# Reliable Multilingual Orthopedic Decision Support from Clinical Narratives: Language-Aware Adaptation and Verification-Guided Deferral

arXiv: https://arxiv.org/abs/2605.31512 · HTML: https://mlsanity.com/paper/2605.31512/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 20 of its 46 references that are in the map.
- First posted on arXiv 29 May 2026.

## Builds on (most cited first, 20 of 20)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter](https://mlsanity.com/paper/1910.01108/) (2019)
- [Unsupervised Cross-lingual Representation Learning at Scale](https://mlsanity.com/paper/1911.02116/) (2019)
- [Domain Specialization as the Key to Make Large Language Models Disruptive: A Comprehensive Survey](https://mlsanity.com/paper/2305.18703/) (2023)
- [Towards Building Multilingual Language Model for Medicine](https://mlsanity.com/paper/2402.13963/) (2024)
- [Benchmarking large language models for biomedical natural language processing applications and recommendations](https://mlsanity.com/paper/2305.16326/) (2023)
- [IndicGenBench: A Multilingual Benchmark to Evaluate Generation Capabilities of LLMs on Indic Languages](https://mlsanity.com/paper/2404.16816/) (2024)
- [A Unified Review of Deep Learning for Automated Medical Coding](https://mlsanity.com/paper/2201.02797/) (2022)
- [MEDDxAgent: A Unified Modular Agent Framework for Explainable Automatic Differential Diagnosis](https://mlsanity.com/paper/2502.19175/) (2025)
- [A Modular Approach for Clinical SLMs Driven by Synthetic Data with Pre-Instruction Tuning, Model Merging, and Clinical-Tasks Alignment](https://mlsanity.com/paper/2505.10717/) (2025)
- [Trustworthy and Practical AI for Healthcare: A Guided Deferral System with Large Language Models](https://mlsanity.com/paper/2406.07212/) (2024)
- [Uncertainty-aware Medical Diagnostic Phrase Identification and Grounding](https://mlsanity.com/paper/2404.06798/) (2024)
- [Pattern Recognition or Medical Knowledge? The Problem with Multiple-Choice Questions in Medicine](https://mlsanity.com/paper/2406.02394/) (2024)
- [Towards Omni-RAG: Comprehensive Retrieval-Augmented Generation for Large Language Models in Medical Applications](https://mlsanity.com/paper/2501.02460/) (2025)
- [Multi-OphthaLingua: A Multilingual Benchmark for Assessing and Debiasing LLM Ophthalmological QA in LMICs](https://mlsanity.com/paper/2412.14304/) (2024)
- [DeepSeek-Inspired Exploration of RL-based LLMs and Synergy with Wireless Networks: A Survey](https://mlsanity.com/paper/2503.09956/) (2025)
- [Aligning AI Research with the Needs of Clinical Coding Workflows: Eight Recommendations Based on US Data Analysis and Critical Review](https://mlsanity.com/paper/2412.18043/) (2024)
- [Temporal Relation Extraction in Clinical Texts: A Span-based Graph Transformer Approach](https://mlsanity.com/paper/2503.18085/) (2025)
- [Exploring the Consistency, Quality and Challenges in Manual and Automated Coding of Free-text Diagnoses from Hospital Outpatient Letters](https://mlsanity.com/paper/2311.10856/) (2023)
- [AutoMedEval: Harnessing Language Models for Automatic Medical Capability Evaluation](https://mlsanity.com/paper/2505.11887/) (2025)

## Abstract (arXiv)

Multilingual orthopedic decision support remains challenging in low-resource healthcare settings, where clinical narratives contain specialized terminology, mixed scripts, incomplete evidence, label imbalance and language-dependent documentation patterns. This article presents a reliability-oriented framework for classifying free-text orthopedic notes in English, Hindi and Punjabi. We compare task-aligned multilingual transformer encoders, a task-fine-tuned DistilBERT baseline, zero-shot instruction-tuned large language models (LLMs) and a domain-adaptive encoder, IndicBERT-HPA. IndicBERT-HPA augments IndicBERT with language-aware orthopedic adapter heads to support clinically relevant multilingual representation learning. Evaluation extends beyond aggregate accuracy to per-class performance, ROC-AUC, AUPRC, expected calibration error, cross-language stability and robustness under controlled balanced and natural-prevalence distributions. The evaluated zero-shot LLMs remain substantially less effective than task-adapted encoders for closed-set classification, with language-dependent instability. Under natural clinical prevalence, IndicBERT-HPA achieves the strongest overall performance, reaching an averaged Macro-F1 of 0.8792, Macro-AUROC of 0.894 and AUPRC of 0.902. We further implement a deterministic selective-verification layer combining confidence gating, evidence-consistency checking and language-risk screening. On a randomly selected held-out 5,000-record subset, it achieves 84.4% selective accuracy and 0.76 selective Macro-F1 at 72.3% coverage, compared with 71.5% accuracy and 0.65 Macro-F1 for accept-all prediction. These results support reliability-oriented multilingual clinical decision support with explicit deferral.