Enhancing Fracture Detection in Remote Settings: Evaluating the Efficacy of FIXUS AI Deep Learning Algorithms in Identifying Fifth Metatarsal Fractures Using Mixed-Quality X-rays

Category: Midfoot/Forefoot; Trauma Introduction/Purpose: The diagnosis of fractures can be challenging in specific medical settings due to limited expertise or time. While deep learning has shown promising results, its use is confined to the quality of images and the hassle of importing images to th...

Full description

Saved in:
Bibliographic Details
Main Authors: Atta Taseh MD, Alireza Gholipour PhD, Mani Eftekhari, Alireza Ebrahimi MD, Alexandra F. Flaherty MD, MS, Sumner Jones, Varun Nukala, Gregory R. Waryasz MD, Daniel Guss MD, MBA, John Y. Kwon MD, Christopher W. DiGiovanni MD, Lorena Bejarano-Pineda MD, Soheil Ashkani-Esfahani MD
Format: Article
Language:English
Published: SAGE Publishing 2024-12-01
Series:Foot & Ankle Orthopaedics
Online Access:https://doi.org/10.1177/2473011424S00269
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850132992784072704
author Atta Taseh MD
Alireza Gholipour PhD
Mani Eftekhari
Alireza Ebrahimi MD
Alexandra F. Flaherty MD, MS
Alexandra F. Flaherty MD, MS
Sumner Jones
Varun Nukala
Gregory R. Waryasz MD
Daniel Guss MD, MBA
John Y. Kwon MD
Christopher W. DiGiovanni MD
Lorena Bejarano-Pineda MD
Soheil Ashkani-Esfahani MD
author_facet Atta Taseh MD
Alireza Gholipour PhD
Mani Eftekhari
Alireza Ebrahimi MD
Alexandra F. Flaherty MD, MS
Alexandra F. Flaherty MD, MS
Sumner Jones
Varun Nukala
Gregory R. Waryasz MD
Daniel Guss MD, MBA
John Y. Kwon MD
Christopher W. DiGiovanni MD
Lorena Bejarano-Pineda MD
Soheil Ashkani-Esfahani MD
author_sort Atta Taseh MD
collection DOAJ
description Category: Midfoot/Forefoot; Trauma Introduction/Purpose: The diagnosis of fractures can be challenging in specific medical settings due to limited expertise or time. While deep learning has shown promising results, its use is confined to the quality of images and the hassle of importing images to the models. This study aims to develop a model to detect fifth metatarsal fractures based on cell phone photos of radiographs directly taken from a regular screen. Methods: retrospective case-control study was conducted including patients aged > 18 years with fifth metatarsal fractures (n=1240) (Fx), and healthy controls (n=1224) (NoF). Three view radiographs (anterior, posterior, and lateral) were obtained from the Electronic Health Record (EHR) in PNG format. To generate a mixed-quality dataset, Android and iOS smartphones (SP) were used to create two separate datasets for each Fx and NoF group. Two separate deep learning models on each EHR, SP, and combined datasets were developed using Inception V3 architecture (Figure 1.). The models were also tested on a separate SP dataset (SP-test) that was not included in the development process. Area Under the Receiver Operating Characteristics Curve (AUC) along with other performance metrics were calculated and reported. Continuous data were presented as median (interquartile range), and a p-value of < 0.05 was considered to be significant. Results: Baseline analysis revealed differences between the groups with a median age of 56 years (36-68) for the Fx group, and 62 years (51-72) for the NoF group (p < 0.001). Similarly, the racial composition of the groups was also different (Fx: 84.8% white; NoF: 92.2% white; p < 0.001). Initially, the SP model showed the best performance (Youden Index (YI): 0.92, AUC: 0.99) followed by the EHR (YI: 0.74, AUC: 0.96), and combined (YI: 0.52, AUC: 0.97) models. When tested on the SP-test dataset the EHR model’s performance dropped markedly showing a YI of 0.33, an AUC of 0.78, and a sensitivity of 0.49 (Table 1.). However, the SP and combined models continued to perform optimally (YI: 0.94, AUC: 0.99; YI: 0.78, AUC: 0.98, respectively). Conclusion: This study highlights the crucial role of image quality in developing deep learning models for detecting fifth metatarsal fractures. Our findings demonstrate a markedly reduced performance of the EHR model in identifying fractures within lower-quality images. This emphasizes the need for training algorithms on images of varying quality to create more generalizable models capable of operating effectively across diverse settings.
format Article
id doaj-art-491ab898f4cd4ac292cee9d41623af21
institution OA Journals
issn 2473-0114
language English
publishDate 2024-12-01
publisher SAGE Publishing
record_format Article
series Foot & Ankle Orthopaedics
spelling doaj-art-491ab898f4cd4ac292cee9d41623af212025-08-20T02:32:05ZengSAGE PublishingFoot & Ankle Orthopaedics2473-01142024-12-01910.1177/2473011424S00269Enhancing Fracture Detection in Remote Settings: Evaluating the Efficacy of FIXUS AI Deep Learning Algorithms in Identifying Fifth Metatarsal Fractures Using Mixed-Quality X-raysAtta Taseh MDAlireza Gholipour PhDMani EftekhariAlireza Ebrahimi MDAlexandra F. Flaherty MD, MSAlexandra F. Flaherty MD, MSSumner JonesVarun NukalaGregory R. Waryasz MDDaniel Guss MD, MBAJohn Y. Kwon MDChristopher W. DiGiovanni MDLorena Bejarano-Pineda MDSoheil Ashkani-Esfahani MDCategory: Midfoot/Forefoot; Trauma Introduction/Purpose: The diagnosis of fractures can be challenging in specific medical settings due to limited expertise or time. While deep learning has shown promising results, its use is confined to the quality of images and the hassle of importing images to the models. This study aims to develop a model to detect fifth metatarsal fractures based on cell phone photos of radiographs directly taken from a regular screen. Methods: retrospective case-control study was conducted including patients aged > 18 years with fifth metatarsal fractures (n=1240) (Fx), and healthy controls (n=1224) (NoF). Three view radiographs (anterior, posterior, and lateral) were obtained from the Electronic Health Record (EHR) in PNG format. To generate a mixed-quality dataset, Android and iOS smartphones (SP) were used to create two separate datasets for each Fx and NoF group. Two separate deep learning models on each EHR, SP, and combined datasets were developed using Inception V3 architecture (Figure 1.). The models were also tested on a separate SP dataset (SP-test) that was not included in the development process. Area Under the Receiver Operating Characteristics Curve (AUC) along with other performance metrics were calculated and reported. Continuous data were presented as median (interquartile range), and a p-value of < 0.05 was considered to be significant. Results: Baseline analysis revealed differences between the groups with a median age of 56 years (36-68) for the Fx group, and 62 years (51-72) for the NoF group (p < 0.001). Similarly, the racial composition of the groups was also different (Fx: 84.8% white; NoF: 92.2% white; p < 0.001). Initially, the SP model showed the best performance (Youden Index (YI): 0.92, AUC: 0.99) followed by the EHR (YI: 0.74, AUC: 0.96), and combined (YI: 0.52, AUC: 0.97) models. When tested on the SP-test dataset the EHR model’s performance dropped markedly showing a YI of 0.33, an AUC of 0.78, and a sensitivity of 0.49 (Table 1.). However, the SP and combined models continued to perform optimally (YI: 0.94, AUC: 0.99; YI: 0.78, AUC: 0.98, respectively). Conclusion: This study highlights the crucial role of image quality in developing deep learning models for detecting fifth metatarsal fractures. Our findings demonstrate a markedly reduced performance of the EHR model in identifying fractures within lower-quality images. This emphasizes the need for training algorithms on images of varying quality to create more generalizable models capable of operating effectively across diverse settings.https://doi.org/10.1177/2473011424S00269
spellingShingle Atta Taseh MD
Alireza Gholipour PhD
Mani Eftekhari
Alireza Ebrahimi MD
Alexandra F. Flaherty MD, MS
Alexandra F. Flaherty MD, MS
Sumner Jones
Varun Nukala
Gregory R. Waryasz MD
Daniel Guss MD, MBA
John Y. Kwon MD
Christopher W. DiGiovanni MD
Lorena Bejarano-Pineda MD
Soheil Ashkani-Esfahani MD
Enhancing Fracture Detection in Remote Settings: Evaluating the Efficacy of FIXUS AI Deep Learning Algorithms in Identifying Fifth Metatarsal Fractures Using Mixed-Quality X-rays
Foot & Ankle Orthopaedics
title Enhancing Fracture Detection in Remote Settings: Evaluating the Efficacy of FIXUS AI Deep Learning Algorithms in Identifying Fifth Metatarsal Fractures Using Mixed-Quality X-rays
title_full Enhancing Fracture Detection in Remote Settings: Evaluating the Efficacy of FIXUS AI Deep Learning Algorithms in Identifying Fifth Metatarsal Fractures Using Mixed-Quality X-rays
title_fullStr Enhancing Fracture Detection in Remote Settings: Evaluating the Efficacy of FIXUS AI Deep Learning Algorithms in Identifying Fifth Metatarsal Fractures Using Mixed-Quality X-rays
title_full_unstemmed Enhancing Fracture Detection in Remote Settings: Evaluating the Efficacy of FIXUS AI Deep Learning Algorithms in Identifying Fifth Metatarsal Fractures Using Mixed-Quality X-rays
title_short Enhancing Fracture Detection in Remote Settings: Evaluating the Efficacy of FIXUS AI Deep Learning Algorithms in Identifying Fifth Metatarsal Fractures Using Mixed-Quality X-rays
title_sort enhancing fracture detection in remote settings evaluating the efficacy of fixus ai deep learning algorithms in identifying fifth metatarsal fractures using mixed quality x rays
url https://doi.org/10.1177/2473011424S00269
work_keys_str_mv AT attatasehmd enhancingfracturedetectioninremotesettingsevaluatingtheefficacyoffixusaideeplearningalgorithmsinidentifyingfifthmetatarsalfracturesusingmixedqualityxrays
AT alirezagholipourphd enhancingfracturedetectioninremotesettingsevaluatingtheefficacyoffixusaideeplearningalgorithmsinidentifyingfifthmetatarsalfracturesusingmixedqualityxrays
AT manieftekhari enhancingfracturedetectioninremotesettingsevaluatingtheefficacyoffixusaideeplearningalgorithmsinidentifyingfifthmetatarsalfracturesusingmixedqualityxrays
AT alirezaebrahimimd enhancingfracturedetectioninremotesettingsevaluatingtheefficacyoffixusaideeplearningalgorithmsinidentifyingfifthmetatarsalfracturesusingmixedqualityxrays
AT alexandrafflahertymdms enhancingfracturedetectioninremotesettingsevaluatingtheefficacyoffixusaideeplearningalgorithmsinidentifyingfifthmetatarsalfracturesusingmixedqualityxrays
AT alexandrafflahertymdms enhancingfracturedetectioninremotesettingsevaluatingtheefficacyoffixusaideeplearningalgorithmsinidentifyingfifthmetatarsalfracturesusingmixedqualityxrays
AT sumnerjones enhancingfracturedetectioninremotesettingsevaluatingtheefficacyoffixusaideeplearningalgorithmsinidentifyingfifthmetatarsalfracturesusingmixedqualityxrays
AT varunnukala enhancingfracturedetectioninremotesettingsevaluatingtheefficacyoffixusaideeplearningalgorithmsinidentifyingfifthmetatarsalfracturesusingmixedqualityxrays
AT gregoryrwaryaszmd enhancingfracturedetectioninremotesettingsevaluatingtheefficacyoffixusaideeplearningalgorithmsinidentifyingfifthmetatarsalfracturesusingmixedqualityxrays
AT danielgussmdmba enhancingfracturedetectioninremotesettingsevaluatingtheefficacyoffixusaideeplearningalgorithmsinidentifyingfifthmetatarsalfracturesusingmixedqualityxrays
AT johnykwonmd enhancingfracturedetectioninremotesettingsevaluatingtheefficacyoffixusaideeplearningalgorithmsinidentifyingfifthmetatarsalfracturesusingmixedqualityxrays
AT christopherwdigiovannimd enhancingfracturedetectioninremotesettingsevaluatingtheefficacyoffixusaideeplearningalgorithmsinidentifyingfifthmetatarsalfracturesusingmixedqualityxrays
AT lorenabejaranopinedamd enhancingfracturedetectioninremotesettingsevaluatingtheefficacyoffixusaideeplearningalgorithmsinidentifyingfifthmetatarsalfracturesusingmixedqualityxrays
AT soheilashkaniesfahanimd enhancingfracturedetectioninremotesettingsevaluatingtheefficacyoffixusaideeplearningalgorithmsinidentifyingfifthmetatarsalfracturesusingmixedqualityxrays