Biomedical Informatics Review

Biomedical Informatics Review

Share

This page focuses on applications of computing in Biology and Medicine.

This page focuses on the reporting the latest developments in computing solutions to problems in Biology and Medicine. In particular, we focus on applications of machine learning, data analytics and mining in Bioinformatics, medical image and signal analysis in automated diagnosis systems.

Operating as usual

22/05/2022

"Whole slide images are graphs": Our paper on effectiveness of graph modelling of WSIs in & Graph Neural Networks ( ) for receptor status prediction of from routine WSIs is accepted in Medical Image Analysis.
See: https://arxiv.org/abs/2110.06042

TL;DR
Motivation: Whole slide images are big and holistic modelling of interactions between different tissue components using only slide level labels is difficult with conventional "patch-then-aggregate" approaches used for weakly supervised learning in CPath.

2. We introduce a flexible based framework ( +) that can holistically model WSIs for different large scale prediction problems
3. For HER2 status prediction, the method gives AUC 0.75-0.8 and can be used for case triaging & advanced ordering of tests

4. To demonstrate the flexibility of SlideGraph, our implementation uses different local features For ER status predictionwith an AUC of 0.88
5. Once you extract local features, the method is lightning fast allowing large scale experiments (

R**T: robustness evaluation and enhancement toolbox for computational pathology 22/05/2022

Unless appropriately trained, or models in Computational Pathology can be fragile to natural or adversarial subvisual perturbations. Our paper presents Robustness Evaluation and Enhancement Toolbox (R**T) for to help address this challenge.
TL;DR
1/R**T provides a suite of algorithmic strategies for enabling robustness assessment of a trained predictive model with respect to specialized image transformations such as staining, compression, focusing/blurring, changes in spatial resolution, etc.
2/R**T enables efficient and robust training of pipelines by providing a custom implementation of Adversarial Training for Free.
3/Tutorials/implementation available.
Read: https://academic.oup.com/bioinformatics/advance-article-abstract/doi/10.1093/bioinformatics/btac315/6582557
Thanks to all authors. Fayyaz Minhas

R**T: robustness evaluation and enhancement toolbox for computational pathology AbstractMotivation. Digitization of pathology laboratories through digital slide scanners and advances in deep learning approaches for objective histological as

Fayyaz Minhas on Twitter 27/12/2021

Watch our video series on surviving survival analysis and prediction at

Fayyaz Minhas on Twitter “Interested in survival analysis and prediction? We present a tutorial on how to survive survival analysis covering concepts, coding and machine learning with a focus on its use in computational pathology! https://t.co/OI7SY50SXv Thanks to volunteer contributors! ”

Want your school to be the top-listed School/college?