The determination of the tumor's extent is a major challenge in brain tumour treatment planning and measurement. Non-invasive magnetic resonance imaging (MRI) has evolved as a first-line diagnostic... Show More

The determination of the tumor's extent is a major challenge in brain tumour treatment planning and measurement. Non-invasive magnetic resonance imaging (MRI) has evolved as a first-line diagnostic tool for brain malignancies without the use of ionising radiation. Manually segmenting the extent of a brain tumour from 3D MRI volumes is a time-consuming process that significantly relies on the experience of the operator. As a result, we suggested a modified UNet structure based on residual networks that use periodic shuffling at the encoder region of the original UNet and sub-pixel convolution at the decoder section in this research. The proposed UNet was tested on BraTS Challenge 2017 with high-grade glioma (HGG). The model was tested on BraTS 2017 and 2018 datasets. Tumour core (TC), whole tumour (WT), and enhancing core (EC) were the three major labels to be segmented. The test results shown that proposed UNet outperform the existing techniques.

A BCI theoretical idea is to construct an output feature or task for a user using brain signals. These signals are then transmitted to the machine where the required task is performed. In this... Show More

A BCI theoretical idea is to construct an output feature or task for a user using brain signals. These signals are then transmitted to the machine where the required task is performed. In this work, we present a mental task classification model that focuses on the notion of transfer learning and addresses the issues of data scarcity, choice of model selection, and low-performance measure. To decide the optimal network for feature extraction, we used five different pre-trained networks including VGG16, VGG19, ResNet101, ResNet18, and ResNet50. For the classification, the suggested model experiments with three baseline classifiers namely support vector machine, decision tree, and random forest. The model's experimental evaluation is done on the publicly available Keirn and Aunon databases. From the experiment, it is observed that features extracted from the transfer learning models help to identify the five different mental tasks efficiently. The highest average accuracy of 81.25% is attained on ResNet50 based features with a random forest classifier.

Good health of bones is of the utmost importance to human beings. Smart materials like lead zirconate titanate (PZT) patches are small in size and carry less weight, which makes them most apt for... Show More

Good health of bones is of the utmost importance to human beings. Smart materials like lead zirconate titanate (PZT) patches are small in size and carry less weight, which makes them most apt for biomedical structural health monitoring (BSHM). In the past, focus on the development of low-cost non-invasive techniques for real-time monitoring of critical bones has been undertaken as an alternative to current diagnosis techniques such as dual x-ray absorptiometry (DEXA), which is not portable and emits radiations. This paper presents a study to evaluate a previously developed non-bonded piezo sensor (NBPS)-based diagnostic technique for non-invasive detection of osteoporosis, in the framework of the electro-mechanical impedance (EMI) technique. As part of the study, the experimental trials in the paper are performed for comparing DEXA and bone electro-mechano gram (BEMG) on healthy subjects as well as those with osteoporosis. It was found that BEMG identified structural system for healthy and osteoporotic subjects were quite different leading to a new technique to identify osteoporosis.

literature shows that Blink Rate, Blink Duration, and Percentage of Eye Closure (PERCLOS) are the indicators of drowsiness, but the quantification of these parameters, inter-individual differences... Show More

literature shows that Blink Rate, Blink Duration, and Percentage of Eye Closure (PERCLOS) are the indicators of drowsiness, but the quantification of these parameters, inter-individual differences, and scientific or the physiological validation of the results have not been addressed. This study attempts to resolve these problems by the systems dynamic approach by modelling the oculomotor system. Autoregressive model of the EOG blink signatures during active and drowsy states are used to approximate and model the system. The impulse response of the active blink signal shows under damped response with the damping ratio of 0.61-0.75, (p<0.0005), and Drowsy blink signal shows a critically damped behavior with the damping ratio of 1, (p<0.0005). It is Clinically correlated that the continuous bombarding of the neuronal impulses from the brain acts as the stimulus for the blink, Hence during the drowsy phase, the response of the Oculomotor system is sluggish (Damping Ratio is high) thus causing increased Blink duration.

This paper proposes a new approach for non-invasive cuff-less arterial Blood Pressure (BP) estimation using pulse transit time (PTT). The ECG and PPG signals were acquired at a sampling rate of... Show More

This paper proposes a new approach for non-invasive cuff-less arterial Blood Pressure (BP) estimation using pulse transit time (PTT). The ECG and PPG signals were acquired at a sampling rate of 500Hz. Standard cuff based Sphygmomanometer used to take reference BP and heart rate simultaneously. The hardware for the acquiring the ECG and PPG signals were designed and fabricated and were made and study was carried out with 60 subject during various activities. The objective of this work is to estimate the Systolic BP and Diastolic BP using PTT techniques and to apply regression analysis using machine learning methods for estimating the BP, compare the results with recording simultaneously carried out using the standard devices. The proposed work concludes that AdaBoost algorithm has highest accuracy in estimating systolic and diastolic BP values. The readings obtained are in accordance with the AHA standards and are in acceptable limits and can be used for measuring BP in wearable devices.

Patients suffering from diabetes have to bear several other disorders due to this. Diabetic Retinopathy is one such disorder which affects diabetic patients. This disorder affects the patient’s eye... Show More

Patients suffering from diabetes have to bear several other disorders due to this. Diabetic Retinopathy is one such disorder which affects diabetic patients. This disorder affects the patient’s eye leading to permanent blindness if left untreated. Another disorder is exudates in which lipid residues leak out from damaged capillaries. It appears as yellow flecks. Hard exudates can lead to life threatening disorders. Detecting Hard exudates help the Ophthalmologist to diagnose the severity of the patient’s condition and in turn help in better medication. This paper presents a method to adjust the contrast of the image which in turn helps in detecting the hard exudates which can be used for further processing. In this work, initially Otsu algorithm is applied and then compared with Machine Learning techniques due to the disadvantage of Otsu.

One of the most common neurodegenerative disorders of the present age is Parkinson’s Disease or Parkinsonism. To estimate its advancement in the patient, huge amounts of data are being collected and... Show More

One of the most common neurodegenerative disorders of the present age is Parkinson’s Disease or Parkinsonism. To estimate its advancement in the patient, huge amounts of data are being collected and studied to draw out inferences. The types of data generally studied towards that end are vocal data, body movement data, eye movement data, handwriting and drawing patterns, etc. In this work, the use of a Deep Neural Network has been proposed which can predict the Unified Parkinson's Disease Rating Scale (UPDRS) both motor and total by studying vocal data from UCI Machine Learning Repository. Both 2 layered as well as 3 layered networks were studied and it was found that the performance of 3-layer Deep Neural Network having 10, 20, 10 neurons in different layers was found to be the best with an accuracy of 97% and 99.62% for motor UPDRS and total UPDRS respectively. The other three parameters MSE, MAE and RMSE also showed improvement in the 3 layered model as compared to the 2 layered model.