Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual Conference
-
Conf Proc IEEE Eng Med Biol Soc · Aug 2016
Sleep stage classification by non-contact vital signs indices using Doppler radar sensors.
Disturbed sleep has become more common in recent years. To improve the quality of sleep, undergoing sleep observation has gained interest as a means to resolve possible problems. In this paper, we evaluate a non-restrictive and non-contact method for classifying real-time sleep stages and report on its potential applications. ⋯ Classification accuracy was 66.4% for simply identifying wake and sleep, 57.1% for three stages (wake, REM, and NREM) and 34% for four stages (wake, REM, LIGHT, and DEEP). This is a novel system for measuring HRs, HRV, body movements, and respiratory intervals and for measuring high sensitivity pulse waves using two radar signals. It simplifies measurement of sleep stages and may be employed at nursing care facilities or by the general public to improve sleep quality.
-
Conf Proc IEEE Eng Med Biol Soc · Aug 2016
Support vectors machine classification of surface electromyography for non-invasive naturally controlled hand prostheses.
The scientific researches in human rehabilitation techniques have continually evolved to offer again the mobility and freedom lost to disability. Many systems managed by myoelectric signals intended to mimic the movement of the human arm still have results considered partial, which makes it subject of many researches. The use of Natural Interfaces Signal Processing methods makes possible to design systems capable of offering prosthesis in a more natural and intuitive way. ⋯ The Root Mean Square (RMS) value feature is extracted of the signal and it serves as input data for the classification with SVM. The classification stage used three types of kernel functions (linear, polynomial, radial basis) for comparison of the results. The average accuracy reached for the classification of seventeen distinct movements of 83.7% was achieved using the SVM linear classifier, 80.8% was achieved using the SVM polynomial classifier and 85.1% was achieved using the SVM radial basis classifier.
-
Conf Proc IEEE Eng Med Biol Soc · Aug 2016
Artery/vein classification of retinal blood vessels using feature selection.
Automated classification of retinal vessels in fundus images is the first step towards measurement of retinal characteristics that can be used to screen and diagnose vessel abnormalities for cardiovascular and retinal disorders. This paper presents a novel approach to vessel classification to compute the artery/vein ratio (AVR) for all blood vessel segments in the fundus image. The features extracted are then subjected to a selection procedure using Random Forests (RF) where the features that contribute most to classification accuracy are chosen as input to a polynomial kernel Support Vector Machine (SVM) classifier. ⋯ The SVM is then subjected to one time training using 10-fold cross validation on images randomly selected from the VICAVR dataset before testing on an independent test dataset, derived from the same database. An Area Under the ROC Curve (AUC) of 97.2% was obtained on an average of 100 runs of the algorithm. The proposed algorithm is robust due to the feature selection procedure, and it is possible to get similar accuracies across many datasets.
-
Conf Proc IEEE Eng Med Biol Soc · Aug 2016
A cost-effective, non-invasive system for pressure monitoring during epidural needle insertion: Design, development and bench tests.
Epidural blockade procedures have gained large acceptance during last decades. However, the insertion of the needle during epidural blockade procedures is challenging, and there is an increasing alarming risk in accidental dural puncture. One of the most popular approaches to minimize the mentioned risk is to detect the epidural space on the base of the loss of resistance (LOR) during the epidural needle insertion. ⋯ Hence, on the base of a peculiar algorithm, the system automatically detects LOR providing visual and acoustic feedbacks to the operator improving the safety of the procedure. Experiments have been performed to characterize the measurement device and to validate the whole system. Notice that the proposed solution is able to perform an effective detection of the LOR.
-
Conf Proc IEEE Eng Med Biol Soc · Aug 2016
Extraction of medically interpretable features for classification of malignancy in breast thermography.
Thermography, with high-resolution cameras, is being re-investigated as a possible breast cancer screening imaging modality, as it does not have the harmful radiation effects of mammography. This paper focuses on automatic extraction of medically interpretable non-vascular thermal features. We design these features to differentiate malignancy from different non-malignancy conditions, including hormone sensitive tissues and certain benign conditions, which have an increased thermal response. ⋯ On a dataset of around 78 subjects with cancer and 187 subjects without cancer, that have some benign diseases and conditions with thermal responses, we are able to get around 99% specificity while having 100% sensitivity. This indicates a potential break-through in thermographic screening for breast cancer. This shows promise for undertaking a comparison to mammography with larger numbers of subjects with more data variations.