Case-Based Reasoning (CBR) is a form of analogical reasoning in which the information for a (new) query case is determined based on the known cases in a database with established information. While deep machine learning techniques of AI have demonstrated state of the art results in many fields, their transparency status of those hidden layers have cast double in many applications, especially in the medical field, where clinicians need to know the reasons of decision making delegated by a computer system. This study aims to provide a visual explanation while performing classification of endoscopic oesophageal videos. Towards this end, this work integrates the interpretation and decision making together by producing a set of profiles that in appearance resemble the training samples and hence explain the outcome of classification,in an attempt to allay the concerns that using a different model to explain the predictions while employing varying priors from the original network. Furthermore, different from many explainable networks that highlight key regions or points of the input that activate the network, this work is based on whole training images, i.e. case-based, where each training image belongs to one of the classes. Preliminary results have demonstrated the classification accuracy of 95% for training and 75% for testing while applying 500 training data (with 10% for testing split randomly) for each of three classes of ‘cancer’, ‘high grade’ and ‘suspicious’ of oesophageal squamous cancer from endoscopy videos. When training with 2000 samples for the two classes of ‘high grade’ and ‘suspicious’, the testing result delivers an accuracy of 77%, implying the impact of sample sizes. Future work includes collection of large annotated datasetand improving classification accuracy.
|Title of host publication||Proceedings of the 24th UK Symposium on Case-Based Reasoning|
|Publication status||Accepted/In press - 22 Nov 2019|
- deep learning
- visual recognition