Multi-Task Distributed Learning Using Vision Transformer With Random Patch Permutation

Article Computer Science, Artificial Intelligence

Multi-task vision transformer using low-level chest X-ray feature corpus for COVID-19 diagnosis and severity quantification

Sangjoon Park et al.

Summary: The development of a novel Multi-task ViT model enables the diagnosis and severity quantification of COVID-19 using CXR data with outstanding performance and generalization capability.

MEDICAL IMAGE ANALYSIS (2022)

添加到收藏夹

Article Computer Science, Information Systems

Privacy-preserving inpainting for outsourced image

Fang Cao et al.

Summary: In this article, a framework of privacy-preserving inpainting for outsourced image and an encrypted-image inpainting scheme are proposed. Experimental results show that our privacy-preserving inpainting framework can be effectively applied in secure cloud computing, and the proposed encrypted-image inpainting scheme achieves comparable visual quality of inpainted results with some typical inpainting schemes in plaintext domain.

INTERNATIONAL JOURNAL OF DISTRIBUTED SENSOR NETWORKS (2021)

添加到收藏夹

Article Computer Science, Artificial Intelligence

BS-Net: Learning COVID-19 pneumonia severity on a large chest X-ray dataset

Alberto Signoroni et al.

Summary: This study presents an end-to-end deep learning architecture for predicting the degree of lung compromise in COVID-19 patients on Chest X-ray images. The proposed semi-quantitative scoring system shows significant prognostic value and outperforms human annotators. The BS-Net demonstrates high accuracy and self-attentive behavior, showcasing its potential for computer-assisted monitoring in clinical settings.

MEDICAL IMAGE ANALYSIS (2021)

添加到收藏夹

Proceedings Paper Computer Science, Information Systems

Unleashing the Tiger: Inference Attacks on Split Learning

Dario Pasquini et al.

Summary: This study exposes vulnerabilities in the split learning protocol, demonstrating its insecurity against attacks from malicious servers and clients. The attacks have been tested on various datasets and threat scenarios, showing that they can overcome security defenses.

CCS '21: PROCEEDINGS OF THE 2021 ACM SIGSAC CONFERENCE ON COMPUTER AND COMMUNICATIONS SECURITY (2021)

添加到收藏夹

Article Engineering, Electrical & Electronic