Soo-Yeon Ji

Automated ventricular systems segmentation in brain CT images by combining low-level segmentation and high-level template matching

BackgroundAccurate analysis of CT brain scans is vital for diagnosis and treatment of Traumatic Brain Injuries (TBI). Automatic processing of these CT brain scans could speed up the decision making process, lower the cost of healthcare, and reduce the chance of human error. In this paper, we focus on automatic processing of CT brain images to segment and identify the ventricular systems. The segmentation of ventricles provides quantitative measures on the changes of ventricles in the brain that form vital diagnosis information.MethodsFirst all CT slices are aligned by detecting the ideal midlines in all images. The initial estimation of the ideal midline of the brain is found based on skull symmetry and then the initial estimate is further refined using detected anatomical features. Then a two-step method is used for ventricle segmentation. First a low-level segmentation on each pixel is applied on the CT images. For this step, both Iterated Conditional Mode (ICM) and Maximum A Posteriori Spatial Probability (MASP) are evaluated and compared. The second step applies template matching algorithm to identify objects in the initial low-level segmentation as ventricles. Experiments for ventricle segmentation are conducted using a relatively large CT dataset containing mild and severe TBI cases.ResultsExperiments show that the acceptable rate of the ideal midline detection is over 95%. Two measurements are defined to evaluate ventricle recognition results. The first measure is a sensitivity-like measure and the second is a false positive-like measure. For the first measurement, the rate is 100% indicating that all ventricles are identified in all slices. The false positives-like measurement is 8.59%. We also point out the similarities and differences between ICM and MASP algorithms through both mathematically relationships and segmentation results on CT images.ConclusionThe experiments show the reliability of the proposed algorithms. The novelty of the proposed method lies in its incorporation of anatomical features for ideal midline detection and the two-step ventricle segmentation method. Our method offers the following improvements over existing approaches: accurate detection of the ideal midline and accurate recognition of ventricles using both anatomical features and spatial templates derived from Magnetic Resonance Images.

A survey of cloud-based network intrusion detection analysis

As network traffic grows and attacks become more prevalent and complex, we must find creative new ways to enhance intrusion detection systems (IDSes). Recently, researchers have begun to harness both machine learning and cloud computing technology to better identify threats and speed up computation times. This paper explores current research at the intersection of these two fields by examining cloud-based network intrusion detection approaches that utilize machine learning algorithms (MLAs). Specifically, we consider clustering and classification MLAs, their applicability to modern intrusion detection, and feature selection algorithms, in order to underline prominent implementations from recent research. We offer a current overview of this growing body of research, highlighting successes, challenges, and future directions for MLA-usage in cloud-based network intrusion detection approaches.

A comparative analysis of multi-level computer-assisted decision making systems for traumatic injuries

BackgroundThis paper focuses on the creation of a predictive computer-assisted decision making system for traumatic injury using machine learning algorithms. Trauma experts must make several difficult decisions based on a large number of patient attributes, usually in a short period of time. The aim is to compare the existing machine learning methods available for medical informatics, and develop reliable, rule-based computer-assisted decision-making systems that provide recommendations for the course of treatment for new patients, based on previously seen cases in trauma databases. Datasets of traumatic brain injury (TBI) patients are used to train and test the decision making algorithm. The work is also applicable to patients with traumatic pelvic injuries.MethodsDecision-making rules are created by processing patterns discovered in the datasets, using machine learning techniques. More specifically, CART and C4.5 are used, as they provide grammatical expressions of knowledge extracted by applying logical operations to the available features. The resulting rule sets are tested against other machine learning methods, including AdaBoost and SVM. The rule creation algorithm is applied to multiple datasets, both with and without prior filtering to discover significant variables. This filtering is performed via logistic regression prior to the rule discovery process.ResultsFor survival prediction using all variables, CART outperformed the other machine learning methods. When using only significant variables, neural networks performed best. A reliable rule-base was generated using combined C4.5/CART. The average predictive rule performance was 82% when using all variables, and approximately 84% when using significant variables only. The average performance of the combined C4.5 and CART system using significant variables was 89.7% in predicting the exact outcome (home or rehabilitation), and 93.1% in predicting the ICU length of stay for airlifted TBI patients.ConclusionThis study creates an efficient computer-aided rule-based system that can be employed in decision making in TBI cases. The rule-bases apply methods that combine CART and C4.5 with logistic regression to improve rule performance and quality. For final outcome prediction for TBI cases, the resulting rule-bases outperform systems that utilize all available variables.

A multi-level intrusion detection method for abnormal network behaviors

Abnormal network traffic analysis has become an increasingly important research topic to protect computing infrastructures from intruders. Yet, it is challenging to accurately discover threats due to the high volume of network traffic. To have better knowledge about network intrusions, this paper focuses on designing a multi-level network detection method. Mainly, it is composed of three steps as (1) understanding hidden underlying patterns from network traffic data by creating reliable rules to identify network abnormality, (2) generating a predictive model to determine exact attack categories, and (3) integrating a visual analytics tool to conduct an interactive visual analysis and validate the identified intrusions with transparent reasons.To verify our approach, a broadly known intrusion dataset (i.e. NSL-KDD) is used. We found that the generated rules maintain a high performance rate and provide clear explanations. The proposed predictive model resulted about 96% of accuracy in detecting exact attack categories. With the interactive visual analysis, a significant difference among the attack categories was discovered by visually representing attacks in separated clusters. Overall, our multi-level detection method is well-suited for identifying hidden underlying patterns and attack categories by revealing the relationship among the features of network traffic data.

Detection of P, QRS, and T Components of ECG using wavelet transformation

Electrocardiogram (ECG) signals are composed of five important waves: P, Q, R, S, and T. Sometimes, a sixth wave (U) may follow T. Q, R, and S are grouped together to form the QRS-complex. Detection of these waves is a vital step in ECG signal analysis to extract hidden patterns. Many prior studies have focused only on detection of the QRS-complex, because P and T waves are sparse and harder to isolate from the signal. In this paper, we develop an algorithm to detect all five waves - P, Q, R, S, and T in ECG signals using wavelet transformation. The accuracy for P wave detection is 99.5%, 99.8% for QRS complex, and 99.2% for T waves.

Automated segmentation of lateral ventricles in brain CT images

It is estimated that every year, 1.5 million people in the United States sustain a traumatic brain injury (TBI). Over 50,000 of these patients will not survive, and many others will be left permanently disabled. TBI is known to be accompanied by an increase in intracranial pressure (ICP), as the presence of hematomas compresses the brain tissue. Severe ICP can be fatal, and so must be monitored. This typically requires cranial trepanation, a risky procedure for the patient. However, some signs of increased ICP are visible on medical scans. For example, the lateral ventricles may change in size and position, depending on the location of the original injury. In this paper, we focus on automatic processing of CT brain images to segment and identify the lateral ventricles, using both iterated conditional models (ICM) and maximum a posteriori spatial probability (MASP). The ideal midline of the brain is found via exhaustive search based on skull symmetry and tissue features. The horizontal shift in the ventricles associated with increased ICP can then later be calculated based on the ideal midline. The novelty of the proposed method lies in its combination of anatomical features with template matching against MRI images, its stepwise improvement of the detected actual midline, and its comparison of two existing methods, ICM and MASP, for ventricle detection. The relatively large size of the CT dataset used for testing increases the reliability of the results.

A Comparative Medical Informatics Approach to Traumatic Pelvic Injuries

Traumatic pelvic injury is one of the life-threatening injuries because it is often associated with serious hemorrhage. Therefore, it is necessary to provide immediate medical treatments to the pelvic injury patients. But it is often difficult to make a decision because of a lot of similar cases existed. To help this problem, there are several medical applications designed to provide optimized management of hemorrhage during pelvic injuries. Even though they are useful, it is still necessary to have computer-aided system which assists either evaluating the severity of trauma or blood loss and making the most reliable medical treatments depending on the current status of patient by comparing the current trauma case with previously occurred cases, and helping trauma surgeons make more reliable and immediate decisions. Also it has been suggested that having a trauma system with the emphasis on optimal resource utilization and decision-making through computer aided decision-making systems offers the best chance to reduce the cost of trauma care [2]. In this paper, we designed an efficient computer assisted trauma decision making system for traumatic pelvic injuries. More specifically, a rule-based system is designed to create a reliable method of making predictions/recommendations on the status and outcome (ICU days) of treatments of pelvic trauma injuries using nonlinear regression methods and C4.5.

Towards A Differential Privacy and Utility Preserving Machine Learning Classifier

Abstract Many organizations transact in large amounts of data often containing personal identifiable information (PII) and various confidential data. Such organizations are bound by state, federal, and international laws to ensure that the confidentiality of both individuals and sensitive data is not compromised. However, during the privacy preserving process, the utility of such datasets diminishes even while confidentiality is achieved--a problem that has been defined as NP-Hard. In this paper, we investigate a differential privacy machine learning ensemble classifier approach that seeks to preserve data privacy while maintaining an acceptable level of utility. The first step of the methodology applies a strong data privacy granting technique on a dataset using differential privacy. The resulting perturbed data is then passed through a machine learning ensemble classifier, which aims to reduce the classification error, or, equivalently, to increase utility. Then, the association between increasing the number of weak decision tree learners and data utility, which informs us as to whether the ensemble machine learner would classify more correctly is examined. As results, we found that a combined adjustment of the privacy granting noise parameters and an increase in the number of weak learners in the ensemble machine might lead to a lower classification error.

Wavelet based analysis of physiological signals for prediction of severity of hemorrhagic shock

Rapid detection and treatment of hemorrhagic injuries are important factors in decreasing mortality in the battlefield and civilian trauma settings. In this study, novel features based on discrete wavelet transformation (DWT) were used to analyze physiological signals for prediction of central hypovolemia severity in humans. These features were defined based on approximate and detailed DWT coefficients extracted from physiological signals such as the electrocardiogram (ECG), arterial blood pressure (ABP), and thoracic impedance (IZT and DZT) signals, collected on healthy humans exposed to a hemorrhage model called lower body negative pressure (LBNP). The LBNP protocol consisted of applying 0, -15, -30, -45, -60, -70 mm Hg pressure to the lower half of the body, for 5 minutes at each stage. These LBNP levels were divided into three classes: mild, moderate, and severe. Machine learning algorithms were applied to predict the severity of blood loss based on the features extracted from the physiological signals. One of the objectives of this study was to compare the utility of using multiple physiological signals in prediction of the severity of hypovolemia as opposed to only using ECG. The classification results indicate that SVM has the highest accuracy at 82%. SVMs average precision and recall for all three classes are 79.2% and 79.8%, respectively. This shows that the wavelet-based method using multiple signals has the ability of rapidly determining the degree of volume loss, providing a potential tool for real-time remote triage and decision making in victims of trauma.

Designing an Internet Traffic Predictive Model by Applying a Signal Processing Method

Detection of abnormal internet traffic has become a significant area of research in network security. Due to its importance, many predictive models are designed by utilizing machine learning algorithms. The models are well designed to show high performances in detecting abnormal internet traffic behaviors. However, they may not guarantee reliable detection performances for new incoming abnormal internet traffic because they are designed using raw features from imbalanced internet traffic data. Since internet traffic is non-stationary time-series data, it is difficult to identify abnormal internet traffic with the raw features. In this study, we propose a new approach to detecting abnormal internet traffic. Our approach begins with extracting hidden, but important, features by utilizing discrete wavelet transformation. Then, statistical analysis is performed to filter out irrelevant and less important features. Only statistically significant features are used to design a reliable predictive model with logistic regression. A comparative analysis is conducted to determine the importance of our approach by measuring accuracy, sensitivity, and the Area Under the receiver operating characteristic Curve. From the analysis, we found that our model detects abnormal internet traffic successfully with high accuracy.