Enregistré dans:
| Auteurs principaux: | , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | https://arxiv.org/abs/2508.16839 |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866909905009508352 |
|---|---|
| author | Vassef, Shayan Shimegekar, Soorya Ram Goyal, Abhay Saha, Koustuv Zonooz, Pi Kumar, Navin |
| author_facet | Vassef, Shayan Shimegekar, Soorya Ram Goyal, Abhay Saha, Koustuv Zonooz, Pi Kumar, Navin |
| contents | Clinical ML workflows are often fragmented and inefficient: triage, task selection, and model deployment are handled by a patchwork of task-specific networks. These pipelines are rarely aligned with data-science practice, reducing efficiency and increasing operational cost. They also lack data-driven model identification (from imaging/tabular inputs) and standardized delivery of model outputs. We present a framework that employs a single vision-language model (VLM) in two complementary, modular roles.
First (Solution 1): the VLM acts as an aware model-card matcher that routes an incoming image to the appropriate specialist model via a three-stage workflow (modality -> primary abnormality -> model-card ID). Reliability is improved by (i) stage-wise prompts enabling early termination via "None"/"Other" and (ii) a calibrated top-2 answer selector with a stage-wise cutoff. This raises routing accuracy by +9 and +11 percentage points on the training and held-out splits, respectively, compared with a baseline router, and improves held-out calibration (lower Expected Calibration Error, ECE).
Second (Solution 2): we fine-tune the same VLM on specialty-specific datasets so that one model per specialty covers multiple downstream tasks, simplifying deployment while maintaining performance. Across gastroenterology, hematology, ophthalmology, pathology, and radiology, this single-model deployment matches or approaches specialized baselines.
Together, these solutions reduce data-science effort through more accurate selection, simplify monitoring and maintenance by consolidating task-specific models, and increase transparency via per-stage justifications and calibrated thresholds. Each solution stands alone, and in combination they offer a practical, modular path from triage to deployment. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2508_16839 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | One VLM, Two Roles: Stage-Wise Routing and Specialty-Level Deployment for Clinical Workflows Vassef, Shayan Shimegekar, Soorya Ram Goyal, Abhay Saha, Koustuv Zonooz, Pi Kumar, Navin Artificial Intelligence Clinical ML workflows are often fragmented and inefficient: triage, task selection, and model deployment are handled by a patchwork of task-specific networks. These pipelines are rarely aligned with data-science practice, reducing efficiency and increasing operational cost. They also lack data-driven model identification (from imaging/tabular inputs) and standardized delivery of model outputs. We present a framework that employs a single vision-language model (VLM) in two complementary, modular roles. First (Solution 1): the VLM acts as an aware model-card matcher that routes an incoming image to the appropriate specialist model via a three-stage workflow (modality -> primary abnormality -> model-card ID). Reliability is improved by (i) stage-wise prompts enabling early termination via "None"/"Other" and (ii) a calibrated top-2 answer selector with a stage-wise cutoff. This raises routing accuracy by +9 and +11 percentage points on the training and held-out splits, respectively, compared with a baseline router, and improves held-out calibration (lower Expected Calibration Error, ECE). Second (Solution 2): we fine-tune the same VLM on specialty-specific datasets so that one model per specialty covers multiple downstream tasks, simplifying deployment while maintaining performance. Across gastroenterology, hematology, ophthalmology, pathology, and radiology, this single-model deployment matches or approaches specialized baselines. Together, these solutions reduce data-science effort through more accurate selection, simplify monitoring and maintenance by consolidating task-specific models, and increase transparency via per-stage justifications and calibrated thresholds. Each solution stands alone, and in combination they offer a practical, modular path from triage to deployment. |
| title | One VLM, Two Roles: Stage-Wise Routing and Specialty-Level Deployment for Clinical Workflows |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2508.16839 |