Classification of pulsar signals using ensemble gradient boosting algorithms based on asymmetric under-sampling method

Tariq

et al. 2022

Journal of Oncology

Lung cancer is the deadliest cancer killing almost 1.8 million people in 2020. The new cases are expanding alarmingly. Early lung cancer manifests itself in the form of nodules in the lungs. One of the most widely used techniques for both lung cancer early and noninvasive diagnosis is computed tomography (CT). However, the intensive workload of radiologists to read a large number of scans for nodules detection gives rise to issues like false detection and missed detection. To overcome these issues, we proposed an innovative strategy titled adaptive boosting self-normalized multiview convolution neural network (AdaBoost-SNMV-CNN) for lung cancer nodules detection across CT scans. In AdaBoost-SNMV-CNN, MV-CNN function as a baseline learner while the scaled exponential linear unit (SELU) activation function normalizes the layers by considering their neighbors' information and a special drop-out technique (α-dropout). The proposed method was trained and tested using the widely Lung Image Database Consortium and Image Database Resource Initiative (LIDC-IDRI) and Early Lung Cancer Action Program (ELCAP) datasets. AdaBoost-SNMV-CNN achieved an accuracy of 92%, sensitivity of 93%, and specificity of 92% for lung nodules detection on the LIDC-IDRI dataset. Meanwhile, on the ELCAP dataset, the accuracy for detecting lung nodules was 99%, sensitivity 100%, and specificity 98%. AdaBoost-SNMV-CNN outperformed the majority of the model in accuracy, sensitivity, and specificity. The multiviews confer the model’s good generalization and learning ability for diverse features of lung nodules, the model architecture is simple, and has a minimal computational time of around 102 minutes. We believe that AdaBoost-SNMV-CNN has good accuracy for the detection of lung nodules and anticipate its potential application in the noninvasive clinical diagnosis of lung cancer. This model can be of good assistance to the radiologist and will be of interest to researchers involved in the designing and development of advanced systems for the detection of lung nodules to accomplish the goal of noninvasive diagnosis of lung cancer.

Section: Resultsmentioning

confidence: 99%

“…Additionally, we define the true positive rate (TPR) and false positive rate (FPR) as described in equations ( 13) and (14), respectively. e TPR and FPR are used to plot the receiver operating characteristic (ROC) curve, which was used to evaluate the trained models on both datasets [30].…”

Section: Performance Metricsmentioning

confidence: 99%

Lung Cancer Nodules Detection via an Adaptive Boosting Algorithm Based on Self-Normalized Multiview Convolutional Neural Network

Tariq

et al. 2022

Journal of Oncology

Proceedings of the Eighth International Conference on Research in Intelligent Computing in Engineering

“…2) Classification of Pulsars using Extreme Gradient Boosting and Light Gradient Boosting Tariq, et al [10] used HTRU2 [11] and LOTAAS -1 datasets. For handling the class imbalance problem, asymmetric under sampling method was applied.…”

Section: E Classification Of Pulsars 1) Machine Learning For Classify...mentioning

confidence: 99%

Emerging Trends In Pulsar Star Studies: A Synthesis Of Machine Learning Techniques In Pulsar Star Research

Thanu,

Subha

2022

The pulsar is an extremely magnetized gyrating neutron star having a radius of 10 -15 km. Pulsars provide the indirect evidence of the gravitational wave's existence. So, to study the gravitational waves identification of pulsars is mandatory. Pulsars are considered as the Universe's gift. Pulsars provide scientists and researchers with information of the physics of neutron stars, which are thought to be the densest materials in the universe. The reason why astronomers give importance to the pulsars, because they are the leading edge of the research, based on the gravity. All pulsars produce marginally distinct emission pattern and it varies to some extent with every rotation. Hence, a promising signal detection is termed as a candidate, which is averaged based on every rotation of the pulsars. Any absence of the additional information, implies that each candidate is a real pulsar. The valid signals are extremely hard to detect due to noise and radio frequency interference (RFI). To clear up with this issue, Machine Learning (ML) algorithms were used for automatically classifying, identifying and many other process of pulsar candidates. This survey paper talks about different techniques used by different researchers for the pulsar star classification, identification and still more, using ML techniques.

Res. Astron. Astrophys.

Research on Pulsar Time Steered Atomic Time Algorithm Based on DPLL

Zheng,

Liu,

Shen

et al. 2024

In today's society, there is a wide demand for high-precision and high-stability time service in the fields of electric power, communication, transportation and finance. At present, the time standard in various countries is mainly based on atomic clocks, but the frequency drift of atomic clocks will affect the long-term stability performance. Compared with atomic clocks, millisecond pulsars have better long-term stability and can complement with the excellent short-term stability of atomic clocks. In order to improve the long-term stability of the atomic time scale, and then improve the timing accuracy, this paper proposes an algorithm for steering the Atomic clock ensemble (ACE) by Ensemble pulsar time (EPT) based on Digital phase locked loop (DPLL). Firstly, the ACE and EPT are generated by ALGOS algorithm, then the ACE is steered by EPT based on DPLL to calibrate the long-term frequency drift of the atomic clock, so that the generated steered atomic time follows both the short-term stability characteristics of ACE and the long-term stability characteristics of EPT, and finally, the steered atomic time is used to calibrate the local cesium clock. The experimental results show that the long-term stability of atomic time after steering is improved by 2 orders of magnitude compared with that before steering, and the daily drift of local cesium clock after calibration is less than 9.47ns in 3 years, 3 orders of magnitude higher than that before calibration on accuracy.