Developing interactive explanatory models for cancer prognosis

Lead Research Organisation: Imperial College London
Department Name: Dept of Computing


Prognosis is a key task in medicine, across multiple different domains. It guides discussions with patients, risk-adapted treatment decisions and planning for the future. Machine Learning techniques can provide prognostic predictions, and existing techniques allow some degree of explanation of these. But often machines and humans may have different experiences and access to different data. Therefore we need to enable machine-human interaction and sharing of explanations to develop shared explanations and models.

Today's AI landscape is permeated by plentiful data and dominated by powerful methods with the potential to impact a wide range of human sectors, including healthcare. Yet, this potential is hindered by the opacity of most data-centric AI methods and it is widely acknowledged that AI cannot fully benefit society without addressing its widespread inability to explain its outputs, causing human mistrust and doubts regarding its regulatory and ethical compliance. Extensive research efforts are currently being devoted towards explainable AI, but they are mostly fragmented, tailored to specific settings and focused on engineering static explanations that cannot respond to and benefit from human input. This project will aim developing interactive explanations in the clinical setting of neuro-oncology that can be deployed alongside data-centric AI methods built from data for patients with brain tumours to explain their outputs by providing justifications in their support. These can be progressively questioned by humans and refined as a result of their feedback, within conversational explanatory exchanges between humans and machines. The feedback in turn can be integrated to improve successive interactions. These interactive explanations for prognosis in brain tumour patients will be realised using computational argumentation as the underpinning technology.


10 25 50

Studentship Projects

Project Reference Relationship Related To Start End Student Name
EP/S023283/1 01/04/2019 30/09/2027
2454822 Studentship EP/S023283/1 05/10/2020 04/10/2024 Dekai (Kai) Zhang