Diverse Representation Knowledge Distillation for Efficient Edge AI Teledermatology in Skin Disease Diagnosis

Access to dermatological care in rural areas is limited due to a shortage of specialists. While AI-powered teledermatology offers a solution, it faces challenges from unreliable internet connectivity. Edge AI offers a promising approach that enables inferences locally on mobile devices. However, hig...

Full description

Saved in:
Bibliographic Details
Main Authors: Andreas Winata, Nur Afny Catur Andryani, Alexander Agung Santoso Gunawan, and Ford Lumban Gaol
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/11044323/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850169010824413184
author Andreas Winata
Nur Afny Catur Andryani
Alexander Agung Santoso Gunawan
and Ford Lumban Gaol
author_facet Andreas Winata
Nur Afny Catur Andryani
Alexander Agung Santoso Gunawan
and Ford Lumban Gaol
author_sort Andreas Winata
collection DOAJ
description Access to dermatological care in rural areas is limited due to a shortage of specialists. While AI-powered teledermatology offers a solution, it faces challenges from unreliable internet connectivity. Edge AI offers a promising approach that enables inferences locally on mobile devices. However, high-performance AI models are often large, which makes them difficult to deploy on mobile devices. While recent research primarily concentrated on small models or cloud inference, this research addresses the underexplored application of knowledge distillation on mobile devices. By conducting applied research aiming at compressing the AI model while maintaining accuracy, this research proposes a novel adoption framework for skin disease inference using edge computing. The framework applies knowledge distillation to compress a high-performance teacher model into a small student model, followed by deployment on mobile devices for local inference. Key contributions include a novel framework for edge-based inference, evaluation of pre-trained models on the ISIC 2019 and Fitzpatrick17k-C datasets, and practical deployment on mobile devices. Model performance was evaluated using accuracy, precision, recall, and F1-score, while the prototype was measured using model size, compression ratio, and inference time. The experiments demonstrate that the scenario of distilling RegNetY32GF into MobileNetV2 results in the most efficient model, maintaining both accuracy and model size. The prototype evaluation shows the practicality of the proposed framework with a compression rate of 55.88 on the ISIC 2019 dataset, with inference time reduced by approximately 352 times. These results enable efficient inference on mobile devices with limited resources.
format Article
id doaj-art-09bf05de3a6f468eae4b0d50c9d7256e
institution OA Journals
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-09bf05de3a6f468eae4b0d50c9d7256e2025-08-20T02:20:51ZengIEEEIEEE Access2169-35362025-01-011310661810663310.1109/ACCESS.2025.358122511044323Diverse Representation Knowledge Distillation for Efficient Edge AI Teledermatology in Skin Disease DiagnosisAndreas Winata0https://orcid.org/0009-0009-9467-1748Nur Afny Catur Andryani1Alexander Agung Santoso Gunawan2and Ford Lumban Gaol3https://orcid.org/0000-0002-5116-5708Computer Science Department, BINUS Graduate ProgramÓDoctor of Computer Science, Bina Nusantara University, Jakarta, IndonesiaComputer Science Department, BINUS Graduate ProgramÓDoctor of Computer Science, Bina Nusantara University, Jakarta, IndonesiaComputer Science Department, School of Computer Science, Bina Nusantara University, Jakarta, IndonesiaComputer Science Department, BINUS Graduate ProgramÓDoctor of Computer Science, Bina Nusantara University, Jakarta, IndonesiaAccess to dermatological care in rural areas is limited due to a shortage of specialists. While AI-powered teledermatology offers a solution, it faces challenges from unreliable internet connectivity. Edge AI offers a promising approach that enables inferences locally on mobile devices. However, high-performance AI models are often large, which makes them difficult to deploy on mobile devices. While recent research primarily concentrated on small models or cloud inference, this research addresses the underexplored application of knowledge distillation on mobile devices. By conducting applied research aiming at compressing the AI model while maintaining accuracy, this research proposes a novel adoption framework for skin disease inference using edge computing. The framework applies knowledge distillation to compress a high-performance teacher model into a small student model, followed by deployment on mobile devices for local inference. Key contributions include a novel framework for edge-based inference, evaluation of pre-trained models on the ISIC 2019 and Fitzpatrick17k-C datasets, and practical deployment on mobile devices. Model performance was evaluated using accuracy, precision, recall, and F1-score, while the prototype was measured using model size, compression ratio, and inference time. The experiments demonstrate that the scenario of distilling RegNetY32GF into MobileNetV2 results in the most efficient model, maintaining both accuracy and model size. The prototype evaluation shows the practicality of the proposed framework with a compression rate of 55.88 on the ISIC 2019 dataset, with inference time reduced by approximately 352 times. These results enable efficient inference on mobile devices with limited resources.https://ieeexplore.ieee.org/document/11044323/Edge computingimage classificationknowledge distillationmobile applicationsteledermatology
spellingShingle Andreas Winata
Nur Afny Catur Andryani
Alexander Agung Santoso Gunawan
and Ford Lumban Gaol
Diverse Representation Knowledge Distillation for Efficient Edge AI Teledermatology in Skin Disease Diagnosis
IEEE Access
Edge computing
image classification
knowledge distillation
mobile applications
teledermatology
title Diverse Representation Knowledge Distillation for Efficient Edge AI Teledermatology in Skin Disease Diagnosis
title_full Diverse Representation Knowledge Distillation for Efficient Edge AI Teledermatology in Skin Disease Diagnosis
title_fullStr Diverse Representation Knowledge Distillation for Efficient Edge AI Teledermatology in Skin Disease Diagnosis
title_full_unstemmed Diverse Representation Knowledge Distillation for Efficient Edge AI Teledermatology in Skin Disease Diagnosis
title_short Diverse Representation Knowledge Distillation for Efficient Edge AI Teledermatology in Skin Disease Diagnosis
title_sort diverse representation knowledge distillation for efficient edge ai teledermatology in skin disease diagnosis
topic Edge computing
image classification
knowledge distillation
mobile applications
teledermatology
url https://ieeexplore.ieee.org/document/11044323/
work_keys_str_mv AT andreaswinata diverserepresentationknowledgedistillationforefficientedgeaiteledermatologyinskindiseasediagnosis
AT nurafnycaturandryani diverserepresentationknowledgedistillationforefficientedgeaiteledermatologyinskindiseasediagnosis
AT alexanderagungsantosogunawan diverserepresentationknowledgedistillationforefficientedgeaiteledermatologyinskindiseasediagnosis
AT andfordlumbangaol diverserepresentationknowledgedistillationforefficientedgeaiteledermatologyinskindiseasediagnosis