Submit to Sensors Review for Sensors Propose a Special Issue

Journal Browser

► Journal Browser

Machine Learning for Sensing and Healthcare

Special Issue Editors
Special Issue Information
Keywords
Published Papers

A special issue of Sensors (ISSN 1424-8220). This special issue belongs to the section "Chemical Sensors".

Deadline for manuscript submissions: closed (30 May 2019) | Viewed by 27717

Share This Special Issue

Special Issue Editor

Dr. Yogeswaran Umasankar

E-Mail Website
Guest Editor

Biomolecular Sciences Institute, Florida International University, 10555 West Flagler Street, Miami, FL 33174, USA
Interests: point-of-care systems; wearable sensor platforms; longitudinal health monitoring; rapid disease diagnostics; real-time health monitoring
Special Issues, Collections and Topics in MDPI journals

Special Issue Information

Dear Colleagues,

This Special Issue focuses on the significance of machine learning as a complementary tool for sensing and healthcare applications. We invite high-quality research papers resolving existing limitations in sensing by a simple algorithm or through a complex set of machine learning tools. Articles may address advancements in pattern recognition techniques, intelligent algorithms, automated data analysis, sensor fusion techniques, and smart sensing, which resolves existing issues in sensing and diagnostics. To provide the readers a focused perspective, articles with machine learning in healthcare should be related to sensing and diagnostics, but not limited to data mining and predictive analytics. Contents may also focus on extensive studies of sensing with the Ecological Momentary Assessment (EMA). Review papers are also welcome, as they provide readers with a comprehensive view of this prospective field.

Dr. Yogeswaran Umasankar
Guest Editor

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Sensors is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2600 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Keywords

biosensors
point-of-care
longitudinal monitoring
clinical diagnosis
intelligent sensors
sensors and big data
machine learning

Published Papers (5 papers)

Download All Papers

Research

15 pages, 3627 KiB

Open AccessArticle

Machine Learning Techniques for Chemical Identification Using Cyclic Square Wave Voltammetry

by Scott N. Dean, Lisa C. Shriver-Lake, David A. Stenger, Jeffrey S. Erickson, Joel P. Golden and Scott A. Trammell

Sensors 2019, 19(10), 2392; https://doi.org/10.3390/s19102392 - 25 May 2019

Cited by 31 | Viewed by 4541

Abstract

Electroanalytical techniques are useful for detection and identification because the instrumentation is simple and can support a wide variety of assays. One example is cyclic square wave voltammetry (CSWV), a practical detection technique for different classes of compounds including explosives, herbicides/pesticides, industrial compounds, and heavy metals. A key barrier to the widespread application of CSWV for chemical identification is the necessity of a high performance, generalizable classification algorithm. Here, machine and deep learning models were developed for classifying samples based on voltammograms alone. The highest performing models were Long Short-Term Memory (LSTM) and Fully Convolutional Networks (FCNs), depending on the dataset against which performance was assessed. When compared to other algorithms, previously used for classification of CSWV and other similar data, our LSTM and FCN-based neural networks achieve higher sensitivity and specificity with the area under the curve values from receiver operating characteristic (ROC) analyses greater than 0.99 for several datasets. Class activation maps were paired with CSWV scans to assist in understanding the decision-making process of the networks, and their ability to utilize this information was examined. The best-performing models were then successfully applied to new or holdout experimental data. An automated method for processing CSWV data, training machine learning models, and evaluating their prediction performance is described, and the tools generated provide support for the identification of compounds using CSWV from samples in the field. Full article

(This article belongs to the Special Issue Machine Learning for Sensing and Healthcare)

► Show Figures

Graphical abstract

21 pages, 1319 KiB

Open AccessArticle

Sparse ECG Denoising with Generalized Minimax Concave Penalty

by Zhongyi Jin, Anming Dong, Minglei Shu and Yinglong Wang

Sensors 2019, 19(7), 1718; https://doi.org/10.3390/s19071718 - 10 Apr 2019

Cited by 23 | Viewed by 4256

Abstract

The electrocardiogram (ECG) is an important diagnostic tool for cardiovascular diseases. However, ECG signals are susceptible to noise, which may degenerate waveform and cause misdiagnosis. In this paper, the ECG noise reduction techniques based on sparse recovery are investigated. A novel sparse ECG denoising framework combining low-pass filtering and sparsity recovery is proposed. Two sparsity recovery algorithms are developed based on the traditional

ℓ_{1}

-norm penalty and the novel generalized minimax concave (GMC) penalty, respectively. Compared with the

ℓ_{1}

-norm penalty, the non-differentiable non-convex GMC penalty has the potential to strongly promote sparsity while maintaining the convexity of the cost function. Moreover, the GMC punishes large values less severely than

ℓ_{1}

-norm, which is utilized to overcome the drawback of underestimating the high-amplitude components for the

ℓ_{1}

-norm penalty. The proposed methods are evaluated on ECG signals from the MIT-BIH Arrhythmia database. The results show that underestimating problem is overcome by the proposed GMC-based method. The GMC-based method shows significant improvement with respect to the average of output signal-to-noise ratio improvement (

S N R_{i m p}

), the average of root mean square error (RMSE) and the percent root mean square difference (PRD) over almost any given SNR compared with the classical methods, thus providing promising approaches for ECG denoising. Full article

(This article belongs to the Special Issue Machine Learning for Sensing and Healthcare)

► Show Figures

Figure 1

15 pages, 3415 KiB

Open AccessArticle

Classifying Parkinson’s Disease Based on Acoustic Measures Using Artificial Neural Networks

by Lucijano Berus, Simon Klancnik, Miran Brezocnik and Mirko Ficko

Sensors 2019, 19(1), 16; https://doi.org/10.3390/s19010016 - 20 Dec 2018

Cited by 66 | Viewed by 4785

Abstract

In recent years, neural networks have become very popular in all kinds of prediction problems. In this paper, multiple feed-forward artificial neural networks (ANNs) with various configurations are used in the prediction of Parkinson’s disease (PD) of tested individuals, based on extracted features from 26 different voice samples per individual. Results are validated via the leave-one-subject-out (LOSO) scheme. Few feature selection procedures based on Pearson’s correlation coefficient, Kendall’s correlation coefficient, principal component analysis, and self-organizing maps, have been used for boosting the performance of algorithms and for data reduction. The best test accuracy result has been achieved with Kendall’s correlation coefficient-based feature selection, and the most relevant voice samples are recognized. Multiple ANNs have proven to be the best classification technique for diagnosis of PD without usage of the feature selection procedure (on raw data). Finally, a neural network is fine-tuned, and a test accuracy of 86.47% was achieved. Full article

(This article belongs to the Special Issue Machine Learning for Sensing and Healthcare)

► Show Figures

Figure 1

13 pages, 4128 KiB

Open AccessArticle

A Digital Shade-Matching Device for Dental Color Determination Using the Support Vector Machine Algorithm

by Minah Kim, Byungyeon Kim, Byungjun Park, Minsuk Lee, Youngjae Won, Choul-Young Kim and Seungrag Lee

Sensors 2018, 18(9), 3051; https://doi.org/10.3390/s18093051 - 12 Sep 2018

Cited by 19 | Viewed by 8611

Abstract

In this study, we developed a digital shade-matching device for dental color determination using the support vector machine (SVM) algorithm. Shade-matching was performed using shade tabs. For the hardware, the typically used intraoral camera was modified to apply the cross-polarization scheme and block the light from outside, which can lead to shade-matching errors. For reliable experiments, a precise robot arm with ±0.1 mm position repeatability and a specially designed jig to fix the position of the VITA 3D-master (3D) shade tabs were used. For consistent color performance, color calibration was performed with five standard colors having color values as the mean color values of the five shade tabs of the 3D. By using the SVM algorithm, hyperplanes and support vectors for 3D shade tabs were obtained with a database organized using five developed devices. Subsequently, shade matching was performed by measuring 3D shade tabs, as opposed to real teeth, with three additional devices. On average, more than 90% matching accuracy and a less than 1% failure rate were achieved with all devices for 10 measurements. In addition, we compared the classification algorithm with other classification algorithms, such as logistic regression, random forest, and k-nearest neighbors, using the leave-pair-out cross-validation method to verify the classification performance of the SVM algorithm. Our proposed scheme can be an optimum solution for the quantitative measurement of tooth color with high accuracy. Full article

(This article belongs to the Special Issue Machine Learning for Sensing and Healthcare)

► Show Figures

Figure 1

15 pages, 5185 KiB

Open AccessArticle

Clutch Pedal Sensorization and Evaluation of the Main Parameters Related to Driver Posture

by Ester Olmeda, Sergio Fuentes del Toro, María Garrosa, Jonatan Pajares Redondo and Vicente Díaz

Sensors 2018, 18(9), 2797; https://doi.org/10.3390/s18092797 - 24 Aug 2018

Cited by 4 | Viewed by 4739

Abstract

An improper decision for the design, selection and adjustment of the components needed to control a vehicle could generate negative effects and discomfort to the driver, where pedals play a very important role. The aim of the study is to provide a first approach to develop an embedded monitoring device in order to evaluate the posture of the driver, the influence of the clutch pedal and to advise about the possible risk. With that purpose in mind, a testbed was designed and two different sets of tests were carried out. The first test collected information about the volunteers who were part of the experiment, like the applied force on the clutch pedal or the body measurements. The second test was carried out to provide new insight into this matter. One of the more significant findings to emerge from this study is that the force applied on the clutch pedal provides enough information to determine correct driver posture. For this reason, a system composed of a pedal force sensor and an acquisition/processing system can fulfil the requirements to create a healthcare system focused on driver posture. Full article

(This article belongs to the Special Issue Machine Learning for Sensing and Healthcare)

► Show Figures

Figure 1

Journal Menu

Journal Browser

Machine Learning for Sensing and Healthcare

Share This Special Issue

Special Issue Editor

Special Issue Information

Keywords

Published Papers (5 papers)

Research

Further Information

Guidelines

MDPI Initiatives

Follow MDPI