Kwan-Hoong Ng

Automated diagnosis of epileptic EEG using entropies

Abstract Epilepsy is a neurological disorder characterized by the presence of recurring seizures. Like many other neurological disorders, epilepsy can be assessed by the electroencephalogram (EEG). The EEG signal is highly non-linear and non-stationary, and hence, it is difficult to characterize and interpret it. However, it is a well-established clinical technique with low associated costs. In this work, we propose a methodology for the automatic detection of normal , pre-ictal , and ictal conditions from recorded EEG signals. Four entropy features namely Approximate Entropy ( ApEn ), Sample Entropy ( SampEn ), Phase Entropy 1 ( S1 ), and Phase Entropy 2 ( S2 ) were extracted from the collected EEG signals. These features were fed to seven different classifiers: Fuzzy Sugeno Classifier (FSC), Support Vector Machine (SVM), K-Nearest Neighbour (KNN), Probabilistic Neural Network (PNN), Decision Tree (DT), Gaussian Mixture Model (GMM), and Naive Bayes Classifier (NBC). Our results show that the Fuzzy classifier was able to differentiate the three classes with a high accuracy of 98.1%. Overall, compared to previous techniques, our proposed strategy is more suitable for diagnosis of epilepsy with higher accuracy.

Algorithms for the Automated Detection of Diabetic Retinopathy Using Digital Fundus Images: A Review

Diabetes is a chronic end organ disease that occurs when the pancreas does not secrete enough insulin or the body is unable to process it properly. Over time, diabetes affects the circulatory system, including that of the retina. Diabetic retinopathy is a medical condition where the retina is damaged because fluid leaks from blood vessels into the retina. Ophthalmologists recognize diabetic retinopathy based on features, such as blood vessel area, exudes, hemorrhages, microaneurysms and texture. In this paper we review algorithms used for the extraction of these features from digital fundus images. Furthermore, we discuss systems that use these features to classify individual fundus images. The classifications efficiency of different DR systems is discussed. Most of the reported systems are highly optimized with respect to the analyzed fundus images, therefore a generalization of individual results is difficult. However, this review shows that the classification results improved has improved recently, and it is getting closer to the classification capabilities of human ophthalmologists.

Coronary CT angiography: current status and continuing challenges.

Coronary CT angiography has been increasingly used in the diagnosis of coronary artery disease owing to rapid technological developments, which are reflected in the improved spatial and temporal resolution of the images. High diagnostic accuracy has been achieved with multislice CT scanners (64 slice and higher), and in selected patients coronary CT angiography is regarded as a reliable alternative to invasive coronary angiography. With high-quality coronary CT imaging increasingly being performed, patients can benefit from an imaging modality that provides a rapid and accurate diagnosis while avoiding an invasive procedure. Despite the tremendous contributions of coronary CT angiography to cardiac imaging, study results reported in the literature should be interpreted with caution as there are some limitations existing within the study design or related to patient risk factors. In addition, some attention must be given to the potential health risks associated with the ionising radiation received during cardiac CT examinations. Radiation dose associated with coronary CT angiography has raised serious concerns in the literature, as the risk of developing malignancy is not negligible. Various dose-saving strategies have been implemented, with some of the strategies resulting in significant dose reduction. The aim of this review is to present an overview of the role of coronary CT angiography on cardiac imaging, with focus on coronary artery disease in terms of the diagnostic and prognostic value of coronary CT angiography. Various approaches for dose reduction commonly recommended in the literature are discussed. Limitations of coronary CT angiography are identified. Finally, future directions and challenges with the use of coronary CT angiography are highlighted.

Prospective versus retrospective ECG-gated multislice CT coronary angiography: a systematic review of radiation dose and diagnostic accuracy.

PURPOSE To perform a systematic review of the radiation dose and diagnostic accuracy of prospective versus retrospective ECG-gated multislice CT coronary angiography. MATERIALS AND METHODS A search of Pubmed/Medline and Sciencedirect databases for English literature was performed to identify studies comparing prospective and retrospective ECG-gated multislice CT angiography in the diagnosis of coronary artery disease. Effective dose, dose length product, image quality and diagnostic value were compared between two groups of studies. RESULTS 22 studies were included for analysis. The mean effective dose of prospective ECG-gated scans was 4.5 mSv (95% CI: 3.6, 5.3 mSv), which is significantly lower than that of retrospective scans, which is 13.8 mSv (95% CI: 11.5, 16.0 mSv) (p<0.001). The mean dose length product was 225 mGycm (95% CI: 188, 262 mGycm) and 822 mGycm (95% CI: 630, 1013 mGycm) for the prospective and retrospective ECG-gated scans, respectively, indicating a statistically significant difference between these two protocols (p<0.0001). The mean sensitivity and specificity of multislice CT angiography in the diagnosis of coronary artery disease was 97.7% (95% CI: 93.7%, 100%) and 92.1% (95% CI: 87.2%, 97%) for prospective ECG-gated scans; 95.2% (95% CI: 91%, 99.5%) and 94.4% (95% CI: 88.5%, 100%) for retrospective ECG-gated scans, respectively, with no significant difference for sensitivity but significant difference for specificity (p=0.047). CONCLUSION Multislice CT coronary angiography with prospective ECG-gating leads to a significant reduction of radiation dose when compared to that of retrospective ECG-gating, while offering comparable image quality and diagnostic value.

An Integrated Index for the Identification of Diabetic Retinopathy Stages Using Texture Parameters

Diabetes is a condition of increase in the blood sugar level higher than the normal range. Prolonged diabetes damages the small blood vessels in the retina resulting in diabetic retinopathy (DR). DR progresses with time without any noticeable symptoms until the damage has occurred. Hence, it is very beneficial to have the regular cost effective eye screening for the diabetes subjects. This paper documents a system that can be used for automatic mass screenings of diabetic retinopathy. Four classes are identified: normal retina, non-proliferative diabetic retinopathy (NPDR), proliferative diabetic retinopathy (PDR), and macular edema (ME). We used 238 retinal fundus images in our analysis. Five different texture features such as homogeneity, correlation, short run emphasis, long run emphasis, and run percentage were extracted from the digital fundus images. These features were fed into a support vector machine classifier (SVM) for automatic classification. SVM classifier of different kernel functions (linear, radial basis function, polynomial of order 1, 2, and 3) was studied. Receiver operation characteristics (ROC) curves were plotted to select the best classifier. Our proposed system is able to identify the unknown class with an accuracy of 85.2%, and sensitivity, specificity, and area under curve (AUC) of 98.9%, 89.5%, and 0.972 respectively using SVM classifier with polynomial kernel of order 3. We have also proposed a new integrated DR index (IDRI) using different features, which is able to identify the different classes with 100% accuracy.

Differentiation of elemental composition of normal and malignant breast tissue by instrumental neutron activation analysis

Multi-elemental quantitative analyses of 15 paired samples of normal and malignant human breast tissue by instrumental neutron activation analysis are reported. The elements, Al, Br, Ca, Cl, Co, Cs, Fe, K, Na, Rb, Zn were detected. Significantly elevated concentration levels were found for Al, Br, Ca, Cl, Cs, K, Na, Zn in malignant compared to normal tissue. Although the role of elemental composition in breast cancer is unclear, this finding may be of importance as another parameter for differentiating normal from malignant tissue.

A systematic review of radiation dose associated with different generations of multidetector CT coronary angiography.

The purpose of this paper is to perform a systematic review on radiation dose reduction in coronary computed tomography (CT) angiography that is done using different generations of multidetector CT (MDCT) scanners ranging from four‐slice to 320‐slice CTs, and have different dose‐saving techniques. The method followed was to search for references on coronary CT angiography (CTA) that had been published in English between 1998 and February 2011. The effective radiation dose reported in each study based on different generations of MDCT scanners was analysed and compared between the types of scanners, gender, exposure factors and scanning protocols. Sixty‐six studies were eligible for inclusion in this analysis. The mean effective dose (ED) for MDCT angiography with retrospective electrocardiogram (ECG) gating without use of any dose‐saving protocol was 6.0 ± 2.8, 10.4 ± 4.90 and 11.8 ± 5.9 mSv for four‐slice, 16‐slice and 64‐slice CTs, respectively. More dose‐saving strategies were applied in recent CT generations including prospective ECG‐gating protocols, application of lower tube voltage and tube current modulation to achieve a noteworthy dose reduction. Prospective ECG‐gating protocol was increasingly used in 64, 125, 256 and 320 slices with corresponding ED of 4.1 ± 1.7, 3.6 ± 0.4, 3.0 ± 1.9 and 7.6 ± 1.6 mSv, respectively. Lower tube voltage and tube current modulation were widely applied in 64‐slice CT and resulted in significant dose reduction (P < 0.05). This analysis has shown that dose‐saving strategies can substantially reduce the radiation dose in CT coronary angiography. The fact that more and more clinicians are opting for dose‐saving strategies in CT coronary angiography indicates an increased awareness of risks associated with high radiation doses among them.

Differentiation between benign and malignant breast lesions using quantitative diffusion-weighted sequence on 3 T MRI

AIM To investigate the capability and diagnostic accuracy of diffusion-weighted imaging (DWI) in differentiating benign from malignant breast lesions using 3 T magnetic resonance imaging (MRI). MATERIALS AND METHODS Women with suspicious or indeterminate breast lesions detected at MRI, mammogram and/or ultrasound were recruited for dynamic contrast-enhanced (DCE)-MRI and DWI prior to their biopsy. Image fusion of DCE-MRI with apparent diffusion coefficient (ADC) map was utilized to select the region of interest (ROI) for ADC calculation in the area that showed the most avid enhancement. DWI was performed using two sets of b-values at 500 and 1000 s/mm(2), respectively. RESULTS Fifty women were recruited and the final analysis comprised 44 breast lesions, 31 of which were malignant and 13 were benign. Significant results were obtained between ADC values of benign and malignant lesions (p < 0.001). The cut-off ADC values for benign and malignant lesions were 1.21 × 10(-3) mm(2)/s for b = 500 s/mm(2) and 1.22 × 10(-3) mm(2)/s for b = 1000 s/mm(2), respectively. The sensitivity of DCE-MRI alone was 100% with a specificity of 66.7%. When DCE-MRI was combined with b = 1000 s/mm(2), the specificity rose to 100%, while only mildly affecting sensitivity (90.6%). No significant correlation was found between ADC values and prognostic factors, such as lymph node metastasis, tumour size, oestrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2) status, and tumour grades. CONCLUSION The present study provides consistent evidence to support DWI as a diagnostic tool for breast lesion characterization. A combination of DCE-MRI with DWI is suggested to improve the sensitivity and specificity of lesion characterization.

The sky is falling.

Medical tourism involves travelling to other countries to avail medical, dental, or surgical care. A combination of various factors, such as exorbitant costs of healthcare in industrialised nations, the increased ease and affordability of international travel, favourable currency exchange rates in the global economy, rapidly improving medical technology and standards of care in many countries as well as the ubiquitous Internet, have led to the recent increase in its popularity. The medical tourism market is estimated to grow by USD 2.2 billion with a corresponding increase of USD 60 billion in the healthcare market [1]. Western Europeans and Canadians bypass the long wait periods that are part of their national health plans by getting medical care abroad. Ten per cent of EU patients seek treatment outside their own country and spend an estimated 12 billion Euro [2]. Medical tourism is a rapidly growing industry even in the so-called developing countries, with countries like Mexico, Brazil, Costa Rica, Dominican Republic, Hungary, India, Israel, Jordan, Lithuania, Malaysia, South Africa, Thailand and the Philippines actively promoting it [3]. According to a UN study, the cost differentials for medical services for a variety of procedures may be: • A heart-valve replacement that would cost USD 200,000 in the US is available for USD 10,000 in India inclusive of the round trip airfare and a vacation package. • A joint replacement in Thailand with eight days of physical therapy at a luxury resort costs less than USD 9,000. • Cosmetic-surgery savings are even greater: A full facelift that would cost USD 20,000 in the US is about USD 1,250 in South Africa [1]. • A PET/CT scan performed in Melbourne inclusive of airfare and accommodation in a 4-star hotel is cheaper than what it costs in Singapore, with some pocket money to spare.

The journey so far

As part of our efforts in establishing biij as one of the important open-access journals dealing with topics relating to radiology, biomedical imaging and biomedical intervention, we have tirelessly pursued the indexing of biij in various relevant scientific indices and databases. For this reason, biij has been a member of CrossRef and has been made searchable by the Google Scholar search engine since it was first launched in July 2005. In that year, biij was only listed in the Directory of Open Access Journals (DOAJ). By the end of 2006, the journal was listed in and/or indexed by Chemical Abstracts Service, Inspec, Index Copernicus International, and eGranary Digital Library. In 2007, it was listed in the Open-J Gate web directory. In December 2007, biij received the best news so far in its history of publication: being accepted for indexing by the Elsevier Bibliographic Databases as of 2008 onwards. In the confirmation letter from Elsevier, the inclusion of biij is “in recognition of the high quality and relevance (of its contents) to the scientific community”. The acceptance means that the contents of biij will now be available in Scopus, EMBASE, EMCare, Compendex, and several other specialised niche databases once the subscription processes have been completed. The inclusion of biij in Scopus is particularly important, as it is one of the largest abstract and citation databases of research literature and web sources, and has developed a large worldwide customer base. This will certainly pave the way for the inclusion of biij in other important indexing services, particularly those provided by Thomson Scientific and National Library of Medicine.