Home > Research > Publications & Outputs > Robust Methods for Real-time Diabetic Foot Ulce...

Links

Text available via DOI:

View graph of relations

Robust Methods for Real-time Diabetic Foot Ulcer Detection and Localization on Mobile Devices

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Published

Standard

Robust Methods for Real-time Diabetic Foot Ulcer Detection and Localization on Mobile Devices. / Goyal, Manu; Reeves, Neil; Rajbhandari, Satyan et al.
In: IEEE Journal of Biomedical and Health Informatics, Vol. 23, No. 4, 31.07.2019, p. 1730-1741.

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Harvard

Goyal, M, Reeves, N, Rajbhandari, S & Yap, MH 2019, 'Robust Methods for Real-time Diabetic Foot Ulcer Detection and Localization on Mobile Devices', IEEE Journal of Biomedical and Health Informatics, vol. 23, no. 4, pp. 1730-1741. https://doi.org/10.1109/jbhi.2018.2868656

APA

Goyal, M., Reeves, N., Rajbhandari, S., & Yap, M. H. (2019). Robust Methods for Real-time Diabetic Foot Ulcer Detection and Localization on Mobile Devices. IEEE Journal of Biomedical and Health Informatics, 23(4), 1730-1741. https://doi.org/10.1109/jbhi.2018.2868656

Vancouver

Goyal M, Reeves N, Rajbhandari S, Yap MH. Robust Methods for Real-time Diabetic Foot Ulcer Detection and Localization on Mobile Devices. IEEE Journal of Biomedical and Health Informatics. 2019 Jul 31;23(4):1730-1741. Epub 2018 Sept 6. doi: 10.1109/jbhi.2018.2868656

Author

Goyal, Manu ; Reeves, Neil ; Rajbhandari, Satyan et al. / Robust Methods for Real-time Diabetic Foot Ulcer Detection and Localization on Mobile Devices. In: IEEE Journal of Biomedical and Health Informatics. 2019 ; Vol. 23, No. 4. pp. 1730-1741.

Bibtex

@article{862cbeca924345909ad1c3b5e96960d2,
title = "Robust Methods for Real-time Diabetic Foot Ulcer Detection and Localization on Mobile Devices",
abstract = "Current practice for diabetic foot ulcers (DFU) screening involves detection and localization by podiatrists. Existing automated solutions either focus on segmentation or classification. In this work, we design deep learning methods for real-time DFU localization. To produce a robust deep learning model, we collected an extensive database of 1775 images of DFU. Two medical experts produced the ground truths of this data set by outlining the region of interest of DFU with an annotator software. Using five-fold cross-validation, overall, faster R-CNN with InceptionV2 model using two-tier transfer learning achieved a mean average precision of 91.8%, the speed of 48 ms for inferencing a single image and with a model size of 57.2 MB. To demonstrate the robustness and practicality of our solution to realtime prediction, we evaluated the performance of the models on a NVIDIA Jetson TX2 and a smartphone app. This work demonstrates the capability of deep learning in real-time localization of DFU, which can be further improved with a more extensive data set.",
author = "Manu Goyal and Neil Reeves and Satyan Rajbhandari and Yap, {Moi Hoon}",
year = "2019",
month = jul,
day = "31",
doi = "10.1109/jbhi.2018.2868656",
language = "English",
volume = "23",
pages = "1730--1741",
journal = " IEEE Journal of Biomedical and Health Informatics",
issn = "2168-2194",
publisher = "IEEE",
number = "4",

}

RIS

TY - JOUR

T1 - Robust Methods for Real-time Diabetic Foot Ulcer Detection and Localization on Mobile Devices

AU - Goyal, Manu

AU - Reeves, Neil

AU - Rajbhandari, Satyan

AU - Yap, Moi Hoon

PY - 2019/7/31

Y1 - 2019/7/31

N2 - Current practice for diabetic foot ulcers (DFU) screening involves detection and localization by podiatrists. Existing automated solutions either focus on segmentation or classification. In this work, we design deep learning methods for real-time DFU localization. To produce a robust deep learning model, we collected an extensive database of 1775 images of DFU. Two medical experts produced the ground truths of this data set by outlining the region of interest of DFU with an annotator software. Using five-fold cross-validation, overall, faster R-CNN with InceptionV2 model using two-tier transfer learning achieved a mean average precision of 91.8%, the speed of 48 ms for inferencing a single image and with a model size of 57.2 MB. To demonstrate the robustness and practicality of our solution to realtime prediction, we evaluated the performance of the models on a NVIDIA Jetson TX2 and a smartphone app. This work demonstrates the capability of deep learning in real-time localization of DFU, which can be further improved with a more extensive data set.

AB - Current practice for diabetic foot ulcers (DFU) screening involves detection and localization by podiatrists. Existing automated solutions either focus on segmentation or classification. In this work, we design deep learning methods for real-time DFU localization. To produce a robust deep learning model, we collected an extensive database of 1775 images of DFU. Two medical experts produced the ground truths of this data set by outlining the region of interest of DFU with an annotator software. Using five-fold cross-validation, overall, faster R-CNN with InceptionV2 model using two-tier transfer learning achieved a mean average precision of 91.8%, the speed of 48 ms for inferencing a single image and with a model size of 57.2 MB. To demonstrate the robustness and practicality of our solution to realtime prediction, we evaluated the performance of the models on a NVIDIA Jetson TX2 and a smartphone app. This work demonstrates the capability of deep learning in real-time localization of DFU, which can be further improved with a more extensive data set.

U2 - 10.1109/jbhi.2018.2868656

DO - 10.1109/jbhi.2018.2868656

M3 - Journal article

VL - 23

SP - 1730

EP - 1741

JO - IEEE Journal of Biomedical and Health Informatics

JF - IEEE Journal of Biomedical and Health Informatics

SN - 2168-2194

IS - 4

ER -