Guardado en:
| Autores principales: | , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | https://arxiv.org/abs/2508.13378 |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866911271061815296 |
|---|---|
| author | Wang, Yiting Wang, Ziwei Zhu, Di Zhong, Jiachen Li, Weiyi |
| author_facet | Wang, Yiting Wang, Ziwei Zhu, Di Zhong, Jiachen Li, Weiyi |
| contents | Small Language Models (SLMs) are a practical option for narrow, workflow-relevant medical imaging utilities where privacy, latency, and cost dominate. We present a governance-ready recipe that combines prompt scaffolds, calibrated abstention, and standards-compliant integration into Picture Archiving and Communication Systems (PACS). Our focus is the assistive task of AP/PA view tagging for chest radiographs. Using four deployable SLMs (Qwen2.5-VL, MiniCPM-V, Gemma 7B, LLaVA 7B) on NIH Chest X-ray, we provide illustrative evidence: reflection-oriented prompts benefit lighter models, whereas stronger baselines are less sensitive. Beyond accuracy, we operationalize abstention, expected calibration error, and oversight burden, and we map outputs to DICOM tags, HL7 v2 messages, and FHIR ImagingStudy. The contribution is a prompt-first deployment framework, an operations playbook for calibration, logging, and change management, and a clear pathway from pilot utilities to reader studies without over-claiming clinical validation. We additionally specify a human-factors RACI, stratified calibration for dataset shift, and an auditable evidence pack to support local governance reviews. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2508_13378 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Governance-Ready Small Language Models for Medical Imaging: Prompting, Abstention, and PACS Integration Wang, Yiting Wang, Ziwei Zhu, Di Zhong, Jiachen Li, Weiyi Computer Vision and Pattern Recognition Small Language Models (SLMs) are a practical option for narrow, workflow-relevant medical imaging utilities where privacy, latency, and cost dominate. We present a governance-ready recipe that combines prompt scaffolds, calibrated abstention, and standards-compliant integration into Picture Archiving and Communication Systems (PACS). Our focus is the assistive task of AP/PA view tagging for chest radiographs. Using four deployable SLMs (Qwen2.5-VL, MiniCPM-V, Gemma 7B, LLaVA 7B) on NIH Chest X-ray, we provide illustrative evidence: reflection-oriented prompts benefit lighter models, whereas stronger baselines are less sensitive. Beyond accuracy, we operationalize abstention, expected calibration error, and oversight burden, and we map outputs to DICOM tags, HL7 v2 messages, and FHIR ImagingStudy. The contribution is a prompt-first deployment framework, an operations playbook for calibration, logging, and change management, and a clear pathway from pilot utilities to reader studies without over-claiming clinical validation. We additionally specify a human-factors RACI, stratified calibration for dataset shift, and an auditable evidence pack to support local governance reviews. |
| title | Governance-Ready Small Language Models for Medical Imaging: Prompting, Abstention, and PACS Integration |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2508.13378 |