Camera-Based Peripheral Edema Measurement Using Machine Learning

Peripheral edema is the most common symptom of heart failure. Reliable measurement of edema and continuous monitoring of trends provide critical clinical information and can be used for averting episodes of acute decompensation and hospitalizations. Based on the edema pitting-test, new videobased methods for measurement of peripheral edema stages are presented. The methods use videos of skin during the pittingtest, which are processed by machine learning or deep learning techniques to provide classification into one of four edema stages. The proposed methods are implemented and evaluated on videos taken on edema simulators. Variations of the proposed models applied to edema simulators yield classification accuracies in the range between 87% and 98%.

[1]  E. Stranden.  A comparison between surface measurements and water displacement volumetry for the quantification of leg edema. , 1981, Journal of the Oslo city hospitals.

[2]  M. Fornage,et al.  Heart Disease and Stroke Statistics—2017 Update: A Report From the American Heart Association , 2017, Circulation.

[3]  B. Riegel,et al.  Self care in patients with chronic heart failure , 2011, Nature Reviews Cardiology.

[4]  T. Higashi,et al.  Validity of a New Quantitative Evaluation Method that Uses the Depth of the Surface Imprint as an Indicator for Pitting Edema , 2017, PloS one.

[5]  Takumi Yamamoto,et al.  Localized Leg Volume Index: A New Method for Body Type–Corrected Evaluation of Localized Leg Lymphedematous Volume Change , 2018, Annals of plastic surgery.

[6]  T. Miyati,et al.  Objective assessment of leg edema using ultrasonography with a gel pad , 2017, PloS one.

[7]  Geoffrey E. Hinton,et al.  Visualizing Data using t-SNE , 2008 .

[8]  Fei-Fei Li,et al.  ImageNet: A large-scale hierarchical image database , 2009, 2009 IEEE Conference on Computer Vision and Pattern Recognition.

[9]  Takumi Yamamoto,et al.  Localized Arm Volume Index: A New Method for Body Type–Corrected Evaluation of Localized Arm Lymphedematous Volume Change , 2017, Annals of plastic surgery.

[10]  D. Mozaffarian,et al.  Heart disease and stroke statistics--2010 update: a report from the American Heart Association. , 2010, Circulation.

[11]  I. Piña,et al.  Forecasting the Impact of Heart Failure in the United States: A Policy Statement From the American Heart Association , 2013, Circulation. Heart failure.

[12]  V. Roger.  Epidemiology of Heart Failure , 2013, Circulation research.

[13]  Syed Muhammad Anwar,et al.  Deep Learning in Medical Image Analysis , 2017 .

[14]  Simon Lucey,et al.  Why do linear SVMs trained on HOG features perform so well? , 2014, ArXiv.

[15]  Krista A. Ehinger,et al.  SUN database: Large-scale scene recognition from abbey to zoo , 2010, 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[16]  H. Folgering,et al.  Volumetric measurements of peripheral oedema in clinical conditions. , 2000, Clinical physiology.

[17]  Hassan Ghasemzadeh,et al.  SmartSock: a wearable platform for context-aware assessment of ankle edema. , 2016, Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual Conference.

[18]  Nitish Srivastava,et al.  Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..

[19]  Andrew Zisserman,et al.  Two-Stream Convolutional Networks for Action Recognition in Videos , 2014, NIPS.

[20]  Fei-Fei Li,et al.  Large-Scale Video Classification with Convolutional Neural Networks , 2014, 2014 IEEE Conference on Computer Vision and Pattern Recognition.

[21]  Oscar Déniz-Suárez,et al.  Face recognition using Histograms of Oriented Gradients , 2011, Pattern Recognit. Lett..

[22]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[23]  F. Massari,et al.  Accuracy of bioimpedance vector analysis and brain natriuretic peptide in detection of peripheral edema in acute and chronic heart failure. , 2016, Heart & lung : the journal of critical care.

[24]  Tao Mei,et al.  Action Recognition by Learning Deep Multi-Granular Spatio-Temporal Video Representation , 2016, ICMR.

[25]  Harlan M. Krumholz,et al.  Recent National Trends in Readmission Rates After Heart Failure Hospitalization , 2010, Circulation. Heart failure.

[26]  Bill Triggs,et al.  Histograms of oriented gradients for human detection , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05).