Dr. Chern Hong Lim

Ph.D.

Monash University Malaysia Sdn Bhd [ Co. Reg. No. : 199801002475 (458601-U)]

Lecturer

NLST (Learn more about this study)

NLST-656

Apr 9, 2020

Explainable Artificial Intelligence Model for Lung Cancer Detection

In the medical domain, explainability was ranked the most desirable feature of a clinical decision support system by the physicians in order to gain the trust of users and patients. However, this is contradict with the state-of-the-art artificial intelligence (AI) based inference models that were implemented and used in this domain where the computation of the final decision is based on “black box” algorithms such as artificial neural network and deep learning. Although some remarkable results have been published and proven by scientific experiments, yet they are lacking the explicit declarative knowledge representation that constructs the underlying explanatory model which will answer the “Why” questions towards the predicted outcome. Such circumstances limited the trustworthiness of a diagnosis system as the algorithms have been shown to be easily misclassifying the inputs with no resemblance to the true categories and generally unreliable in supporting a very crucial decision. In this research, our objective is to extend the study of generating explainable AI in Lung cancer diagnosis by integrating the causality theory and model induction in medical imaging (e.g. CT-Scan) to support the decision making in the Lung cancer detection.