Compact Multimodal Language Models as Robust OCR Alternatives for Noisy Textual Clinical Reports

2025/11/17 by Nikita Neveditsin, Neveditsin, Nikita, Pawan Lingras +6
Computer Science · #FOS: Computer and information sciences #Information Retrieval (cs.IR) #Machine Learning in Healthcare #Multimodal Machine Learning Applications #Topic Modeling

paper · pdf · doi:10.48550/arxiv.2511.13523

openalex publication_date 2025/11/17 · openalex created_date 2025/11/19 · openalex updated_date 2026/07/28

Abstract

Digitization of medical records often relies on smartphone photographs of printed reports, producing images degraded by blur, shadows, and other noise. Conventional OCR systems, optimized for clean scans, perform poorly under such real-world conditions. This study evaluates compact multimodal language models as privacy-preserving alternatives for transcribing noisy clinical documents. Using obstetric ultrasound reports written in regionally inflected medical English common to Indian healthcare settings, we compare eight systems in terms of transcription accuracy, noise sensitivity, numeric accuracy, and computational efficiency. Compact multimodal models consistently outperform both classical and neural OCR pipelines. Despite higher computational costs, their robustness and linguistic adaptability position them as viable candidates for on-premises healthcare digitization.

Related

• Preserving Privacy, Increasing Accessibility, and Reducing Cost: An On-Device Artificial Intelligence Model for Medical Transcription and Note Generation
• A Multimodal Pipeline for Clinical Data Extraction: Applying Vision-Language Models to Scans of Transfusion Reaction Reports
• Evaluating the Effectiveness of Cost-Efficient Large Language Models in Benchmark Biomedical Tasks
• Searching for Best Practices in Medical Transcription with Large Language Model
• Towards Selection of Large Multimodal Models as Engines for Burned-in Protected Health Information Detection in Medical Images