Deep Learning based on MRI for Differentiation of Low- and High-Grade in Low-Stage Renal Cell Carcinoma

The Keras library 32 with Tensorflow backend 33 was used. Our models were trained on a computer with two NVidia Quadro V100 GPU. The code is publicly available on Github at https://github.com/marcelc21399/Fuhrman_repo.

Deep Learning to Distinguish Benign from Malignant Renal Lesions based on Routine MR Imaging

The implementation of our deep learning model was based on the Keras package (45) with the Tensorflow library as our backend (46). Our models were trained on a computer with two NVidia V100 GPUs. To allow other researchers to develop their models, the code is publicly available on Github at https://github.com/intrepi dlemon/renal-mri. The implementation of the radiomics feature extraction was based on “radiomics-develop” package from the Naqa lab in McGill University (47, 48). This code is available for public use on Github at https://github.com/mvallieres/radiomicsdevelop. The radiomics pipeline was developed using Python’s sklearn package. This code is publicly available at https://github. com/subhanik1999/Radiomics-ML.

Covid Diagnosis Model

http://radiology-ai-env.eba-wgmpba4k.us-west-2.elasticbeanstalk.com/covid/

The following model is a convolutional neural network (CNN) that seeks to diagnose COVID-19 from a chest radiograph (CXR). It employs the EfficientNet-B0 architecture and is trained using a collection of over 2500 organically sourced CXRs from the University of Pennsylvania Health System and Rhode Island Hospital, as well as over 8500 publicly sourced CXRs. The model reports a ~91% mean accuracy on an internal test set and ~86% accuracy on an external test set. Our analysis demonstrates the effectiveness and robustness of the model’s ability to detect COVID-19 in chest radiographs.

To use the COVID-19 diagnosis model, the user must upload a chest radiograph as a DICOM image file and click submit. The image will be processed automatically so that the model outputs an evaluation only based on the lung segmentation of the radiograph.