Tuan D. Pham

Development of a brain MRI-based hidden Markov model for dementia recognition.

BackgroundDementia is an age-related cognitive decline which is indicated by an early degeneration of cortical and sub-cortical structures. Characterizing those morphological changes can help to understand the disease development and contribute to disease early prediction and prevention. But modeling that can best capture brain structural variability and can be valid in both disease classification and interpretation is extremely challenging. The current study aimed to establish a computational approach for modeling the magnetic resonance imaging (MRI)-based structural complexity of the brain using the framework of hidden Markov models (HMMs) for dementia recognition.MethodsRegularity dimension and semi-variogram were used to extract structural features of the brains, and vector quantization method was applied to convert extracted feature vectors to prototype vectors. The output VQ indices were then utilized to estimate parameters for HMMs. To validate its accuracy and robustness, experiments were carried out on individuals who were characterized as non-demented and mild Alzheimers diseased. Four HMMs were constructed based on the cohort of non-demented young, middle-aged, elder and demented elder subjects separately. Classification was carried out using a data set including both non-demented and demented individuals with a wide age range.ResultsThe proposed HMMs have succeeded in recognition of individual who has mild Alzheimers disease and achieved a better classification accuracy compared to other related works using different classifiers. Results have shown the ability of the proposed modeling for recognition of early dementia.ConclusionThe findings from this research will allow individual classification to support the early diagnosis and prediction of dementia. By using the brain MRI-based HMMs developed in our proposed research, it will be more efficient, robust and can be easily used by clinicians as a computer-aid tool for validating imaging bio-markers for early prediction of dementia.

Sample entropy and regularity dimension in complexity analysis of cortical surface structure in early Alzheimer's disease and aging

We apply for the first time the sample entropy (SampEn) and regularity dimension model for measuring signal complexity to quantify the structural complexity of the brain on MRI. The concept of the regularity dimension is based on the theory of chaos for studying nonlinear dynamical systems, where power laws and entropy measure are adopted to develop the regularity dimension for modeling a mathematical relationship between the frequencies with which information about signal regularity changes in various scales. The sample entropy and regularity dimension of MRI-based brain structural complexity are computed for early Alzheimers disease (AD) elder adults and age and gender-matched non-demented controls, as well as for a wide range of ages from young people to elder adults. A significantly higher global cortical structure complexity is detected in AD individuals (p<0.001). The increase of SampEn and the regularity dimension are also found to be accompanied with aging which might indicate an age-related exacerbation of cortical structural irregularity. The provided model can be potentially used as an imaging bio-marker for early prediction of AD and age-related cognitive decline.

Computerized assessment of communication for cognitive stimulation for people with cognitive decline using spectral-distortion measures and phylogenetic inference

Therapeutic communication and interpersonal relationships in care homes can help people to improve their mental wellbeing. Assessment of the efficacy of these dynamic and complex processes are necessary for psychosocial planning and management. This paper presents a pilot application of photoplethysmography in synchronized physiological measurements of communications between the care-giver and people with dementia. Signal-based evaluations of the therapy can be carried out using the measures of spectral distortion and the inference of phylogenetic trees. The proposed computational models can be of assistance and cost-effectiveness in caring for and monitoring people with cognitive decline.

Automated CT detection of intestinal abnormalities and ischemia for decision making in emergency medicine

BackgroundEvaluation of computed tomography (CT) for the diagnosis of intestinal wall abnormalities and ischemia is important for clinical decision making in patients with acute abdominal pain to which if surgery should be performed in the emergency department. Interpretation of such information on CT is usually based on visual assessment by medical professionals and still remains a challenge in a variety of settings of the medical emergency care. This paper reports a pilot study in the implementation of image processing methods for automated detection of intestinal wall abnormalities and bowel ischemia, which can be of a potential application for CT-based detection of the intestinal disease.MethodsCT scans of 3 patients of ischemia, one benign and one control subjects were used in this study. Statistical and geometrical features of the CT scans were extracted for pattern classification using two distance measures and the k-nearest neighbor algorithm. The automated detection of intestinal abnormalities and ischemia was carried out using labeled data from the training process with various proportions of training and testing samples to validate the results.ResultsDetection rates of intestinal ischemia and abnormalities are promising in terms of sensitivity and specificity, where the sensitivity is higher than the specificity in all test cases. The overall classification accuracy between the diseased and control subjects can be as high as 100% when all CT scans were included for measuring the difference between a cohort of three patients of ischemia and a single control subject.ConclusionThe proposed approach can be utilized as a computer-aided tool for decision making in the emergency department, where the availability of expert knowledge of the radiologist and surgeon about this complex bowel disease is limited.

Measures of Morphological Complexity of Gray Matter on Magnetic Resonance Imaging for Control Age Grouping

Current brain-age prediction methods using magnetic resonance imaging (MRI) attempt to estimate the physiological brain age via some kind of machine learning of chronological brain age data to perform the classification task. Such a predictive approach imposes greater risk of either over-estimate or under-estimate, mainly due to limited training data. A new conceptual framework for more reliable MRI-based brain-age prediction is by systematic brain-age grouping via the implementation of the phylogenetic tree reconstruction and measures of information complexity. Experimental results carried out on a public MRI database suggest the feasibility of the proposed concept.

Segmentation of mitochondria in intracellular space

Information of cellular organelle location and morphology is essential for cancer simulation. In order to obtain such information, the segmentation of the organelles from electronic microscopy intracellular image is crucial. In this paper, we focus on the segmentation of mitochondria organelle which is one of the most important organelles tightly related to the form of cancer. A simple three-stage strategy for mitochondrial segmentation based on exclusive and morphology properties and Gabor filter is proposed. Experimental results on focused ion beam (FIB) and scanning electron microscope (SEM) images have shown the effectiveness of proposed method.

Biomedical Informatics and Technology

Wakefulness state estimation prior electroencephalography (EEG) measurements is important for more precise interpretation of those signals. Thus, a new method based on approximate entropy (ApEn) using a new range of ApEn parameter values was used to be compared with different spectral-based measures (i.e., relative delta subband power R.δ, relative theta sub-band power R.θ, power ratio between theta and alpha Pθ/α and power ratio between theta and beta Pθ/β) using occipital-alpha rhythm. The performances of the aforementioned measures were determined by using decision threshold value that satisfy the minimum misclassification rate between two groups: fully awake and light drowsy, where each of which was composed from 45 subjects. To determine the performance of the aforementioned measures, the results of the comparisons were based on cross validation method. Our results indicate that ApEn is better than R.δ, R.θ, Pθ/α and Pθ/β in evaluating wakefulness state with 9.24%, 6.57%, 10.25% and 4.74% respectively.

Chaos Analysis of Brain MRI for Studying Mental Disorders

A tendency of increase has been observed for mental disorders in the last years. However, the conventional diagnosis methods like history taking suffer from low objectivity. Recently, some research on the diagnosis of mental disorders focused on cerebral blood flow change has been reported. Data of near-infrared spectroscopy(NIRS) is needed for this analysis, but this device is not available in all medical facilities and the method is not established yet. It is a current challenge to develop a method that can be conducted easily and provides an accurate diagnosis for patients at an easily stage of the disease. In this research, we based our analysis method on brain MRI, which many medical institutions use to take medical images. At first the characteristics of the grey matter distribution have been extracted in order to generate a time series of this feature for every slice image. The time series were further analyzed by extracting the largest Lyapunov Exponent(LLE), a measure of chaos theory. The descriptor of mental health of a subject is finally calculated by averaging the LLE measures for all image slices. We were able to confirm that a significant difference of descriptors can be found for young, middle-age, elderly persons and patients with Alzheimer’s disease. It is an open task for the future to verify if this method can be used for other mental diseases as well. Our long-term perspective is to introduce this diagnostic method to official medical guidelines.

Chaotic Behavior in Intracellular Space: An Implication for Modeling and Simulation of Cancer

Understanding what causes cancer remains the most challenging research in medicine and biology. Biomedical research and clinical trials are spending constant effort in finding out new information regarding anything that has to do with cancer. This is because cancer is a complex disease and our knowledge about cancer is still very primitive. In fact we may never know everything about cancer. We are interested in searching into possible chaotic behavior of cancer in its intracellular space captured by scanning electron microscopy (SEM) and focused ion beam (FIB) images using the theory of chaos. Such a study rarely exists in literature and its finding may give some insight into the identification of an appropriately underlying mechanism for computer modeling and simulation of cancer which is very computationally useful and cost-effective for targeted drug treatment of cancer diseases.

Photoplethysmography technology and its feature visualization for cognitive stimulation assessment

Therapeutic communication is recognized as an alternative cognitive stimulation for people with mental disorders. It is important to measure the effectiveness of such therapeutic treatments. In this paper, we present the use of photoplethysmography (PPG) technology to synchronize communication signals between the care-giver and people with dementia. To gain insights into the communication effect, the largest Lyapunov exponents are extracted from the PPG signals, which are then analyzed by multidimensional scaling to visualize the signal similarity/dissimilarity between the care-giver and participants. Experimental results show that the proposed approach is promising as a useful tool for visual assessment of the influence of the therapy over participants.