Machine Learning Decision Support for Detecting Lipohypertrophy With Bedside Ultrasound: Proof-of-Concept Study

Ela Bandari; Tomas Beuzen; Lara Habashy; Javairia Raza; Xudong Yang; Jordanna Kapeluto; Graydon Meneilly; Kenneth Madden

doi:10.2196/34830

Machine Learning Decision Support for Detecting Lipohypertrophy With Bedside Ultrasound: Proof-of-Concept Study

JMIR Form Res. 2022 May 6;6(5):e34830. doi: 10.2196/34830.

Authors

Ela Bandari^#¹, Tomas Beuzen^#¹, Lara Habashy^#¹, Javairia Raza^#¹, Xudong Yang^#¹, Jordanna Kapeluto^{2

3}, Graydon Meneilly^{2

4}, Kenneth Madden^{2

4

5}

Affiliations

¹ Data Science Program, University of British Columbia, Vancouver, BC, Canada.
² Gerontology and Diabetes Research Laboratory, University of British Columbia, Vancouver, BC, Canada.
³ Division of Endocrinology, Department of Medicine, University of British Columbia, Vancouver, BC, Canada.
⁴ Division of Geriatric Medicine, Department of Medicine, University of British Columbia, Vancouver, BC, Canada.
⁵ Centre for Hip Health and Mobility, Vancouver, BC, Canada.

^# Contributed equally.

PMID: 35404833
PMCID: PMC9123536
DOI: 10.2196/34830

Abstract

Background: The most common dermatological complication of insulin therapy is lipohypertrophy.

Objective: As a proof of concept, we built and tested an automated model using a convolutional neural network (CNN) to detect the presence of lipohypertrophy in ultrasound images.

Methods: Ultrasound images were obtained in a blinded fashion using a portable GE LOGIQ e machine with an L8-18I-D probe (5-18 MHz; GE Healthcare). The data were split into train, validation, and test splits of 70%, 15%, and 15%, respectively. Given the small size of the data set, image augmentation techniques were used to expand the size of the training set and improve the model's generalizability. To compare the performance of the different architectures, the team considered the accuracy and recall of the models when tested on our test set.

Results: The DenseNet CNN architecture was found to have the highest accuracy (76%) and recall (76%) in detecting lipohypertrophy in ultrasound images compared to other CNN architectures. Additional work showed that the YOLOv5m object detection model could be used to help detect the approximate location of lipohypertrophy in ultrasound images identified as containing lipohypertrophy by the DenseNet CNN.

Conclusions: We were able to demonstrate the ability of machine learning approaches to automate the process of detecting and locating lipohypertrophy.

Keywords: diabetes; diagnostic ultrasound; insulin; lipohypertrophy; lipoma; machine learning; ultrasound images.

©Ela Bandari, Tomas Beuzen, Lara Habashy, Javairia Raza, Xudong Yang, Jordanna Kapeluto, Graydon Meneilly, Kenneth Madden. Originally published in JMIR Formative Research (https://formative.jmir.org), 06.05.2022.