Removing Redundant Anatomical Inputs Improves Deep Learning-Based Forensic Dental Age and Sex Estimation
- 1Guangzhou Key Laboratory of Forensic Multi-Omics for Precision Identification, School of Forensic Medicine, Southern Medical University
Abstract
Age and sex estimation from dental radiographs supports forensic identification and age-dependent legal assessment. Multi-input models may combine global and regional representations, but when these inputs are derived from the same radiograph, they may repeat anatomical information rather than provide complementary signals. We developed HMA-Net, a hierarchical multi-stream anatomical network, together with a task-specific input audit that evaluates whether each candidate representation improves at least one prediction task without materially degrading the other. The model was developed using a retrospective single-centre cohort of 4,293 panoramic radiographs from individuals aged 3-18 years and combined a panoramic-image stream with graph-linked tooth representations. On the held-out test set, the final panoramic-plus-permanent-tooth model achieved a mean absolute age error of 0.51 years and a sex-classification accuracy of 85.2%; 99.4% of age estimates were within 2 years of chronological age. Across three independent training runs, removing the jaw-region stream reduced mean absolute age error from 0.68 {+/-} 0.01 to 0.52 {+/-} 0.01 years and increased sex accuracy from 81.3 {+/-} 1.7% to 85.8 {+/-} 0.8%. Adding 20 deciduous-tooth classes provided no age benefit and reduced sex accuracy by 7.9% relative to the permanent-only representation. The mandibular first molars received the greatest tooth-level attention, and their occlusion impaired both prediction tasks. These findings show that anatomically plausible inputs do not necessarily provide incremental predictive value. Task-specific input auditing can improve joint model performance, reduce input redundancy and preserve anatomically traceable predictions. More broadly, this framework provides a testable approach for selecting correlated inputs in medical-imaging systems in which regional views re-present anatomy already contained in the whole image.
Related
- Preprocedural CT-guided anatomical assessment shortens intubation time and improves patient and operator experience during unsedated colonoscopy: a randomised controlled trial
- Prediction of discoid lateral meniscus based on tibial plateau anatomical factors and machine learning
- Deep Learning-Based Anatomical Level Classification of Craniofacial CT Slices: Comparison of CNN Architectures and Ensemble Methods
- BurnAreaNet: a deep learning method for total body surface area estimation from 2D masks
- Development and External Validation of Machine Learning Models for Ionized Calcium Estimation in Hypoalbuminemia
BibTeX
@misc{wu2026removing,
title = {{Removing Redundant Anatomical Inputs Improves Deep Learning-Based Forensic Dental Age and Sex Estimation}},
author = {Z. Wu and W. Shan and W. Zhang and Z. Kang and B. Yang and Z. Liao and Z. Huang and Y. Zhang and X. Yue},
howpublished = {medRxiv},
year = {2026},
month = oct,
doi = {10.64898/2026.10.05.26364771},
url = {https://www.medrxiv.org/content/10.64898/2026.10.05.26364771v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - Removing Redundant Anatomical Inputs Improves Deep Learning-Based Forensic Dental Age and Sex Estimation
AU - Z. Wu
AU - W. Shan
AU - W. Zhang
AU - Z. Kang
AU - B. Yang
AU - Z. Liao
AU - Z. Huang
AU - Y. Zhang
AU - X. Yue
PY - 2026
DA - 2026/10/05
PB - medRxiv
DO - 10.64898/2026.10.05.26364771
UR - https://www.medrxiv.org/content/10.64898/2026.10.05.26364771v1
AB - Age and sex estimation from dental radiographs supports forensic identification and age-dependent legal assessment. Multi-input models may combine global and regional representations, but when these inputs are derived from the same radiograph, they may repeat anatomical information rather than provide complementary signals. We developed HMA-Net, a hierarchical multi-stream anatomical network, together with a task-specific input audit that evaluates whether each candidate representation improves at least one prediction task without materially degrading the other. The model was developed using a retrospective single-centre cohort of 4,293 panoramic radiographs from individuals aged 3-18 years and combined a panoramic-image stream with graph-linked tooth representations. On the held-out test set, the final panoramic-plus-permanent-tooth model achieved a mean absolute age error of 0.51 years and a sex-classification accuracy of 85.2%; 99.4% of age estimates were within 2 years of chronological age. Across three independent training runs, removing the jaw-region stream reduced mean absolute age error from 0.68 {+/-} 0.01 to 0.52 {+/-} 0.01 years and increased sex accuracy from 81.3 {+/-} 1.7% to 85.8 {+/-} 0.8%. Adding 20 deciduous-tooth classes provided no age benefit and reduced sex accuracy by 7.9% relative to the permanent-only representation. The mandibular first molars received the greatest tooth-level attention, and their occlusion impaired both prediction tasks. These findings show that anatomically plausible inputs do not necessarily provide incremental predictive value. Task-specific input auditing can improve joint model performance, reduce input redundancy and preserve anatomically traceable predictions. More broadly, this framework provides a testable approach for selecting correlated inputs in medical-imaging systems in which regional views re-present anatomy already contained in the whole image.
ER -