True Positive Detection Rate Research Articles

An Intelligent Spam Detection Model Based on Artificial Immune System

Spam emails, also known as non-self, are unsolicited commercial or malicious emails, sent to affect either a single individual or a corporation or a group of people. Besides advertising, these may contain links to phishing or malware hosting websites set up to steal confidential information. In this paper, a study of the effectiveness of using a Negative Selection Algorithm (NSA) for anomaly detection applied to spam filtering is presented. NSA has a high performance and a low false detection rate. The designed framework intelligently works through three detection phases to finally determine an email’s legitimacy based on the knowledge gathered in the training phase. The system operates by elimination through Negative Selection similar to the functionality of T-cells’ in biological systems. It has been observed that with the inclusion of more datasets, the performance continues to improve, resulting in a 6% increase of True Positive and True Negative detection rate while achieving an actual detection rate of spam and ham of 98.5%. The model has been further compared against similar studies, and the result shows that the proposed system results in an increase of 2 to 15% in the correct detection rate of spam and ham.

A Novel Approach for Real-Time Gait Events Detection Using Developed Wireless Foot Sensor Module

This article presents a method for the detection of heel strike (HS) and toe off (TO) for overground walking using developed wireless foot sensor module (WFSM). The proposed algorithm uses variation of foot angle measured from WFSM for estimating stance and swing phase during human gait. The significance and novelty of the developed work lies in the fact that the method validates an approach for detecting gait events using foot angle variation alone, computed from a single inertial sensor. For 11 479 steps recorded from 34 healthy volunteers, it showed 100% true positive detection rate with a mean error of –12.9 ± 30.6 ms and 7.9 ± 10.8 ms for HS and TO, respectively, referenced against foot switches. One-way analysis of variance (ANOVA) test for gait cycle time computed from a force sensitive resistor and WFMS showed no significant difference ( p > 0.01) between the two sensor methods. The developed system was benchmarked against Zebris FDM system (Zebris medical GMBH, Germany) and Kistler Force Platform (Kistler Inc., Colorado, USA) for comparing temporal gait parameters. The proposed work also reports an improvement in terms of detection latency and, thus, shows potential for real-time control applications.

Statistical change detection of building energy consumption: Applications to savings estimation

The surge in interval meter data availability and associated activity in energy data analytics has inspired new interest in advanced methods for building efficiency savings estimation. Statistical and machine learning approaches are being explored to improve the energy baseline models used to measure and verify savings. One outstanding challenge is the ability to identify and account for operational changes that may confound savings estimates. In the measurement and verification (M&V) context, ‘non-routine events’ (NREs) cause changes in building energy use that are not attributable to installed efficiency measures, and not accounted for in the baseline model's independent variables. In the M&V process NREs must be accounted for as ‘adjustments’ to appropriately attribute the estimated energy savings to the specific efficiency interventions that were implemented. Currently this is a manual and custom process, conducted using professional judgment and engineering expertise. As such it remains a barrier in scaling and standardizing meter-based savings estimation.In this work, a data driven methodology was developed to (partially) automate, and therefore streamline the process of detecting NREs in the post-retrofit period and making associated savings adjustments. The proposed NRE detection algorithm is based on a statistical change point detection method and a dissimilarity metric. The dissimilarity metric measures the proximity between the actual time series of the post-retrofit energy consumption and the projected baseline, which is generated using a statistical baseline model. The suggested approach for NRE adjustment involves the NRE detection algorithm, the M&V practitioner, and a regression modeling algorithm. The performance of the detection and adjustment algorithm was evaluated using a simulation-generated test data set, and two benchmark algorithms. Results show a high true positive detection rate (75%-100% across the test cases), higher than ideal false positive detection rates (20%-70%), and low errors in energy adjustment (<0.7%). These results indicate that the algorithm holds for helping M&V practitioners to streamline the process of handling NREs. Moreover, the change point algorithm and underlying statistical principles could prove valuable for other building analytics applications such as anomaly detection and fault diagnostics.

A Dynamic Partitioning Algorithm for Sip Detection using a Bottle-Attachable IMU Sensor

Hydration tracking technologies are a promising tool for improving health outcomes across a variety of populations. As a non-wearable solution that is reconfigurable across containers, bottle-attachable inertial measurement unit (IMU) sensors offer numerous advantages versus alternative tracking approaches. This paper proposes a novel dynamic temporal partitioning and classification algorithm for spotting drinks within the streaming data generated by such sensors. By exploiting the distinguishing characteristics of the container’s estimated inclination during drinking, the algorithm identifies candidate drink intervals for subsequent classification using a Threshold-Merge-Discard framework. The proposed approach is benchmarked against a slight variation of a previously introduced sliding window classifier for a series of experiments replicating the intended use case of the device. The new algorithm is shown to increase the true-positive detection rate by 23.7%, while reducing the number of required classification operations by more than an order of magnitude.

Chest radiography or chest CT plus head and neck CT versus 18F-FDG PET/CT for detection of distant metastasis and synchronous cancer in patients with head and neck cancer

ObjectivesCurrent guidelines recommend chest computed tomography (CT) with locoregional CT/magnetic resonance imaging for patients with head and neck squamous cell carcinoma (HNSCC), and 18F-fluorodeoxyglucose (FDG) positron-emission tomography (PET)/CT is suggested for stage III–IV patients. However, whole body screening with 18F-FDG PET/CT may provide better detection of distant metastases and synchronous cancer than conventional imaging. We evaluated the utility of 18F-FDG PET/CT in detecting distant metastasis and synchronous cancer. MethodsThis prospective study enrolled 740 consecutive patients with previously untreated HNSCC diagnosed between September 2010 and December 2015. Synchronous cancer was histologically confirmed and distant metastases were confirmed by biopsy or serial imaging follow-ups. McNemar test was used to compare the true-positive detection rates of chest radiography (CXR) + head and neck CT (hnCT) (A) versus 18F-FDG PET/CT (C) and chest CT + hnCT (B) versus 18F-FDG PET/CT. ResultsDistant metastases and synchronous cancer were found in 23 (3.1%) and 55 (7.4%) patients, respectively. A, B, and C detected distant metastases in 10 (1.3%), 19 (2.6%), and 21 (2.8%) patients, respectively. The absolute differences were 1.5% (A versus C, P = 0.003) and 0.3% (B versus C, P = 0.687). A, B, and C detected synchronous cancer in 15 (2.0%), 22 (2.9%), and 36 (4.9%) patients, respectively. The absolute differences were 2.8% (A versus C, P < 0.001) and 1.4% (B versus C, P = 0.013). Conclusions18F-FDG PET/CT detected more distant metastases and synchronous cancer than CXR + hnCT and more synchronous cancer than chest CT + hnCT.

An assessment of true and false positive detection rates of stepwise epistatic model selection as a function of sample size and number of markers

Association studies have been successful at identifying genomic regions associated with important traits, but routinely employ models that only consider the additive contribution of an individual marker. Because quantitative trait variability typically arises from multiple additive and non-additive sources, utilization of statistical approaches that include main and two-way interaction marker effects of several loci in one model could lead to unprecedented characterization of these sources. Here we examine the ability of one such approach, called the Stepwise Procedure for constructing an Additive and Epistatic Multi-Locus model (SPAEML), to detect additive and epistatic signals simulated using maize and human marker data. Our results revealed that SPAEML was capable of detecting quantitative trait nucleotides (QTNs) at sample sizes as low as n = 300 and consistently specifying signals as additive and epistatic for larger sizes. Sample size and minor allele frequency had a major influence on SPAEML’s ability to distinguish between additive and epistatic signals, while the number of markers tested did not. We conclude that SPAEML is a useful approach for providing further elucidation of the additive and epistatic sources contributing to trait variability when applied to a small subset of genome-wide markers located within specific genomic regions identified using a priori analyses.

A cooperative approach with improved performance for a global intrusion detection systems for internet service providers

Typical perimeter-based intrusion detection systems do not provide the user with the necessary preventive protection measures. In addition, many of the available solutions still need to improve their true-positive detection rates and reduce the proportion of false-positive alarms. Therefore, internet service providers, utilising this type of device to defend their assets and subscribers against malicious traffic, may be induced by them to make incorrect decisions. In this paper, we propose a global intrusion detection system, based upon the BGP protocol that establishes a cooperative federation whose members are distributed autonomous intrusion detection elements. These elements are able to propagate alarms of potential threatening flows traversing their respective autonomous systems. We present the architecture for the described approach and an analytical model based upon Dempster-Shafer’s combination rule, in order to evaluate specific performance metrics. The results show significant improvements over the assessed metrics, highlighting the advantage of using the proposed solution as a frontline to prevent cyberattacks.

Learning deep features of spectrogram to estimate acoustic density of a fish species inhabitating Posidonia oceanica seagrasses

In field acoustic investigations on seagrass ecology, characteristic /kwa/ sound production of a fish species systematically present in meadows constituted by Posidonia oceanica species were recorded (Mediterranean, since 1995). Machine learning is investigated to automate estimation of acoustic density of the /kwa/ species. A multi-model deep neural network is implemented that directly inputs the spectrograms of signals recorded on hydrophones above and within the canopy. The network is a combination of 1D convolutional neural networks and a recurrent neural network with long short-term memory units and fully connected networks. Such architecture is able to process spectral information in the convolutional network and temporal information in the recurrent network, thereby making the best use of information in both time and frequency domains. In spite of high background noise level, the true-positive detection rate is shown to be excellent. With respect to human expert supervision, false-positive detections were, for the major part, faint sounds produced by distant fishes of the same species, as verified through careful re-listening. The apparent error rate is 8.4 ×10−2 on a partially mislabeled dataset of three days (7-10 pm). The true error rate, after label correction, is estimated at 1×10−2.

Of daemons and men: reducing false positive rate in intrusion detection systems with file system footprint analysis

In this work, we propose a methodology for reducing false alarms in file system intrusion detection systems, by taking into account the daemon’s file system footprint. More specifically, we experimentally show that sequences of outliers can serve as a distinguishing characteristic between true and false positives, and we show how analysing sequences of outliers can lead to lower false positive rates, while maintaining high detection rates. Based on this analysis, we developed an anomaly detection filter that learns outlier sequences using k-nearest neighbours with normalised longest common subsequence. Outlier sequences are then used as a filter to reduce false positives on the $$FI^2DS$$ file system intrusion detection system. This filter is evaluated on both overlapping and non-overlapping sequences of outliers. In both cases, experiments performed on three real-world web servers and a honeynet show that our approach achieves significant false positive reduction rates (up to 50 times), without any degradation of the corresponding true positive detection rates.

Wireless Smartphone Control using Electromyography and Automated Gesture Recognition.

In this paper, a wearable, wireless system is demonstrated using electromyography (EMG) signals for realtime control of a smartphone device. The system allows gesturebased control of a smartphone or tablet computer without physical contact, direct line of sight, or significant movement. Additionally, automated gesture detection is shifted to the smartphone, eliminating the need for robust computing hardware. The electronic system and gesture prediction algorithm are described, and measured results are presented and for multiple users. The system demonstrates a maximum true positive detection rate of 92% for a trained user, using three distinct hand gestures. The EMG-based detection system serves as a proof-of-concept for providing wireless, gesture-based control of computer interfaces using low-cost consumer hardware.

Evaluation of Takagi-Sugeno-Kang fuzzy method in entropy-based detection of DDoS attacks

The detection of distributed denial of service (DDoS) attacks based on internet traffic anomalies is a method which is general in nature and can detect unknown or zero-day attacks. One of the statistical characteristics used for this purpose is network traffic entropy: a sudden change in entropy may indicate a DDoS attack. However, this approach often gives false positives, and this is the main obstacle to its wider deployment within network security equipment. In this paper, we propose a new, two-step method for detection of DDoS attacks. This method combines the approaches of network traffic entropy and the Takagi-Sugeno-Kang fuzzy system. In the first step, the detection process calculates the entropy distribution of the network packets. In the second step, the Takagi-Sugeno-Kang fuzzy system (TSK-FS) method is applied to these entropy values. The performance of the TSK-FS method is compared with that of the typically used approach, in which cumulative sum (CUSUM) change point detection is applied directly to entropy time series. The results show that the TSK-FS DDoS detector reaches enhanced sensitivity and robustness in the detection process, achieving a high true-positive detection rate and a very low false-positive rate. As it is based on entropy, this combined method retains its generality and is capable of detecting various types of attack.

Detecting Smoking Events Using Accelerometer Data Collected Via Smartwatch Technology: Validation Study.

BackgroundSmoking is the leading cause of preventable death in the world today. Ecological research on smoking in context currently relies on self-reported smoking behavior. Emerging smartwatch technology may more objectively measure smoking behavior by automatically detecting smoking sessions using robust machine learning models.ObjectiveThis study aimed to examine the feasibility of detecting smoking behavior using smartwatches. The second aim of this study was to compare the success of observing smoking behavior with smartwatches to that of conventional self-reporting.MethodsA convenience sample of smokers was recruited for this study. Participants (N=10) recorded 12 hours of accelerometer data using a mobile phone and smartwatch. During these 12 hours, they engaged in various daily activities, including smoking, for which they logged the beginning and end of each smoking session. Raw data were classified as either smoking or nonsmoking using a machine learning model for pattern recognition. The accuracy of the model was evaluated by comparing the output with a detailed description of a modeled smoking session.ResultsIn total, 120 hours of data were collected from participants and analyzed. The accuracy of self-reported smoking was approximately 78% (96/123). Our model was successful in detecting 100 of 123 (81%) smoking sessions recorded by participants. After eliminating sessions from the participants that did not adhere to study protocols, the true positive detection rate of the smartwatch based-detection increased to more than 90%. During the 120 hours of combined observation time, only 22 false positive smoking sessions were detected resulting in a 2.8% false positive rate.ConclusionsSmartwatch technology can provide an accurate, nonintrusive means of monitoring smoking behavior in natural contexts. The use of machine learning algorithms for passively detecting smoking sessions may enrich ecological momentary assessment protocols and cessation intervention studies that often rely on self-reported behaviors and may not allow for targeted data collection and communications around smoking events.

Rapid Line-Extraction Method for SAR Images Based on Edge-Field Features

This letter proposes a rapid line-extraction (RLE) method for synthetic aperture radar (SAR) images. RLE first transforms an image in the space domain into an image in the frequency domain. Then, using the central-slice theorem, RLE skilfully maps the image in the frequency domain into a parameter space, which effectively accelerates the straight-line extraction process. Unlike the traditional Hough transform, RLE is performed directly on an edge-field image rather than on a binary edge map. Theoretical analysis proves the advantages of using the edge-field map. Notably, the computational complexity can be greatly reduced relative to the complexity of obtaining a binary edge map, and the method can efficiently avoid the negative influence of false edges in the binary edge map. More importantly, because speckle, clutter, and blurred edges in real-world images decrease the sharpness of peaks, edge-field images that include the strength and direction information of SAR images are adopted to reduce the diffusion of peaks and improve the detection accuracy. Experimental studies show that RLE works independently, is robust to noise, has low computational complexity, achieves high true-positive detection rates, and yields satisfactory detection precision.

Proof-of-concept robot platform for exploring automated harvesting of sugar snap peas

Currently, sugar snap peas are harvested manually. In high-cost countries like Norway, such a labour-intensive practise implies particularly large costs for the farmer. Hence, automated alternatives are highly sought after. This project explored a concept for robotic autonomous identification and tracking of sugar snap pea pods. The approach was based on a combination of visible–near infrared reflection measurements and image analysis, along with visual servoing. A proof-of-concept harvesting platform was implemented by mounting a robotic arm with hand-mounted sensors on a mobile unit. The platform was tested under plastic greenhouse conditions on potted plants of the sugar snap pea variety Cascadia using LED-lights and a partial shade. The results showed that it was feasible to differentiate the pods from the surrounding foliage using the light reflection at the spectral range around 970 nm combined with elementary image segmentation and shape modelling methods. The proof-of-concept harvesting platform was tested on 48 representative agricultural environments comprising dense canopy, varying pod sizes, partial occlusions and different working distances. A set of 104 images were analysed during the teleoperation experiment. The true positive detection rate was 93 and 87% for images acquired at long distances and at close distances, respectively. The robot arm achieved a success rate of 54% for autonomous visual servoing to a pre-grasp pose around targeted pods on 22 untouched scenarios. This study shows the potential of developing a prototype robot for semi-automated sugar snap pea harvesting.

Automatic detection of new tumors and tumor burden evaluation in longitudinal liver CT scan studies.

Radiological longitudinal follow-up of liver tumors in CT scans is the standard of care for disease progression assessment and for liver tumor therapy. Finding new tumors in the follow-up scan is essential to determine malignancy, to evaluate the total tumor burden, and to determine treatment efficacy. Since new tumors are typically small, they may be missed by examining radiologists. We describe a new method for the automatic detection and segmentation of new tumors in longitudinal liver CT studies and for liver tumors burden quantification. Its inputs are the baseline and follow-up CT scans, the baseline tumors delineation, and a tumor appearance prior model. Its outputs are the new tumors segmentations in the follow-up scan, the tumor burden quantification in both scans, and the tumor burden change. Our method is the first comprehensive method that is explicitly designed to find new liver tumors. It integrates information from the scans, the baseline known tumors delineations, and a tumor appearance prior model in the form of a global convolutional neural network classifier. Unlike other deep learning-based methods, it does not require large tagged training sets. Our experimental results on 246 tumors, of which 97 were new tumors, from 37 longitudinal liver CT studies with radiologist approved ground-truth segmentations, yields a true positive new tumors detection rate of 86 versus 72% with stand-alone detection, and a tumor burden volume overlap error of 16%. New tumors detection and tumor burden volumetry are important for diagnosis and treatment. Our new method enables a simplified radiologist-friendly workflow that is potentially more accurate and reliable than the existing one by automatically and accurately following known tumors and detecting new tumors in the follow-up scan.

Prostate fiducial marker detection with the use of multi-parametric magnetic resonance imaging

Abstract Background and purpose The introduction of a magnetic resonance (MR)-only workflow in radiotherapy requires that fiducial markers, used for position verification, can be detected on MR images. Here we evaluate a model for marker detection in prostate cancer patients by combining information from our hospital standard multi-parametric (mp-) MRI protocol (T1-weighted – T1w, T2-weighted – T2w, B 0 ) with dedicated sequences (balanced steady-state free precession sequence – bTFE, susceptibility weighted imaging – SWI). Materials and methods Thirty two patients scheduled for external-beam radiotherapy received a mp-MRI and computed-tomography; the latter was used as ground truth location of the markers. A logistic regression model was implemented for marker detection by combining features from all imaging sequences. The performance of the individual and combined sequences was assessed by determining true and false positive detections. Results The combination of different sequences (mp-MRI) resulted in a better performance than the best imaging sequence alone (bTFE). Combining mp-MRI+bTFE resulted in good accuracy and a true positive detection rate of 0.94. Conclusions The standard mp-MRI provides valuable information to detect fiducial markers. The combination of different sequences outperforms the use of a single dedicated sequence. We recommend the addition of a bTFE to the standard mp-MRI protocol to improve fiducial marker detection.

Chroma key background detection for digital video using statistical correlation of blurring artifact

Video editing software can be used to combine different videos into one video using the chroma key feature. However, detecting such video manipulation is an understudied topic. Digital forgers may present a manipulated video from chroma key composition as evidence in court, thus creating a severe problem. In this paper, we propose a blind detection technique designed to analyze the statistical correlation of blurring artifact extracted from a digital video. This technique allows us to detect chroma key forgery in a manipulated composite video. Findings from our experiments demonstrate that the proposed detection technique can adequately determine the presence of manipulation on a video using chroma key, with a true positive detection rate of 91.12% and a false positive detection rate of 1.95%.

A Wireless Headstage for Combined Optogenetics and Multichannel Electrophysiological Recording

This paper presents a wireless headstage with real-time spike detection and data compression for combined optogenetics and multichannel electrophysiological recording. The proposed headstage, which is intended to perform both optical stimulation and electrophysiological recordings simultaneously in freely moving transgenic rodents, is entirely built with commercial off-the-shelf components, and includes 32 recording channels and 32 optical stimulation channels. It can detect, compress and transmit full action potential waveforms over 32 channels in parallel and in real time using an embedded digital signal processor based on a low-power field programmable gate array and a Microblaze microprocessor softcore. Such a processor implements a complete digital spike detector featuring a novel adaptive threshold based on a Sigma-delta control loop, and a wavelet data compression module using a new dynamic coefficient re-quantization technique achieving large compression ratios with higher signal quality. Simultaneous optical stimulation and recording have been performed in-vivo using an optrode featuring 8 microelectrodes and 1 implantable fiber coupled to a 465-nm LED, in the somatosensory cortex and the Hippocampus of a transgenic mouse expressing ChannelRhodospin (Thy1::ChR2-YFP line 4) under anesthetized conditions. Experimental results show that the proposed headstage can trigger neuron activity while collecting, detecting and compressing single cell microvolt amplitude activity from multiple channels in parallel while achieving overall compression ratios above 500. This is the first reported high-channel count wireless optogenetic device providing simultaneous optical stimulation and recording. Measured characteristics show that the proposed headstage can achieve up to 100% of true positive detection rate for signal-to-noise ratio (SNR) down to 15 dB, while achieving up to 97.28% at SNR as low as 5 dB. The implemented prototype features a lifespan of up to 105 minutes, and uses a lightweight (2.8 g) and compact [Formula: see text] rigid-flex printed circuit board.

Extracting and reasoning about implicit behavioral evidences for detecting fraudulent online transactions in e-Commerce

With the explosive growth of e-Commerce worldwide, there are also growing concerns about collusive fraudulent transaction attacks in e-Commerce. The main contribution of our research work is the design of a novel detection framework that can reason about implicit online user behavior for detecting collusive fraudulent transactions. Based on real transactional and user behavioral data collected from one of the largest e-Commerce platforms in the world, our experimental results confirm that the proposed detection framework can achieve an average true positive detection rate of 83% while the false alarm rate is kept at as low as 2.4%. To the best of our knowledge, this is one of the largest scale studies toward the detection of fraudulent transactions in e-Commerce. The managerial implication of our study is that administrators of e-Commerce platforms can apply our framework to detect and prevent fraudulent transaction attacks, and hence fair electronic trading is upheld in the ever expanding e-Commerce world.

Extracting Line Features in SAR Images Through Image Edge Fields

Conventional line detection methods are mainly based on the binary edge map. This letter proposes a new line detection method that directly extracts line features from the image edge fields of the synthetic aperture radar (SAR) images. In the proposed method, the strength and direction of each field point are first obtained using a ratio-based edge filter. Then, the accumulation weight of the field point is jointly computed using its strength and direction. The direction of a field point on the line is essentially the orientation of the line. Furthermore, a field point on a strong line should be distinguished from a field point on a weak line. Thus, the accumulation weights of different field points are not equal. By summing up the accumulation weights, the straight lines in the SAR image space are directly converted into several local peaks in the parameter space. A sort-window peak detection method is proposed to suppress the spurious secondary peaks in the parameter space. The experimental results show that the proposed line detection method is robust to noise and has a good antiocclusion ability. The proposed method performs well in terms of true positive detection rate and detection accuracy for both synthetic and real-world images.