal-sensitive input channel preserves intensity cues from dental materials. Across three validation runs, EviDent-CBCT achieves

0.666\pm0.006

merged evidence set-F1 and

0.402\pm0.003

RadFact-Lite-Dental logical-F1, versus

0.371\pm0.018

for the strongest controlled direct baseline. In the ODIN 2026 challenge, it ranked second in automated evaluation and third in blinded clinical Arena comparison on the hidden test set. These results support the discrete evidence record as an effective and auditable interface for CBCT report generation.

Abstract

Explore similar work

EviDent-CBCT: Evidence-Bottlenecked Report Generation from Dental CBCT under Non-Exhaustive Report Supervision

Clinical Reasoning Under a Partially Observed Objective in Cone Beam CT Report Generation

Field-of-View Extension in Dental Cone-Beam CT via Implicit Neural Representations and Diffusion Model-Based Refinement