An Efficient CNN for Hand X-Ray Overall Scoring of Rheumatoid Arthritis

Article Computer Science, Artificial Intelligence

Multi-modality medical image fusion technique using multi-objective differential evolution based deep neural networks

Manjit Kaur et al.

Summary: The proposed approach utilizes NSCT and Xception for feature extraction and fusion of multi-modality medical images, outperforming competitive approaches in experimental results.

JOURNAL OF AMBIENT INTELLIGENCE AND HUMANIZED COMPUTING (2021)

Add to Collection

Article Computer Science, Hardware & Architecture

DNetUnet: a semi-supervised CNN of medical image segmentation for super-computing AI service

Kuo-Kun Tseng et al.

Summary: In this paper, a new convolutional neural network architecture named DNetUnet is proposed, which combines U-Nets with different down-sampling levels and a new dense block as feature extractor. DNetUnet is a semi-supervised learning method that enhances algorithm generalization ability from unlabelled data and improves performance on mobile platform through integrating distillation technique.

JOURNAL OF SUPERCOMPUTING (2021)

Add to Collection

Article Radiology, Nuclear Medicine & Medical Imaging

MIScnn: a framework for medical image segmentation with convolutional neural networks and deep learning

Dominik Mueller et al.

Summary: MIScnn is an open-source Python library designed to provide an intuitive API for fast construction of medical image segmentation pipelines, offering high configurability and multiple open interfaces for full pipeline customization. By running cross-validation on the Kidney Tumor Segmentation Challenge 2019 data set, MIScnn demonstrated powerful predictive capabilities based on the standard 3D U-Net model, showing that researchers can quickly set up a complete medical image segmentation pipeline with just a few lines of code.

BMC MEDICAL IMAGING (2021)

Add to Collection

Article Cell Biology

Wavelet decomposition facilitates training on small datasets for medical image classification by deep learning

Axel H. Masquelin et al.

Summary: Utilizing multi-level DWT composition instead of early convolutional layers in a DNN may improve image classification, reduce the number of network parameters requiring evaluation, and result in a substantially faster convergence rate. Multi-level DWT performed better than convolutional layers when utilizing multiple kernel resolutions.

HISTOCHEMISTRY AND CELL BIOLOGY (2021)

Add to Collection

Article Rheumatology