Eyelid Measurements: Smartphone-Based Artificial Intelligence-Assisted Prediction (Preprint)

BACKGROUND Margin reflex distance 1(MRD1), margin reflex distance 2 (MRD2), and levator muscle function (LF) are crucial for ptosis evaluation and management. Manual measurements of MRD1, MRD2, and LF are time-consuming, subjective, and prone to human error. Smartphone-based artificial intelligence (AI) image processing is a potential solution to overcome these limitations. OBJECTIVE We proposed the first smartphone-based AI-assisted image processing algorithm for MRD1, MRD2, and LF measurements. METHODS This observational study included 822 eyes of 411 volunteers aged over 18 years from August 1, 2020, to April 30, 2021. Six orbital photographs (bilateral primary gaze, up-gaze, and down-gaze) were taken using a smartphone (iPhone 11 pro max). The gold standard measurements and normalized eye photographs were obtained from these orbital photographs and compiled using AI-assisted software to create MRD1, MRD2 and LF models. RESULTS The Pearson correlation coefficients between the gold standard measurements and the predicted values obtained with the MRD1 and MRD2 models were excellent (r = 0.91, and 0.88, respectively) and with the LF model were good (r = 0.73). The intraclass correlation coefficient results showed excellent agreement between the gold standard measurements and the values predicted by the MRD1and MRD2 models (0.90, and 0.84, respectively), and substantial agreement with the LF model (0.69). The mean absolute errors were 0.35 mm, 0.37 mm, and 1.06 mm for MRD1, MRD2, and LF models, respectively. The 95% limits of agreement were -0.94 to 0.94 mm for the MRD1 model; -0.92 to 1.03 mm for the MRD2 model; and -0.63 to 2.53 mm for the LF model. CONCLUSIONS In this study, we proposed the first smartphone-based AI-assisted image processing algorithm for eyelid measurements. MRD1, MRD2, and LF measures can be taken in a quick, objective, and convenient manner. Furthermore, by using a smartphone, the examiner can check these measurements anywhere and at any time, which facilitates data collection.

Download Full-text

Reproducibility and Validity of the 10-Meter Shuttle Ride Test in Wheelchair-Using Children and Adolescents With Cerebral Palsy

Physical Therapy ◽

10.2522/ptj.20120513 ◽

2013 ◽

Vol 93 (7) ◽

pp. 967-974 ◽

Cited By ~ 29

Author(s):

Olaf Verschuren ◽

Maremka Zwinkels ◽

Marjolijn Ketelaar ◽

Femke Reijnders-van Son ◽

Tim Takken

Keyword(s):

Cerebral Palsy ◽

Cardiorespiratory Fitness ◽

Gold Standard ◽

Pearson Correlation ◽

Intraclass Correlation ◽

Correlation Coefficients ◽

Peak Oxygen Uptake ◽

Repeated Measurements ◽

Manual Wheelchair ◽

Intraclass Correlation Coefficients

BackgroundFor children with cerebral palsy (CP) who are able to walk or run, the 10-m shuttle run test is currently the test of choice to assess cardiorespiratory fitness. This test, however, has not yet been examined in wheelchair-using youth with CP.ObjectiveThe purpose of this study was to investigate the test-retest reproducibility and validity of the 10-m shuttle ride test (SRiT) in youth with CP.DesignRepeated measurements of the SRiT were obtained.MethodsTwenty-three individuals with spastic CP (18 boys, 5 girls; mean age=13.3 years, SD=3.6 years) using a manual wheelchair for at least part of the day participated in this study. During the study, all participants performed one graded arm exercise test (GAET) and 2 identical SRiTs within 2 weeks. Peak oxygen uptake (V̇o2peak), peak heart rate (HRpeak), and respiratory exchange ratio (RER) were recorded. Intraclass correlation coefficients (2,1), the smallest detectable difference, and the limits of agreement (LOA) were calculated. The association between the results of the SRiT and GAET was tested using Pearson correlation coefficients.ResultsIntraclass correlation coefficients (.99, 95% confidence interval=.98–1.00) for all variables indicated highly acceptable reproducibility. The LOA analysis revealed satisfactory levels of agreement. The SRiT variables demonstrated strong, significant positive correlations for V̇o2peak values obtained during the SRiT and the GAET (r=.84, P<.01).LimitationsAlthough the GAET is considered the gold standard, the cardiorespiratory demand during the GAET was significantly lower compared with during the SRiT. Future studies should determine whether the GAET can still be accepted as the gold standard for upper-extremity exercise.ConclusionsThe SRiT is a reproducible and valid test for measuring cardiorespiratory fitness in youth with spastic CP who self-propel a manual wheelchair.

Download Full-text

Validation of a 3D Camera System for Cycling Analysis

Sensors ◽

10.3390/s21134473 ◽

2021 ◽

Vol 21 (13) ◽

pp. 4473

Author(s):

Robson Dias Scoz ◽

Thiago Roberto Espindola ◽

Mateus Freitas Santiago ◽

Paulo Rui de Oliveira ◽

Bruno Mazziotti Oliveira Alves ◽

...

Keyword(s):

Gold Standard ◽

Sports Medicine ◽

Pearson Correlation ◽

Intraclass Correlation ◽

Correlation Coefficients ◽

Camera System ◽

Camera Systems ◽

3D Camera ◽

High Degree ◽

Very High

Background: Kinematic analysis aimed toward scientific investigation or professional purposes is commonly unaffordable and complex to use. Objective: The purpose of this study was to verify concurrent validation between a cycling-specific 3D camera and the gold-standard 3D general camera systems. Methods: Overall, 11 healthy amateur male triathletes were filmed riding their bicycles with Vicon 3D cameras and the Retul 3D cameras for bike fitting analysis simultaneously. All 18 kinematic measurements given by the bike fitting system were compared with the same data given by Vicon cameras through Pearson correlation (r), intraclass correlation coefficients (ICC), standard error measurements (SEM), and Bland–Altman (BA) analysis. Confidence intervals of 95% are given. Results: A very high correlation between cameras was found on six of 18 measurements. All other presented a high correlation between cameras (between 0.7 and 0.9). In total, six variables indicate a SEM of less than one degree between systems. Only two variables indicate a SEM higher than two degrees between camera systems. Overall, four measures indicate bias tendency according to BA. Conclusions: The cycling-specific led-emitting 3D camera system tested revealed a high or very high degree of correlation with the gold-standard 3D camera system used in laboratory motion capture. In total, 14 measurements of this equipment could be used in sports medicine clinical practice and even by researchers of cycling studies.

Download Full-text

Human Activity Classification for Surveillance using Machine Learning and Image Processing Algorithm

International Journal of Computer Sciences and Engineering ◽

10.26438/ijcse/v7i6.343346 ◽

2019 ◽

Vol 7 (6) ◽

pp. 343-346

Author(s):

Arvind Malge ◽

Mallikarjuna Shastry P.M

Keyword(s):

Machine Learning ◽

Image Processing ◽

Human Activity ◽

Processing Algorithm ◽

Image Processing Algorithm ◽

Activity Classification

Download Full-text

Design of an Automatic Light Intensity Control System and Image Processing Algorithm for the Development of an Assistive Drug Screening Device

SSRN Electronic Journal ◽

10.2139/ssrn.3472593 ◽

2019 ◽

Author(s):

John Louis Cruz ◽

Edryl Aljecera ◽

Arlenvy Ayson ◽

Dennis Jefferson Amora ◽

Shearyl Arenas ◽

...

Keyword(s):

Image Processing ◽

Control System ◽

Light Intensity ◽

Drug Screening ◽

Processing Algorithm ◽

Image Processing Algorithm ◽

Intensity Control ◽

Screening Device

Download Full-text

Image processing algorithm of computer-aided diagnosis in lung cancer screening by CT

Systems and Computers in Japan ◽

10.1002/scj.20156 ◽

2005 ◽

Vol 36 (7) ◽

pp. 40-53 ◽

Cited By ~ 2

Author(s):

Shinji Yamamoto

Keyword(s):

Lung Cancer ◽

Image Processing ◽

Cancer Screening ◽

Lung Cancer Screening ◽

Computer Aided Diagnosis ◽

Processing Algorithm ◽

Image Processing Algorithm ◽

Computer Aided ◽

Aided Diagnosis

Download Full-text

Internal quality evaluation of chestnut using nuclear magnetic resonance

International Journal of Food Engineering ◽

10.1515/ijfe-2019-0389 ◽

2020 ◽

Vol 0 (0) ◽

Author(s):

Soo Hyun Park ◽

Sang Ha Noh ◽

Michael J. McCarthy ◽

Seong Min Kim

Keyword(s):

Image Processing ◽

Nuclear Magnetic Resonance ◽

Magnetic Resonance ◽

Processing Algorithm ◽

Internal Quality ◽

Soluble Solid Content ◽

Mr Images ◽

Image Processing Algorithm ◽

Internal Defects ◽

Nuclear Magnetic

AbstractThis study was carried out to develop a prediction model for soluble solid content (SSC) of intact chestnut and to detect internal defects using nuclear magnetic resonance (NMR) relaxometry and magnetic resonance imaging (MRI). Inversion recovery and Carr–Purcell–Meiboom–Gill (CPMG) pulse sequences used to determine the longitudinal (T1) and transverse (T2) relaxation times, respectively. Partial least squares regression (PLSR) was adopted to predict SSCs of chestnuts with NMR data and histograms from MR images. The coefficient of determination (R2), root mean square error of prediction (RMSEP), ratio of prediction to deviation (RPD), and the ratio of error range (RER) of the optimized model to predict SSC were 0.77, 1.41 °Brix, 1.86, and 11.31 with a validation set. Furthermore, an image-processing algorithm has been developed to detect internal defects such as decay, mold, and cavity using MR images. The classification applied with the developed image processing algorithm was over 94% accurate to classify. Based on the results obtained, it was determined that the NMR signal could be applied for grading several levels by SSC, and MRI could be used to evaluate the internal qualities of chestnuts.

Download Full-text

Image processing and machine learning for telehealth craniosynostosis screening in newborns

Journal of Neurosurgery Pediatrics ◽

10.3171/2020.9.peds20605 ◽

2021 ◽

pp. 1-8

Author(s):

Markus J. Bookland ◽

Edward S. Ahn ◽

Petronella Stoltz ◽

Jonathan E. Martin

Keyword(s):

Image Processing ◽

Intraclass Correlation ◽

Correlation Coefficients ◽

Asymmetry Index ◽

Optical Measurements ◽

Shape Classification ◽

Data Set ◽

Intraclass Correlation Coefficients ◽

Optical Scanner ◽

Cranial Shape

OBJECTIVE The authors sought to evaluate the accuracy of a novel telehealth-compatible diagnostic software system for identifying craniosynostosis within a newborn (< 1 year old) population. Agreement with gold standard craniometric diagnostics was also assessed. METHODS Cranial shape classification software accuracy was compared to that of blinded craniofacial specialists using a data set of open-source (n = 40) and retrospectively collected newborn orthogonal top-down cranial images, with or without additional facial views (n = 339), culled between April 1, 2008, and February 29, 2020. Based on image quality, midface visibility, and visibility of the cranial equator, 351 image sets were deemed acceptable. Accuracy, sensitivity, and specificity were calculated for the software versus specialist classification. Software agreement with optical craniometrics was assessed with intraclass correlation coefficients. RESULTS The cranial shape classification software had an accuracy of 93.3% (95% CI 86.8–98.8; p < 0.001), with a sensitivity of 92.0% and specificity of 94.3%. Intraclass correlation coefficients for measurements of the cephalic index and cranial vault asymmetry index compared to optical measurements were 0.95 (95% CI 0.84–0.98; p < 0.001) and 0.67 (95% CI 0.24–0.88; p = 0.003), respectively. CONCLUSIONS These results support the use of image processing–based neonatal cranial deformity classification software for remote screening of nonsyndromic craniosynostosis in a newborn population and as a substitute for optical scanner– or CT-based craniometrics. This work has implications that suggest the potential for the development of software for a mobile platform that would allow for screening by telemedicine or in a primary care setting.

Download Full-text