Radiologists' Relationship to Convolutional Neural Networks

Convolutional neural networks (CNNs) can be trained to analyse visual imagery a lot easier than any other artificial neural networks, making them especially important to the future of radiology, according to a new analysis published in Current Problems in Diagnostic Radiology.

“A major strength of the CNN architecture is that no explicit feature identification is required,” wrote authors Rebecca J. Mieloszyk, PhD, Puneet Bhargava, MD, of the department of radiology at the University of Washington School of Medicine in Seattle. “We need not specify that houses have straight edges and sloped rooves in order for a CNN to learn to recognize them. Rather than defining specific features or rules for image classification, CNN design decisions include sizing of the input image, depth of the network, and the classification loss function. These are much more readily tuned than hand-crafted feature vectors.”

CNNs have already been trained to identify common objects and animals such as houses, cats and dogs, similar to the way smartphone users can search their photo library for these terms. While there are still a few errors from time to time, overall the technology has proved to be extremely accurate. This is fantastic news for radiologists, as explained by Mieloszyk and Bhargava:

“Projects have demonstrated CNN use in image acquisition, segmentation, nodule detection, captioning, and classification tasks… CNN-based tools have been used to reliably reconstruct magnetic resonance images from k-space. Automated segmentation methods using CNNs have also been investigated in several radiology settings. These approaches represent a significant improvement over the typically supervised atlases or rule-based guidelines necessary to train machine-assisted medical image segmentation tools.”

Eastmed Radiology in Auckland is also excited about this new technology. As a clinic that offers a comprehensive range of x-rays including hand x-ray, and ultrasound scans, they understand the significance of artificial intelligence in medical diagnostics in the near future.