Training and validation of a novel non-invasive imaging system for ruling out malignancy in canine subcutaneous and cutaneous masses using machine learning in 664 masses

Access the article

Dank, G. et al (2023) Training and validation of a novel non-invasive imaging system for ruling out malignancy in canine subcutaneous and cutaneous masses using machine learning in 664 masses. Frontiers in Veterinary Science, 10, p.1164438. https://doi.org/10.3389/fvets.2023.1164438

Summary

This study reports the training and validation of a test to rule out malignancy in cutaneous and subcutaneous masses using an imaging system. The training and validation process used machine learning, which is a type of artificial intelligence using algorithms and statistical models to enable computers to “learn” and make predictions without being explicitly programmed to do so.

Machine learning can be subdivided based on whether the data has already been labelled (supervised learning) or whether the machine is left to identify patterns in the data (unsupervised learning). Deep learning is a further type of machine learning which use neural networks, with multiple layers to analyse complex relationships and patterns in data.

The algorithms are ‘trained’ on large data sets to identify patterns and associations and then use this information to make predictions about new data sets. They should always be validated on a second set of data which should be representative of the range of cases on which they will be used in practice.

The use of machine learning is becoming more common, especially in the interpretation of diagnostic imaging.  Although the methodology may be new when critically evaluating a paper the primary aim is to assess the diagnostic accuracy of a test in order to assess whether it will be useful in guiding diagnosis in your practice.

A general checklist for evaluating a diagnostic study is available.

Support with assessing the relevance of this paper

We have produced a guide that you can use to work through this paper as a practice.  For those less confident in assessing published papers there is a partially completed guide.

For those new to journal clubs, we have a step-by-step guide you might find useful: Setting up and running a journal club