Terms Of False Negative Rate Research Articles

Refactoring and performance analysis of the main CNN architectures: using false negative rate minimization to solve the clinical images melanoma detection problem

BackgroundMelanoma is one of the deadliest tumors in the world. Early detection is critical for first-line therapy in this tumor pathology and it remains challenging due to the need for histological analysis to ensure correctness in diagnosis. Therefore, multiple computer-aided diagnosis (CAD) systems working on melanoma images were proposed to mitigate the need of a biopsy. However, although the high global accuracy is declared in literature results, the CAD systems for the health fields must focus on the lowest false negative rate (FNR) possible to qualify as a diagnosis support system. The final goal must be to avoid classification type 2 errors to prevent life-threatening situations. Another goal could be to create an easy-to-use system for both physicians and patients.ResultsTo achieve the minimization of type 2 error, we performed a wide exploratory analysis of the principal convolutional neural network (CNN) architectures published for the multiple image classification problem; we adapted these networks to the melanoma clinical image binary classification problem (MCIBCP). We collected and analyzed performance data to identify the best CNN architecture, in terms of FNR, usable for solving the MCIBCP problem. Then, to provide a starting point for an easy-to-use CAD system, we used a clinical image dataset (MED-NODE) because clinical images are easier to access: they can be taken by a smartphone or other hand-size devices. Despite the lower resolution than dermoscopic images, the results in the literature would suggest that it would be possible to achieve high classification performance by using clinical images. In this work, we used MED-NODE, which consists of 170 clinical images (70 images of melanoma and 100 images of naevi). We optimized the following CNNs for the MCIBCP problem: Alexnet, DenseNet, GoogleNet Inception V3, GoogleNet, MobileNet, ShuffleNet, SqueezeNet, and VGG16.ConclusionsThe results suggest that a CNN built on the VGG or AlexNet structure can ensure the lowest FNR (0.07) and (0.13), respectively. In both cases, discrete global performance is ensured: 73% (accuracy), 82% (sensitivity) and 59% (specificity) for VGG; 89% (accuracy), 87% (sensitivity) and 90% (specificity) for AlexNet.

The value of sentinel lymph-node biopsy after neoadjuvant therapy: an overview.

We conducted a systematic review to analyse the performance of the sentinel lymph-node biopsy (SLNB) after the neoadjuvant chemotherapy, compared to axillary lymph-node dissection, in terms of false-negative rate (FNR) and sentinel lymph-node identification rate (SLNIR), sensitivity, negative predictive value (NPV), need for axillary lymph-node dissection (ALND), morbidity, preferences, and costs. MEDLINE, Embase, Scopus, and The Cochrane Library were searched. We assessed the quality of the included systematic reviews using AMSTAR2 tool, and estimated the degree of overlapping of the individual studies on the included reviews. Six systematic reviews with variable quality were selected. We observed a very high overlapping degree across the included reviews. The FNR and the SLNIR were quite consistent (FNR 13-14%; SLNIR ~ 90% or higher). In women with initially clinically node-negative breast cancer, the FNR was better (6%), with similar SLNIR (96%). The included reviews did not consider the other prespecified outcomes. It would be reasonable to suggest performing an SLNB in patients treated with NACT, adjusting the procedure to the previous marking of the affected lymph node, using double tracer, and biopsy of at least three sentinel lymph nodes. More well-designed research is needed. PROSPERO registration number: CRD42020114403.

An Effective Probabilistic Technique for DDoS Detection in OpenFlow Controller

Distributed denial of service (DDoS) attacks have always been a nightmare for network infrastructure for the last two decades. Existing network infrastructure is lacking in identifying and mitigating the attack due to its inflexible nature. Currently, software-defined networking (SDN) is more popular due to its ability to monitor and dynamically configure network devices based on the global view of the network. In SDN, the control layer is accountable for forming all decisions in the network and data plane for just forwarding the message packets. The unique property of SDN has brought a lot of excitement to network security researchers for preventing DDoS attacks. In this article, for the identification of DDoS attacks in the OpenFlow controller, a probabilistic technique with a central limit theorem has been utilized. This method primarily detects resource depletion attacks, for which the DARPA dataset is used to train the probabilistic model. In different attack scenarios, the probabilistic approach outperforms the entropy-based method in terms of false negative rate (FNR). The emulation results demonstrate the efficacy of the approach, by reducing the FNR by 98% compared to 78% in the existing entropy mechanism, at 50% attack rate.

Emotion Processing by Applying a Fuzzy-Based Vader Lexicon and a Parallel Deep Belief Network Over Massive Data

Emotion processing has been a very intense domain of investigation in data analysis and NLP during the previous few years. Currently, the algorithms of the deep neural networks have been applied for opinion mining tasks with good results. Among various neuronal models applied for opinion mining a deep belief network (DBN) model has gained more attention. In this proposal, we have developed a combined classifier based on fuzzy Vader lexicon and a parallel deep belief network for emotion analysis. We have implemented multiple pretreatment techniques to improve the quality and soundness of the data and eliminate disturbing data. Afterward, we have performed a semi-automatic dataset labeling using a combination of two different methods: Mamdani’s fuzzy system and Vader lexicon. As well, we have applied four feature extractors, which are: GloVe, TFIDF (Trigram), TFIDF (Bigram), TFIDF (Unigram) with the aim of transforming each incoming tweet into a digital value vector. In addition, we have integrated three feature selectors, namely: The ANOVA method, the chi-square approach and the mutual information technique with the objective of selecting the most relevant features. Further, we have implemented the DBN as classifier for classifying each inputted tweet into three categories: neutral, positive or negative. At the end, we have deployed our proposed approach in parallel way employing both Hadoop and Spark framework with the purpose of overcoming the problem of long runtime of massive data. Furthermore, we have carried out a comparison between our newly suggested hybrid approach and alternative hybrid models available in the literature. From the experimental findings, it was found that our suggested vague parallel approach is more powerful than the baseline patterns in terms of false negative rate (1.33%), recall (99.75%), runtime (32.95s), convergence, stability, F1 score (99.53%), accuracy (98.96%), error rate (1.04%), kappa-Static (99.1%), complexity, false positive rate (0.25%), precision rate (97.59%) and specificity rate (98.67%). As a conclusion, our vague parallel approach outperforms baseline and deep learning models, as well as certain other approaches chosen from the literature.

Rainfall-Induced Landslide Prediction Using Machine Learning Models: The Case of Ngororero District, Rwanda.

Landslides fall under natural, unpredictable and most distractive disasters. Hence, early warning systems of such disasters can alert people and save lives. Some of the recent early warning models make use of Internet of Things to monitor the environmental parameters to predict the disasters. Some other models use machine learning techniques (MLT) to analyse rainfall data along with some internal parameters to predict these hazards. The prediction capability of the existing models and systems are limited in terms of their accuracy. In this research paper, two prediction modelling approaches, namely random forest (RF) and logistic regression (LR), are proposed. These approaches use rainfall datasets as well as various other internal and external parameters for landslide prediction and hence improve the accuracy. Moreover, the prediction performance of these approaches is further improved using antecedent cumulative rainfall data. These models are evaluated using the receiver operating characteristics, area under the curve (ROC-AUC) and false negative rate (FNR) to measure the landslide cases that were not reported. When antecedent rainfall data is included in the prediction, both models (RF and LR) performed better with an AUC of 0.995 and 0.997, respectively. The results proved that there is a good correlation between antecedent precipitation and landslide occurrence rather than between one-day rainfall and landslide occurrence. In terms of incorrect predictions, RF and LR improved FNR to 10.58% and 5.77% respectively. It is also noted that among the various internal factors used for prediction, slope angle has the highest impact than other factors. Comparing both the models, LR model’s performance is better in terms of FNR and it could be preferred for landslide prediction and early warning. LR model’s incorrect prediction rate FNR = 9.61% without including antecedent precipitation data and 3.84% including antecedent precipitation data.

A Study on Impact of Dimensionality Reduction on Naïve Bayes Classifier

Objectives: The time complexity of the machine learning algorithm is directly proportionate to the dimension of the dataset. In this paper, he impacts of dimensionality of the dataset on the machine learning algorithm, Naive-Bayes Classifier is evaluated with all feature subsets to analyze whether there is any variations in the performance. Methods/Statistical Analysis: Naive Bayes Classifier is taken for the study to evaluate its variations in terms of its performance in correctly classified instances and incorrectly classified instances. Pima Indian Type II diabetes dataset is taken for the experimental study. Confusion matrix will be formulated for the performance of Naive-Bayes Classifier using 10-fold cross validation for each run. The study exhibits the impact of the dimensionality on the performance of Naive-Bayes Classifier. Findings: The Naive Bayes classifier classifies the patient records either as diabetes or as non-diabetes using the values of the feature set. It is a probabilistic approach of classifying the patient records into the binary class. It is found that there is an impact on the performance of Naive Bayes Classifier due to the dimensionality of the feature set it terms of Classification accuracy, number of true positives, true negatives, false positives and false negatives. The incorrect classification is certainly dangerous. Whereas the valid classification facilitates the healthcare systems in terms of planning effective course of treatment which will save the life of the patient. The invalid classification will lead to a wrong diagnosis while formulating the treatment plan and it will lead to loss of life. Hence, the invalid classification in terms of false negative rate is to be viewed very seriously. In this paper, the study shows that there is an impact on the performance of Naive Bayes Classifier due to the higher dimensionality of the dataset. Application/Improvements: They will be used in medical Informatics for the quality diagnosis and effective treatment planning. The focus on the false positive rate in the classification accuracy of Naive Bayes Classifier will notably help the healthcare systems to diagnose the patients accurately to save life.

An improved Android malware detection scheme based on an evolving hybrid neuro-fuzzy classifier (EHNFC) and permission-based features

The increasing number of Android devices and users has been attracting the attention of different types of attackers. Malware authors create new versions of malware from previous ones by implementing code obfuscation techniques. Obfuscated malware is potentially contributed to the exponential increase in the number of generated malware variants. Detection of obfuscated malware is a continuous challenge because it can easily evade the signature-based malware detectors, and behaviour-based detectors are not able to detect them accurately. Therefore, an efficient technique for obfuscated malware detection in Android-based smartphones is needed. In the literature on Android malware classification, few malware detection approaches are designed with the capability of detecting obfuscated malware. However, these malware detection approaches were not equipped with the capacity to improve their performance by learning and evolving their malware detection rules. Based on the concept of evolving soft computing systems, this paper proposes an evolving hybrid neuro-fuzzy classifier (EHNFC) for Android malware classification using permission-based features. The proposed EHNFC not only has the capability of detecting obfuscated malware using fuzzy rules, but can also evolve its structure by learning new malware detection fuzzy rules to improve its detection accuracy when used in detection of more malware applications. To this end, an evolving clustering method for adapting and evolving malware detection fuzzy rules was modified to incorporate an adaptive procedure for updating the radii and centres of clustered permission-based features. This modification to the evolving clustering method enhances cluster convergence and generates rules that are better tailored to the input data, hence improving the classification accuracy of the proposed EHNFC. The experimental results for the proposed EHNFC show that the proposal outperforms several state-of-the-art obfuscated malware classification approaches in terms of false negative rate (0.05) and false positive rate (0.05). The results also demonstrate that the proposal detects the Android malware better than other neuro-fuzzy systems (viz., the adaptive neuro-fuzzy inference system and the dynamic evolving neuro-fuzzy system) in terms of accuracy (90%).

Multi-task ant system for multi-object parameter estimation and its application in cell tracking

Inspired by ant's stochastic behavior in search for multiple food sources, we propose a cooperating multi-task ant system for tracking multiple synthetic objects as well as multiple real cells in a bio-medical field. In our framework, each ant colony is assumed and assigned to fulfill a given task to estimate the state of an object. Furthermore, two ant levels are used, i.e., ant individual level and ant cooperation level. In the ant individual level, ants within one colony perform independently, and the motion of each individual is probabilistically determined by both its intended motion modes and the likelihood function score. In the ant cooperation level, each ant adjusts individual state within its influence region according to heuristic information of all other ants within the same colony, while the global best template at current iteration is found among all ant colonies and utilized to update ant model probability, influence region, and probability of fulfilling task. Our algorithm is validated by comparing it to the-state-of-art algorithms, and specifically the improved tracking performance in terms of false negative rate (up to 10.0%) and false negative rate (up to 2.1%) is achieved based on the studied three real cell image sequences.

Sentinel Lymph Node Mapping in Melanoma

Management of cutaneous melanoma has changed after introduction in the clinical routine of sentinel lymph node biopsy (SLNB) for nodal staging. By defining the nodal basin status, SLNB provides a powerful prognostic information. Nevertheless, some debate still surrounds the accuracy of this procedure in terms of false-negative rate. Several large-scale studies have reported a relatively high false-negative rate (5.6%-21%), correctly defined as the proportion of false-negative results with respect to the total number of "actual" positive lymph nodes. In this review, we identified all the technical aspects that the nuclear medicine physician, the surgeon, and the pathologist should take into account to improve accuracy of the procedure and minimize the false-negative rate. In particular, SPECT/CT imaging detects more SLNs than those found by planar lymphoscintigraphy. Furthermore, the nuclear medicine community should reach a consensus on the radioactive counting rate threshold to better guide the surgeon in identifying the lymph nodes with the highest likelihood of housing metastases ("true biologic SLNs"). Analysis of the harvested SLNs by conventional techniques is also a further potential source for error. More accurate SLN analysis (eg, molecular analysis by reverse transcriptase-polymerase chain reaction) and more extensive SLN sampling identify more positive nodes, thus reducing the false-negative rate.The clinical factors identifying patients at higher-risk local recurrence after a negative SLNB include older age at diagnosis, deeper lesions, histological ulceration, and head-neck anatomic location of the primary lesion.The clinical impact of a false-negative SLNB on the prognosis of melanoma patients remains controversial, because the majority of studies have failed to demonstrate overall statistically significant disadvantage in melanoma-specific survival for false-negative SLNB patients compared with true-positive SLNB patients.When new more effective drugs will be available in the adjuvant setting for stage III melanoma patients, the implication of an accurate staging procedure for the sentinel lymph nodes will be crucial for both patients and clinicians. Standardization and accuracy of SLN identification, removal, and analysis are required.

1-D fast normalized cross-correlation using additions

A new fast algorithm for the computation of the normalized cross-correlation (NCC) is presented. For a search window of size M and a template of size N, our fast NCC requires only approximately 3N⋅(M−N+1) additions/subtractions without multiplications. Numerical results with 100,000 test signals show that the use of the fast NCC instead of the traditional approaches for the determination of the degree of similarity between a test signal and a reference signal (template) brings about a significant improvement in terms of false negative rate, identification rate and computational cost without a significant increase in false positive rate, especially when signal to noise ratio (SNR) is higher than 4 dB.

Fast Normalized Cross-Correlation

Normalized cross-correlation has been used extensively for many signal processing applications, but the traditional normalized correlation operation does not meet speed requirements for time-critical applications. In this paper, a new fast algorithm for the computation of the normalized cross-correlation (NCC) without using multiplications is presented. For a search window of size M and a template of size N the fast NCC requires only approximately 2N?(M?N+1) additions/subtractions without multiplications. Simulation results with 100,000 test signals show that the use of the fast NCC instead of the traditional approaches for the determination of the degree of similarity between a test signal and a reference signal (template) brings about a significant improvement in terms of false negative rate, identification rate and computational cost without a significant increase in false positive rate, especially when the signal-to-noise ratio (SNR) is higher than 3 dB.

APPLICATIONS AND STATISTICAL PROPERTIES OF MINIMUM SIGNIFICANT DIFFERENCE-BASED CRITERION TESTING IN A TOXICITY TESTING PROGRAM

As a follow up to the recommendations of the September 1995 SETAC Pellston Workshop on Whole Effluent Toxicity (WET) on test methods and appropriate endpoints, this paper will discuss the applications and statistical properties of using a statistical criterion of minimum significant difference (MSD). We examined the upper limits of acceptable MSDs as acceptance criterion in the case of normally distributed data. The implications of this approach are examined in terms of false negative rate as well as false positive rate. Results indicated that the proposed approach has reasonable statistical properties. Reproductive data from short-term chronic WET test with Ceriodaphnia dubia tests were used to demonstrate the applications of the proposed approach. The data were collected by the North Carolina Department of Environment, Health, and Natural Resources (Raleigh, NC, USA) as part of their National Pollutant Discharge Elimination System program.

Applications and statistical properties of minimum significant difference‐based criterion testing in a toxicity testing program

AbstractAs a follow up to the recommendations of the September 1995 SETAC Pellston Workshop on Whole Effluent Toxicity (WET) on test methods and appropriate endpoints, this paper will discuss the applications and statistical properties of using a statistical criterion of minimum significant difference (MSD). We examined the upper limits of acceptable MSDs as acceptance criterion in the case of normally distributed data. The implications of this approach are examined in terms of false negative rate as well as false positive rate. Results indicated that the proposed approach has reasonable statistical properties. Reproductive data from short‐term chronic WET test with Ceriodaphnia dubia tests were used to demonstrate the applications of the proposed approach. The data were collected by the North Carolina Department of Environment, Health, and Natural Resources (Raleigh, NC, USA) as part of their National Pollutant Discharge Elimination System program.