Unstructured medical image query using big data - An epilepsy case study.

Big data technologies are critical to the medical field which requires new frameworks to leverage them. Such frameworks would benefit medical experts to test hypotheses by querying huge volumes of unstructured medical data to provide better patient care. The objective of this work is to implement and examine the feasibility of having such a framework to provide efficient querying of unstructured data in unlimited ways. The feasibility study was conducted specifically in the epilepsy field. The proposed framework evaluates a query in two phases. In phase 1, structured data is used to filter the clinical data warehouse. In phase 2, feature extraction modules are executed on the unstructured data in a distributed manner via Hadoop to complete the query. Three modules have been created, volume comparer, surface to volume conversion and average intensity. The framework allows for user-defined modules to be imported to provide unlimited ways to process the unstructured data hence potentially extending the application of this framework beyond epilepsy field. Two types of criteria were used to validate the feasibility of the proposed framework - the ability/accuracy of fulfilling an advanced medical query and the efficiency that Hadoop provides. For the first criterion, the framework executed an advanced medical query that spanned both structured and unstructured data with accurate results. For the second criterion, different architectures were explored to evaluate the performance of various Hadoop configurations and were compared to a traditional Single Server Architecture (SSA). The surface to volume conversion module performed up to 40 times faster than the SSA (using a 20 node Hadoop cluster) and the average intensity module performed up to 85 times faster than the SSA (using a 40 node Hadoop cluster). Furthermore, the 40 node Hadoop cluster executed the average intensity module on 10,000 models in 3h which was not even practical for the SSA. The current study is limited to epilepsy field and further research and more feature extraction modules are required to show its applicability in other medical domains. The proposed framework advances data-driven medicine by unleashing the content of unstructured medical data in an efficient and unlimited way to be harnessed by medical experts.

A simulation to analyze feature selection methods utilizing gene ontology for gene expression classification.

Gene expression profile classification is a pivotal research domain assisting in the transformation from traditional to personalized medicine. A major challenge associated with gene expression data classification is the small number of samples relative to the large number of genes. To address this problem, researchers have devised various feature selection algorithms to reduce the number of genes. Recent studies have been experimenting with the use of semantic similarity between genes in Gene Ontology (GO) as a method to improve feature selection. While there are few studies that discuss how to use GO for feature selection, there is no simulation study that addresses when to use GO-based feature selection. To investigate this, we developed a novel simulation, which generates binary class datasets, where the differentially expressed genes between two classes have some underlying relationship in GO. This allows us to investigate the effects of various factors such as the relative connectedness of the underlying genes in GO, the mean magnitude of separation between differentially expressed genes denoted by δ, and the number of training samples. Our simulation results suggest that the connectedness in GO of the differentially expressed genes for a biological condition is the primary factor for determining the efficacy of GO-based feature selection. In particular, as the connectedness of differentially expressed genes increases, the classification accuracy improvement increases. To quantify this notion of connectedness, we defined a measure called Biological Condition Annotation Level BCAL(G), where G is a graph of differentially expressed genes. Our main conclusions with respect to GO-based feature selection are the following: (1) it increases classification accuracy when BCAL(G) ≥ 0.696; (2) it decreases classification accuracy when BCAL(G) ≤ 0.389; (3) it provides marginal accuracy improvement when 0.389

Curvature analysis of perisylvian epilepsy.

PURPOSE: We assess whether alterations in the convolutional anatomy of the deep perisylvian area (DPSA) might indicate focal epileptogenicity. MATERIALS AND METHODS: The DPSA of each hemisphere was segmented on MRI and a 3D gray-white matter interface (GWMI) geometrical model was constructed. Comparative visual and quantitative assessment of the convolutional anatomy of both the left and right DPSA models was performed. Both the density of thorn-like contours (peak percentage) and coarse interface curvatures was computed using Gaussian curvature and shape index, respectively. The proposed method was applied to a total of 14 subjects; 7 patients with an epileptogenic DPSA and 7 non-epileptic subjects. RESULTS: A high peak percentage correlated well with the epileptogenic DPSA. It distinguished between patients and non-epileptic subjects (P = 0.029) and identified laterality of the epileptic focus in all but one case. A diminished regional curvature also identified epileptogenicity (P = 0.016) and, moreover, its laterality (P = 0.001). CONCLUSION: An increased peak percentage from a global view of the GWMI of the DPSA provides some indication of a propensity toward a focal or regional DPSA epileptogenicity. A diminished convolutional anatomy (i.e., smoothing effect) appears also to coincide with the epileptogenic site in the DPSA and to distinguish laterality.

Knowledge-based localization of hippocampus in human brain MRI.

We present a novel and efficient method for localization of human brain structures such as hippocampus. Landmark localization is important for segmentation and registration. This method follows a statistical roadmap, consisting of anatomical landmarks, to reach the desired structures. Using a set of desired and undesired landmarks, identified on a training set, we estimate Gaussian models and determine optimal search areas for desired landmarks. The statistical models form a set of rules to evaluate the extracted landmarks during the search procedure. When applied on 900 MR images of 10 epileptic patients, this method demonstrated an overall success rate of 83%.

Content-based image database system for epilepsy.

We have designed and implemented a human brain multi-modality database system with content-based image management, navigation and retrieval support for epilepsy. The system consists of several modules including a database backbone, brain structure identification and localization, segmentation, registration, visual feature extraction, clustering/classification and query modules. Our newly developed anatomical landmark localization and brain structure identification method facilitates navigation through an image data and extracts useful information for segmentation, registration and query modules. The database stores T1-, T2-weighted and FLAIR MRI and ictal/interictal SPECT modalities with associated clinical data. We confine the visual feature extractors within anatomical structures to support semantically rich content-based procedures. The proposed system serves as a research tool to evaluate a vast number of hypotheses regarding the condition such as resection of the hippocampus with a relatively small volume and high average signal intensity on FLAIR. Once the database is populated, using data mining tools, partially invisible correlations between different modalities of data, modeled in database schema, can be discovered. The design and implementation aspects of the proposed system are the main focus of this paper.

Design, implementation and accuracy of a prototype for medical augmented reality.

OBJECTIVE: This paper is focused on prototype development and accuracy evaluation of a medical Augmented Reality (AR) system. The accuracy of such a system is of critical importance for medical use, and is hence considered in detail. We analyze the individual error contributions and the system accuracy of the prototype. MATERIALS AND METHODS: A passive articulated arm is used to track a calibrated end-effector-mounted video camera. The live video view is superimposed in real time with the synchronized graphical view of CT-derived segmented object(s) of interest within a phantom skull. The AR accuracy mostly depends on the accuracy of the tracking technology, the registration procedure, the camera calibration, and the image scanning device (e.g., a CT or MRI scanner). RESULTS: The accuracy of the Microscribe arm was measured to be 0.87 mm. After mounting the camera on the tracking device, the AR accuracy was measured to be 2.74 mm on average (standard deviation = 0.81 mm). After using data from a 2-mm-thick CT scan, the AR error remained essentially the same at an average of 2.75 mm (standard deviation = 1.19 mm). CONCLUSIONS: For neurosurgery, the acceptable error is approximately 2-3 mm, and our prototype approaches these accuracy requirements. The accuracy could be increased with a higher-fidelity tracking system and improved calibration and object registration. The design and methods of this prototype device can be extrapolated to current medical robotics (due to the kinematic similarity) and neuronavigation systems.

Continence Index: a new screening questionnaire to predict the probability of future incontinence in older women in the community.

PURPOSE: Urinary incontinence (UI) is a chronic, costly condition that impairs quality of life. To identify older women most at risk, the Medical Epidemiologic and Social Aspects of Aging (MESA) datasets were mined to create a set of questions that can reliably predict future UI. METHODS: MESA data were collected during four household interviews at approximately 1 year intervals. Factors associated with becoming incontinent at the second interview (HH2) were identified using logistic regression (construction datasets). Based on p values and odds ratios, eight potential predictive factors with their 256 combinations and corresponding prediction probabilities formed the Continence Index. Its predictive and discriminatory capability was tested against the same cohort's outcome in the fourth survey (HH4 validation datasets). Sensitivity analysis, area under receiver operating characteristic (ROC) curve, predicted probabilities and confidence intervals were used to statistically validate the Continence Index. RESULTS: Body mass index, sneezing, post-partum UI, urinary frequency, mild UI, belief of developing UI in the future, difficulty stopping urinary stream and remembering names emerged as the strongest predictors of UI. The confidence intervals for prediction probabilities strongly agreed between construction and validation datasets. Calculated sensitivity, specificity, false-positive and false-negative values revealed that the areas under the ROCs (0.802 and 0.799) for the construction and validation datasets, respectively, indicated good discriminatory capabilities of the index as a predictor. CONCLUSION: The Continence Index will help identify older women most at risk of UI in order to apply targeted prevention strategies in women that are most likely to benefit.

Development and human factors analysis of neuronavigation vs. augmented reality.

This paper is focused on the human factors analysis comparing a standard neuronavigation system with an augmented reality system. We use a passive articulated arm (Microscribe, Immersion technology) to track a calibrated end-effector mounted video camera. In real time, we superimpose the live video view with the synchronized graphical view of CT-derived segmented object(s) of interest within a phantom skull. Using the same robotic arm, we have developed a neuronavigation system able to show the end-effector of the arm on orthogonal CT scans. Both the AR and the neuronavigation systems have been shown to be within 3mm of accuracy. A human factors study was conducted in which subjects were asked to draw craniotomies and answer questions to gage their understanding of the phantom objects. The human factors study included 21 subjects and indicated that the subjects performed faster, with more accuracy and less errors using the Augmented Reality interface.

A radiographic comparison of progressive and conventional loading on crestal bone loss and density in single dental implants: a randomized controlled trial study.

OBJECTIVE: Crestal bone loss is a biological complication in implant dentistry. The aim of this study was to compare the effect of progressive and conventional loading on crestal bone height and bone density around single osseointegrated implants in the posterior maxilla by a longitudinal radiographic assessment technique. MATERIALS AND METHODS: Twenty micro thread implants were placed in 10 patients (two implants per patient). One of the two implants in each patient was assigned to progressive and the other to conventional loading groups. Eight weeks after surgery, conventional implants were restored with a metal ceramic crown and the progressive group underwent a progressive loading protocol. The progressive loading group took different temporary acrylic crowns at 2, 4 and 6 months. After eight months, acrylic crowns were replaced with a metal ceramic crown. Computer radiography of both progressive and conventional implants was taken at 2, 4, 6, and 12 months. Image analysis was performed to measure the height of crestal bone loss and bone density. RESULTS: The mean values of crestal bone loss at month 12 were 0.11 (0.19) mm for progressively and 0.36 (0.36) mm for conventionally loaded implants, with a statistically significant difference (P < 0.05) using Wilcoxon sign rank. Progressively loaded group showed a trend for higher bone density gain compared to the conventionally loaded group, but when tested with repeated measure ANOVA, the differences were not statistically significant (P > 0.05). CONCLUSION: The progressive group showed less crestal bone loss in single osseointegrated implant than the conventional group. Bone density around progressively loaded implants showed increase in crestal, middle and apical areas.

Novel application of statistical methods to identify new urinary incontinence risk factors.

Longitudinal data for studying urinary incontinence (UI) risk factors are rare. Data from one study, the hallmark Medical, Epidemiological, and Social Aspects of Aging (MESA), have been analyzed in the past; however, repeated measures analyses that are crucial for analyzing longitudinal data have not been applied. We tested a novel application of statistical methods to identify UI risk factors in older women. MESA data were collected at baseline and yearly from a sample of 1955 men and women in the community. Only women responding to the 762 baseline and 559 follow-up questions at one year in each respective survey were examined. To test their utility in mining large data sets, and as a preliminary step to creating a predictive index for developing UI, logistic regression, generalized estimating equations (GEEs), and proportional hazard regression (PHREG) methods were used on the existing MESA data. The GEE and PHREG combination identified 15 significant risk factors associated with developing UI out of which six of them, namely, urinary frequency, urgency, any urine loss, urine loss after emptying, subject's anticipation, and doctor's proactivity, are found most highly significant by both methods. These six factors are potential candidates for constructing a future UI predictive index.