Nordic Machine Intelligence

Two-stage mammography classification model using explainable-AI for ROI detection

Fri, 17 Nov 2023 00:00:00 +0100

This study introduces an enhanced version of a two-stage modelling approach using artificial intelligence (AI) for breast cancer detection in mammography screening. Leveraging a large dataset of 2,863,175 mammograms from the Norwegian breast cancer screening program, the approach uses two convolutional neural networks. A key enhancement over the prior methodology is the application of the explainable-AI method Layered GradCam for identifying regions of interest (ROIs) within the mammograms. The second neural network subsequently classifies these ROIs for malignancy. Layered GradCam is also used to display identified cancers to the user. By the AUC criterion, our model performs well, 0.974 for screen-detected and 0.931 for all cancers (screen-detected and interval), compared to a commercial program; 0.959 and 0.918, respectively. Comparisons with the radiologist cores indicates that the model has equal performance with two radiologists, and superior performance to one, for the detection of all cancers (screening- and interval type). Our tests indicate that our model generalizes well for different breast centers, but so far only images from a single manufacturer have been tested.

Deep Graph Learning for Molecules and Materials

Hannes Kneiding, David Balcells — Thu, 21 Dec 2023 00:00:00 +0100

Machine learning approaches have become an important tool in chemistry and materials science for the accurate and efficient prediction of physical properties. Most notably among them are graph neural networks that leverage the inherent graph structure of molecules and materials in order to achieve state-of-the-art accuracy. In this perspective we give a brief introduction to the theoretical foundations of graph neural networks for molecular structures and their specific applications in chemistry and materials science. We conclude with a short outlook discussing remaining research questions as well as opportunities for further developments of the field.

Frequency-Histogram Coarse Graining in Elementary and 2-Dimensional Cellular Automata

Sanyam Jain, Stefano Nichele — Wed, 22 May 2024 00:00:00 +0200

Cellular automata and other discrete dynamical systems have long been studied as models of emergent complexity. Recently, neural cellular automata have been proposed as models to investigate the emerge of a more general artificial intelligence, thanks to their propensity to support properties such as self-organization, emergence, and open-endedness. However, understanding emergent complexity in large scale systems is an open challenge. How can the important computations leading to emergent complex structures and behaviors be identified? In this work, we systematically investigate a form of dimensionality reduction for 1-dimensional and 2-dimensional cellular automata based on coarse-graining of macrostates into smaller blocks. We discuss selected examples and provide the entire exploration of coarse graining with different filtering levels in the appendix (available also digitally at this link: https://s4nyam.github.io/eca88/. We argue that being able to capture emergent complexity in AI systems may pave the way to open-ended evolution, a plausible path to reach artificial general intelligence.