Improved detection of dry age-related macular degeneration from optical coherence tomography images using adaptive window based feature extraction and weighted ensemble based classification approach.

BACKGROUND: Dry Age-related macular degeneration (AMD), which affects the older population, can lead to blindness when left untreated. Preventing vision loss in elderly needs early identification. Dry-AMD diagnosis is still time-consuming and very subjective, depending on the ophthalmologist. Setting up a thorough eye-screening system to find Dry-AMD is a very difficult task. METHODOLOGY: This study aims to develop a weighted majority voting (WMV) ensemble-based prediction model to diagnose Dry-AMD. The WMV approach combines the predictions from base-classifiers and chooses the class with greatest vote based on assigned weights to each classifier. A novel feature extraction method is used along the retinal pigment epithelium (RPE) layer, with the number of windows calculated for each picture playing an important part in identifying Dry-AMD/normal images using the WMV methodology. Pre-processing using hybrid-median filter followed by scale-invariant feature transform based segmentation of RPE layer and curvature flattening of retina is employed to measure exact thickness of RPE layer. RESULT: The proposed model is trained on 70% of the OCT image database (OCTID) and evaluated on remaining OCTID and SD-OCT Noor dataset. Model has achieved accuracy of 96.15% and 96.94%, respectively. The suggested algorithm's effectiveness in Dry-AMD identification is demonstrated by comparison with alternative approaches. Even though the suggested model is only trained on the OCTID, it has performed well when tested on additional dataset. CONCLUSION: The suggested architecture can be used for quick eye-screening for early identification of Dry-AMD. The recommended method may be applied in real-time since it requires fewer complexity and learning-variables.

Improved detection accuracy of red lesions in retinal fundus images with superlearning approach.

BACKGROUND: Diabetic Retinopathy (DR) is a serious consequence of diabetes that can result to permanent vision loss for a person. Diabetes-related vision impairment can be significantly avoided with timely screening and treatment in its initial phase. The earliest and the most noticeable indications on the surface of the retina are micro-aneurysm and haemorrhage, which appear as dark patches. Therefore, the automatic detection of retinopathy begins with the identification of all these dark lesions. METHOD: In our study, we have developed a clinical knowledge based segmentation built on Early Treatment DR Study (ETDRS). ETDRS is a gold standard for identifying all red lesions using adaptive-thresholding approach followed by different pre-processing steps. The lesions are classified using super-learning approach to improve multi-class detection accuracy. Ensemble based super-learning approach finds optimal weights of base learners by minimizing the cross validated risk-function and it pledges the improved performance compared to base-learners predictions. For multi-class classification, a well informative feature-set based on colour, intensity, shape, size and texture, is developed. In this work, we have handled the data imbalance problem and compared the final accuracy with different synthetic data creation ratios. RESULT: The suggested approach uses publicly available resources to perform quantitative assessments at lesions-level. The overall accuracy of red lesion segregation is 93.5%, which has increased to 97.88% when data imbalance problem is taken care-off. CONCLUSION: The results of our system have achieved competitive performance compared with other modern approaches and handling of data imbalance further increases the performance of it.

Acoustic feature based unsupervised approach of heart sound event detection.

This paper represents an unsupervised approach to detect the positions of S1, S2 heart sound events in a Phonocardiogram (PCG) recording. Insufficiency of correctly annotated heart sound database drives us to investigate unsupervised techniques. Gammatone filter bank features are used to characterize the spectral pattern of fundamental heart sound events from noise contaminated PCG data. An unsupervised spectral clustering technique is employed for segmentation of S1/S2 and non-S1/S2 heart sound events. A Feature winning score is computed to identify the S1/S2 and non-S1/S2 frames. Finally, time based threshold is applied to detect the accurate positions of S1 and S2 heart sounds. The performance of spectral clustering is compared with other clustering methods. The proposed method offers a maximum F1-score of 98% and 92.5% for normal and abnormal PCG data respectively on 2016 PhysioNet/CinC challenge dataset. The heart sound annotation algorithm provided by PhysioNet has been used as the ground truth after hand correction.

A novel approach towards non-obstructive detection and classification of COPD using ECG derived respiration.

The alarming rate of mortality and disability due to Chronic Obstructive Pulmonary Disease (COPD) has become a serious health concern worldwide. The progressive nature of this disease makes it inevitable to detect this disease in its early stages, leads to a greater demand for developing non-obstructive and reliable technology for COPD detection. The use of highly patient-effort dependent, time-consuming, and expensive methods are some major inherent limitations of previous techniques. Lack of knowledge about the disease and inadequacy of proper diagnostic tool for early detection of COPD is another reason behind the 3rd leading cause of death worldwide. For this reason, this study aims to explore the utility of ECG Derived Respiration (EDR) for classification between COPD patients and normal healthy subjects as EDR can be easily extracted from ECG. ECG and respiration signals collected from 30 normal and 30 COPD subjects were analysed. Error calculation and statistical analysis were performed to observe the similarity between original respiration and EDR signal. The morphological pattern changes of respiration and EDR signals were analysed and three different features were extracted from those. Classification was performed by different classifiers employing Decision Tree, Linear Discriminant Analysis (LDA), Support Vector Machine (SVM) and K-Nearest Neighbor (KNN). Apart from obtaining comparable classification performance it was seen that EDR has better potential than the original respiration signal for classification of COPD from normal population.

Arduino-based noise robust online heart-rate detection.

This paper introduces a noise robust real time heart rate detection system from electrocardiogram (ECG) data. An online data acquisition system is developed to collect ECG signals from human subjects. Heart rate is detected using window-based autocorrelation peak localisation technique. A low-cost Arduino UNO board is used to implement the complete automated process. The performance of the system is compared with PC-based heart rate detection technique. Accuracy of the system is validated through simulated noisy ECG data with various levels of signal to noise ratio (SNR). The mean percentage error of detected heart rate is found to be 0.72% for the noisy database with five different noise levels.

Significance of Exhaled Breath Test in Clinical Diagnosis: A Special Focus on the Detection of Diabetes Mellitus.

Analysis of volatile organic compounds (VOCs) emanating from human exhaled breath can provide deep insight into the status of various biochemical processes in the human body. VOCs can serve as potential biomarkers of physiological and pathophysiological conditions related to several diseases. Breath VOC analysis, a noninvasive and quick biomonitoring approach, also has potential for the early detection and progress monitoring of several diseases. This paper gives an overview of the major VOCs present in human exhaled breath, possible biochemical pathways of breath VOC generation, diagnostic importance of their analysis, and analytical techniques used in the breath test. Breath analysis relating to diabetes mellitus and its characteristic breath biomarkers is focused on. Finally, some challenges and limitations of the breath test are discussed.

Empirical mode decomposition based ECG enhancement and QRS detection.

In this paper an Empirical Mode Decomposition (EMD) based ECG signal enhancement and QRS detection algorithm is proposed. Being a non-invasive measurement, ECG is prone to various high and low frequency noises causing baseline wander and power line interference, which act as a source of error in QRS and other feature extraction. EMD is a fully adaptive signal decomposition technique that generates Intrinsic Mode Functions (IMF) as decomposition output. Here, first baseline wander is corrected by selective reconstruction based slope minimization technique from IMFs and then high frequency noise is removed by eliminating a noisy set of lower order IMFs with a statistical peak correction as high frequency noise elimination is accompanied by peak deformation of sharp characteristic waves. Then a set of IMFs are selected that represents QRS region and a nonlinear transformation is done for QRS enhancement. This improves detection accuracy, which is represented in the result section. Thus in this method a single fold processing of each signal is required unlike other conventional techniques.