Integrating CNN along with FAST descriptor for accurate retrieval of medical images with reduced error probability

Article Computer Science, Theory & Methods

A Prior-mask-guided Few-shot Learning for Skin Lesion Segmentation

Junsheng Xiao et al.

Summary: This study proposes a few-shot segmentation network for skin lesion segmentation, which allows for quick and accurate identification of skin lesions, thereby improving patient survival rate.

COMPUTING (2023)

Add to Collection

Article Computer Science, Information Systems

Discriminative Cervical Lesion Detection in Colposcopic Images With Global Class Activation and Local Bin Excitation

Tingting Chen et al.

Summary: In this study, a new cervical lesion detection method for colposcopic images is proposed, which tackles the challenges posed by the specific characteristics of these images. The method utilizes novel feature enhancing mechanisms and achieves more accurate localization and diagnosis. Experimental results demonstrate the superiority of the proposed method over state-of-the-art models on four widely used metrics.

IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS (2022)

Add to Collection

Article Computer Science, Artificial Intelligence

Parallel deep convolutional neural network for content based medical image retrieval

P. Haripriya et al.

Summary: DICOM images play a critical role in healthcare systems for diagnosis and prognosis. Fast retrieval of historical DICOM images based on query images is important for laboratories and doctors. Addressing the issue of imbalanced data sets can improve model accuracy, and different feature extractions may help in this regard.

JOURNAL OF AMBIENT INTELLIGENCE AND HUMANIZED COMPUTING (2021)

Add to Collection

Article Computer Science, Artificial Intelligence

Deep Fuzzy Hashing Network for Efficient Image Retrieval

Huimin Lu et al.

Summary: Our proposed deep fuzzy hashing network (DFHN) combines fuzzy logic technique and DNN to learn effective binary codes that leverage fuzzy rules to model data uncertainties. The generalized hamming distance derived from fuzzy logic theory is utilized in the convolutional and fully connected layers to model outputs, resulting in competitive retrieval accuracy and efficient training speed on large-scale image datasets: CIFAR-10 and NUS-WIDE, compared to state-of-the-art deep hashing approaches.

IEEE TRANSACTIONS ON FUZZY SYSTEMS (2021)

Add to Collection

Article Computer Science, Artificial Intelligence

Recent developments of content-based image retrieval (CBIR)

Xiaoqing Li et al.

Summary: With the advancement of Internet technology and digital devices, Content-Based Image Retrieval (CBIR) has rapidly developed and been widely applied. This paper surveyed the fast developments and applications of CBIR theories and algorithms from 2009 to 2019, focusing on technological advancements in image representation and database search, as well as practical applications in various fields.

NEUROCOMPUTING (2021)

Add to Collection

Article Computer Science, Information Systems

Query-by-example HDR image retrieval based on CNN

Raoua Khwildi et al.

Summary: This paper studies the application of Convolutional Neural Networks (CNN) in HDR image retrieval, with experimental results showing satisfactory performance of CNN descriptors for describing HDR images. Features after Perceptually Uniform encoding are more descriptive than directly extracted features, and the proposed method demonstrates superior performance compared to the latest state-of-the-art technique.

MULTIMEDIA TOOLS AND APPLICATIONS (2021)

Add to Collection

Article Computer Science, Information Systems

A Weakly Supervised Semantic Segmentation Network by Aggregating Seed Cues: The Multi-Object Proposal Generation Perspective

Junsheng Xiao et al.

Summary: This article proposes an innovative three-step framework to address the issue of obtaining high-confidence discriminative regions from multi-label classification networks, and the experiments demonstrate its effectiveness compared to other baseline image segmentation methods.

ACM TRANSACTIONS ON MULTIMEDIA COMPUTING COMMUNICATIONS AND APPLICATIONS (2021)

Add to Collection

Article Computer Science, Information Systems

A Deep Learning Approach for Colonoscopy Pathology WSI Analysis: Accurate Segmentation and Classification

Ruiwei Feng et al.

Summary: The paper introduces a new framework for colonoscopy pathology whole slide image (WSI) analysis, including lesion segmentation and tissue diagnosis. The framework incorporates specific sampling and transfer learning strategies based on the characteristics of colonoscopy pathology WSI, as well as a specific loss function for training the segmentation network.

IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS (2021)

Add to Collection

Article Geography, Physical

An attention-fused network for semantic segmentation of very-high-resolution remote sensing imagery

Xuan Yang et al.

Summary: This paper proposes a multipath attention-fused network structure to address feature fusion challenges in semantic segmentation of remote sensing images. By fusing high-level abstract features and low-level spatial features, the network achieves state-of-the-art performance on two 2D datasets.

ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING (2021)

Add to Collection

Article Geography, Physical

Urban scene understanding based on semantic and socioeconomic features: From high-resolution remote sensing imagery to multi-source geographic datasets

Yu Su et al.

Summary: Scene classification in high-resolution remote sensing imagery provides essential information for urban planning and monitoring. However, traditional methods may overlook geographic boundaries and suffer from the mosaic effect. This paper proposes a universal urban scene understanding framework based on multi-source geographic data, which integrates various data sources to determine urban scene categories.

ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING (2021)

Add to Collection

Article Biochemical Research Methods

A Transfer Learning Based Super-Resolution Microscopy for Biopsy Slice Images: The Joint Methods Perspective

Jintai Chen et al.

Summary: This paper proposes a joint framework consisting of a transfer learning strategy and a deep super-resolution framework for generating high resolution biopsy slice images. Experimental results show that SRFBN+ performs well in generating super-resolution images and CF-Trans is an efficient transfer learning strategy.

IEEE-ACM TRANSACTIONS ON COMPUTATIONAL BIOLOGY AND BIOINFORMATICS (2021)

Add to Collection

Review Radiology, Nuclear Medicine & Medical Imaging