EcoFair: Trustworthy and Energy-Aware Routing for Privacy-Preserving Vertically Partitioned Medical Inference

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Anoosha, Mostafa, Thakker, Dhavalkumar, Paxton, Kuniko, Aslansefat, Koorosh, Mishra, Bhupesh Kumar, Ahmad, Baseer, Kureshi, Rameez Raja
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914426854047744
author Anoosha, Mostafa
Thakker, Dhavalkumar
Paxton, Kuniko
Aslansefat, Koorosh
Mishra, Bhupesh Kumar
Ahmad, Baseer
Kureshi, Rameez Raja
author_facet Anoosha, Mostafa
Thakker, Dhavalkumar
Paxton, Kuniko
Aslansefat, Koorosh
Mishra, Bhupesh Kumar
Ahmad, Baseer
Kureshi, Rameez Raja
contents Privacy-preserving medical inference must balance data locality, diagnostic reliability, and deployment efficiency. This paper presents EcoFair, a simulated vertically partitioned inference framework for dermatological diagnosis in which raw image and tabular data remain local and only modality-specific embeddings are transmitted for server-side multimodal fusion. EcoFair introduces a lightweight-first routing mechanism that selectively activates a heavier image encoder when local uncertainty or metadata-derived clinical risk indicates that additional computation is warranted. The routing decision combines predictive uncertainty, a safe--danger probability gap, and a tabular neurosymbolic risk score derived from patient age and lesion localisation. Experiments on three dermatology benchmarks show that EcoFair can substantially reduce edge-side inference energy in representative model pairings while remaining competitive in classification performance. The results further indicate that selective routing can improve subgroup-sensitive malignant-case behaviour in representative settings without modifying the global training objective. These findings position EcoFair as a practical framework for privacy-preserving and energy-aware medical inference under edge deployment constraints.
format Preprint
id arxiv_https___arxiv_org_abs_2603_26483
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle EcoFair: Trustworthy and Energy-Aware Routing for Privacy-Preserving Vertically Partitioned Medical Inference
Anoosha, Mostafa
Thakker, Dhavalkumar
Paxton, Kuniko
Aslansefat, Koorosh
Mishra, Bhupesh Kumar
Ahmad, Baseer
Kureshi, Rameez Raja
Machine Learning
C.1.4; C.2.8; D.4.6; I.2.11; I.2.6; J.3
Privacy-preserving medical inference must balance data locality, diagnostic reliability, and deployment efficiency. This paper presents EcoFair, a simulated vertically partitioned inference framework for dermatological diagnosis in which raw image and tabular data remain local and only modality-specific embeddings are transmitted for server-side multimodal fusion. EcoFair introduces a lightweight-first routing mechanism that selectively activates a heavier image encoder when local uncertainty or metadata-derived clinical risk indicates that additional computation is warranted. The routing decision combines predictive uncertainty, a safe--danger probability gap, and a tabular neurosymbolic risk score derived from patient age and lesion localisation. Experiments on three dermatology benchmarks show that EcoFair can substantially reduce edge-side inference energy in representative model pairings while remaining competitive in classification performance. The results further indicate that selective routing can improve subgroup-sensitive malignant-case behaviour in representative settings without modifying the global training objective. These findings position EcoFair as a practical framework for privacy-preserving and energy-aware medical inference under edge deployment constraints.
title EcoFair: Trustworthy and Energy-Aware Routing for Privacy-Preserving Vertically Partitioned Medical Inference
topic Machine Learning
C.1.4; C.2.8; D.4.6; I.2.11; I.2.6; J.3
url https://arxiv.org/abs/2603.26483