Human Embryo Classification Using Self-Supervised Learning

Wicaksono, R. Satrio Hariomurti; Septiandri, Ali Akbar; Jamal, Ade

doi:10.1109/aidas53897.2021.9574328

Cited by 6 publications

(3 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Schirris et al [190] utilized a SimCLR-based feature extractor pretrained on histopathology tiles and extended the deep multiple instance learning (DeepMIL) [191] classification framework for homolog-ous recombination deficiency (HRD) and microsatellite instability (MSI) classification on colorectal and cancer datasets. Zhao and Zhou [192] added the fast mixed hard negative sample strategy to rapidly synthesize more hard negative samples [193] through a convex combination for training. The proposed model was pretrained in a self-supervised way on the Chest X-ray of pneumonia dataset and fine-tuned in a supervised way on the COVID-CT dataset.…”

Section: Instance-instance Contrastive Learning For Medical Image Ana...mentioning

confidence: 99%

A Review of Predictive and Contrastive Self-supervised Learning for Medical Images

et al. 2023

View full text Add to dashboard Cite

Over the last decade, supervised deep learning on manually annotated big data has been progressing significantly on computer vision tasks. But, the application of deep learning in medical image analysis is limited by the scarcity of high-quality annotated medical imaging data. An emerging solution is self-supervised learning (SSL), among which contrastive SSL is the most successful approach to rivalling or outperforming supervised learning. This review investigates several state-of-the-art contrastive SSL algorithms originally on natural images as well as their adaptations for medical images, and concludes by discussing recent advances, current limitations, and future directions in applying contrastive SSL in the medical domain.

show abstract

Section: Instance-instance Contrastive Learning For Medical Image Ana...mentioning

confidence: 99%

A Review of Predictive and Contrastive Self-supervised Learning for Medical Images

et al. 2023

View full text Add to dashboard Cite

show abstract

“…In network systems, as in many others, however, legitimate behavior is likely to shift over time or to experience occasional abnormal events that do not indicate an attack. Whenever valid behavior modifies [21], a large number of false positives are generated by non-self negative selection for such detection of intrusions [22]. Therefore, only those systems should employ negative selection on its own.…”

Section: Figure 1 Model Of Idsmentioning

confidence: 99%

Improving Data Transmission Rate with Self Healing Activation Model for Intrusion Detection with Enhanced Quality of Service

Madhuri,

Lakshmi

2023

IJRITCC

View full text Add to dashboard Cite

Several types of attacks can easily compromise a Wireless Sensor Network (WSN). Although not all intrusions can be predicted, they may cause significant damage to the network and its nodes before being discovered. Due to its explosive growth and the infinite scope in terms of applications and processing brought about by 5G, WSN is becoming more and more deeply embedded in daily life. Security breaches, downed services, faulty hardware, and buggy software can all cripple these enormous systems. As a result, the platform becomes unmaintainable when there are a million or more interconnected devices. When it comes to network security, intrusion detection technology plays a crucial role, with its primary function being to constantly monitor the health of a network and, if any aberrant behavior is detected, to issue a timely warning to network administrators. The current network's availability and dependability are directly tied to the efficacy and timeliness of the Intrusion Detection System (IDS). An Intrusion-Tolerant system would incorporate self-healing mechanisms to restore compromised data. System attributes such as readiness for accurate service, supply identical and correct data, confidentiality, and availability are necessary for a system to merit trust. In this research, self-healing methods are considered that can detect intrusions and can remove with intellectual strategies that can make a system fully autonomous and fix any problems it encounters. In this study, a new architecture for an Intrusion Tolerant Self Healing Activation Model for Improved Data Transmission Rate (ITSHAM-IDTR) is proposed for accurate detection of intrusions and self repairing the network for better performance, which boosts the server's performance quality and enables it to mend itself without any intervention from the administrator. When compared to the existing paradigm, the proposed model performs in both self-healing and increased data transmission rates..

show abstract

“…Residual Neural Network (ResNet) variance, Proposed by He et al [12] in 2016, is the most frequently used architecture as a base network [19]. A recent study done by Wicaksono et al [20]. used ResNet50 as a base network to compare multiple contrastive learning methods.…”

Section: Related Workmentioning

confidence: 99%

Multi-Class Intrusion Detection Using Two-Channel Color Mapping in IEEE 802.11 Wireless Network

et al. 2022

Self Cite

View full text Add to dashboard Cite

The rise of interconnected devices through wireless networks provides two sides consequences. On one side, it helps many human tasks; on the other hand, the prone wireless medium opens the vulnerable system to be exploited by adversaries. An Intrusion Detection System (IDS) is one method to inspect the network traffic by leveraging state-of-the-art anomaly detection techniques. Deep learning models have been utilized to distinguish the benign and malicious traffic. However, projecting the tabular data into images before the image classification has been the main challenge of leveraging deep learning for IDS purposes. We propose the novel projection of tabular data into 2-coded color mapping for IDS purposes. The proposed method employs a feature selection method to ensure optimal dimensionality. We examined the different number of attribute subsets to obtain the relationship between the attributes. Furthermore, it takes advantage of the Convolutional Neural Network (CNN) model to classify the Wi-Fi attacks. We evaluate the proposed model using the most common Wi-Fi attacks dataset, Aegean Wi-Fi Intrusion Dataset (AWID2). The proposed method achieved an F1 score of 99.73% and a false positive rate of 0.24%. This study highlights the importance of addressing the mapping procedures from tabular data into grid-based data before deep learning training and validates the effectiveness of CNN to detect multiple types of wireless network attacks.

show abstract

Human Embryo Classification Using Self-Supervised Learning

Cited by 6 publications

References 10 publications

A Review of Predictive and Contrastive Self-supervised Learning for Medical Images

A Review of Predictive and Contrastive Self-supervised Learning for Medical Images

Improving Data Transmission Rate with Self Healing Activation Model for Intrusion Detection with Enhanced Quality of Service

Multi-Class Intrusion Detection Using Two-Channel Color Mapping in IEEE 802.11 Wireless Network

Contact Info

Product

Resources

About