Smartphone-Based Pupillometry Using Machine Learning for the Diagnosis of Sports-Related Concussion

<b>Background:</b> Quantitative pupillometry has been proposed as an objective means to diagnose acute sports-related concussion (SRC). <b>Objective:</b> To assess the diagnostic accuracy of a smartphone-based quantitative pupillometer in the acute diagnosis of SRC. <b>...

Full description

Saved in:
Bibliographic Details
Main Authors: Anthony J. Maxin, Bridget M. Whelan, Michael R. Levitt, Lynn B. McGrath, Kimberly G. Harmon
Format: Article
Language:English
Published: MDPI AG 2024-12-01
Series:Diagnostics
Subjects:
Online Access:https://www.mdpi.com/2075-4418/14/23/2723
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1846124262228754432
author Anthony J. Maxin
Bridget M. Whelan
Michael R. Levitt
Lynn B. McGrath
Kimberly G. Harmon
author_facet Anthony J. Maxin
Bridget M. Whelan
Michael R. Levitt
Lynn B. McGrath
Kimberly G. Harmon
author_sort Anthony J. Maxin
collection DOAJ
description <b>Background:</b> Quantitative pupillometry has been proposed as an objective means to diagnose acute sports-related concussion (SRC). <b>Objective:</b> To assess the diagnostic accuracy of a smartphone-based quantitative pupillometer in the acute diagnosis of SRC. <b>Methods:</b> Division I college football players had baseline pupillometry including pupillary light reflex (PLR) parameters of maximum resting diameter, minimum diameter after light stimulus, percent change in pupil diameter, latency of pupil constriction onset, mean constriction velocity, maximum constriction velocity, and mean dilation velocity using a smartphone-based app. When an SRC occurred, athletes had the smartphone pupillometry repeated as part of their concussion testing. All combinations of the seven PLR parameters were tested in machine learning binary classification models to determine the optimal combination for differentiating between non-concussed and concussed athletes. <b>Results</b>: 93 football athletes underwent baseline pupillometry testing. Among these athletes, 11 suffered future SRC and had pupillometry recordings repeated at the time of diagnosis. In the machine learning pupillometry analysis that used the synthetic minority oversampling technique to account for the significant class imbalance in our dataset, the best-performing model was a random forest algorithm with the combination of latency, maximum diameter, minimum diameter, mean constriction velocity, and maximum constriction velocity PLR parameters as feature inputs. This model produced 91% overall accuracy, 98% sensitivity, 84.2% specificity, area under the curve (AUC) of 0.91, and an F1 score of 91.6% in differentiating between baseline and SRC recordings. In the machine learning analysis prior to oversampling of our imbalanced dataset, the best-performing model was k-nearest neighbors using latency, maximum diameter, maximum constriction velocity, and mean dilation velocity to produce 82% accuracy, 40% sensitivity, 87% specificity, AUC of 0.64, and F1 score of 24%. <b>Conclusions</b>: Smartphone pupillometry in combination with machine learning may provide fast and objective SRC diagnosis in football athletes.
format Article
id doaj-art-067c622e297d4ec7bbbfb93edcefcf0e
institution Kabale University
issn 2075-4418
language English
publishDate 2024-12-01
publisher MDPI AG
record_format Article
series Diagnostics
spelling doaj-art-067c622e297d4ec7bbbfb93edcefcf0e2024-12-13T16:24:48ZengMDPI AGDiagnostics2075-44182024-12-011423272310.3390/diagnostics14232723Smartphone-Based Pupillometry Using Machine Learning for the Diagnosis of Sports-Related ConcussionAnthony J. Maxin0Bridget M. Whelan1Michael R. Levitt2Lynn B. McGrath3Kimberly G. Harmon4Department of Neurological Surgery, University of Washington, Seattle, WA 98195, USADepartment of Family Medicine, Sports Medicine Section, University of Washington, Seattle, WA 98195, USADepartment of Neurological Surgery, University of Washington, Seattle, WA 98195, USADepartment of Neurological Surgery, Northern Light Health, Portland, ME 04102, USADepartment of Family Medicine, Sports Medicine Section, University of Washington, Seattle, WA 98195, USA<b>Background:</b> Quantitative pupillometry has been proposed as an objective means to diagnose acute sports-related concussion (SRC). <b>Objective:</b> To assess the diagnostic accuracy of a smartphone-based quantitative pupillometer in the acute diagnosis of SRC. <b>Methods:</b> Division I college football players had baseline pupillometry including pupillary light reflex (PLR) parameters of maximum resting diameter, minimum diameter after light stimulus, percent change in pupil diameter, latency of pupil constriction onset, mean constriction velocity, maximum constriction velocity, and mean dilation velocity using a smartphone-based app. When an SRC occurred, athletes had the smartphone pupillometry repeated as part of their concussion testing. All combinations of the seven PLR parameters were tested in machine learning binary classification models to determine the optimal combination for differentiating between non-concussed and concussed athletes. <b>Results</b>: 93 football athletes underwent baseline pupillometry testing. Among these athletes, 11 suffered future SRC and had pupillometry recordings repeated at the time of diagnosis. In the machine learning pupillometry analysis that used the synthetic minority oversampling technique to account for the significant class imbalance in our dataset, the best-performing model was a random forest algorithm with the combination of latency, maximum diameter, minimum diameter, mean constriction velocity, and maximum constriction velocity PLR parameters as feature inputs. This model produced 91% overall accuracy, 98% sensitivity, 84.2% specificity, area under the curve (AUC) of 0.91, and an F1 score of 91.6% in differentiating between baseline and SRC recordings. In the machine learning analysis prior to oversampling of our imbalanced dataset, the best-performing model was k-nearest neighbors using latency, maximum diameter, maximum constriction velocity, and mean dilation velocity to produce 82% accuracy, 40% sensitivity, 87% specificity, AUC of 0.64, and F1 score of 24%. <b>Conclusions</b>: Smartphone pupillometry in combination with machine learning may provide fast and objective SRC diagnosis in football athletes.https://www.mdpi.com/2075-4418/14/23/2723smartphone pupillometrysports-related concussiondiagnosticsbiomarkerspupillary light reflexdigital health
spellingShingle Anthony J. Maxin
Bridget M. Whelan
Michael R. Levitt
Lynn B. McGrath
Kimberly G. Harmon
Smartphone-Based Pupillometry Using Machine Learning for the Diagnosis of Sports-Related Concussion
Diagnostics
smartphone pupillometry
sports-related concussion
diagnostics
biomarkers
pupillary light reflex
digital health
title Smartphone-Based Pupillometry Using Machine Learning for the Diagnosis of Sports-Related Concussion
title_full Smartphone-Based Pupillometry Using Machine Learning for the Diagnosis of Sports-Related Concussion
title_fullStr Smartphone-Based Pupillometry Using Machine Learning for the Diagnosis of Sports-Related Concussion
title_full_unstemmed Smartphone-Based Pupillometry Using Machine Learning for the Diagnosis of Sports-Related Concussion
title_short Smartphone-Based Pupillometry Using Machine Learning for the Diagnosis of Sports-Related Concussion
title_sort smartphone based pupillometry using machine learning for the diagnosis of sports related concussion
topic smartphone pupillometry
sports-related concussion
diagnostics
biomarkers
pupillary light reflex
digital health
url https://www.mdpi.com/2075-4418/14/23/2723
work_keys_str_mv AT anthonyjmaxin smartphonebasedpupillometryusingmachinelearningforthediagnosisofsportsrelatedconcussion
AT bridgetmwhelan smartphonebasedpupillometryusingmachinelearningforthediagnosisofsportsrelatedconcussion
AT michaelrlevitt smartphonebasedpupillometryusingmachinelearningforthediagnosisofsportsrelatedconcussion
AT lynnbmcgrath smartphonebasedpupillometryusingmachinelearningforthediagnosisofsportsrelatedconcussion
AT kimberlygharmon smartphonebasedpupillometryusingmachinelearningforthediagnosisofsportsrelatedconcussion