Automating the Clock Drawing Test with Deep Learning and Saliency Maps

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

The Clock Drawing Test (CDT) is an important tool in the diagnosis of Cognitive Decline (CD). Using Deep Learning (DL), this test can be automated with a high degree of accuracy, more so where the medium of recording allows the use of temporal information on how the clock was drawn which may not be accessible to clinicians in traditional screening. The high-risk nature of this field makes understanding the reasoning for automated results imperative. A model’s reasoning can often be described using saliency maps, however, there are a number of different methods for generating such maps. Therefore, we propose a methodology to train a DL classifier for use in the CDT which incorporates temporal information and use saliency maps to explain classification predictions. We find that our classifier achieves scores above 98% with F1 for clocks and over 96% F1 on average across a test set of 18 different classes. Our methodology also shows that Integrated Gradients using SmoothGrad produce the best saliency map results visually and statistically.
Original languageEnglish
Title of host publicationProgress in Artificial Intelligence
Subtitle of host publication23rd EPIA Conference on Artificial Intelligence, EPIA 2024, Viana do Castelo, Portugal, September 3–6, 2024, Proceedings, Part II
EditorsManuel Filipe Santos, José Machado, Paulo Novais, Paulo Cortez, Pedro Miguel Moreira
PublisherSpringer
Chapter8
Pages86-97
Number of pages12
ISBN (Electronic)978-3-031-73500-4
ISBN (Print)978-3-031-73499-1
DOIs
Publication statusPublished - 2025

Publication series

NameLecture Notes in Computer Science
Volume14968
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Keywords

  • Cognitive Decline
  • Deep Learning
  • eXplainable Artificial Intelligence

Cite this