Intelligent and Personalised Risk Stratification and Early Diagnosis of Lung Cancer

Lead Research Organisation: King's College London
Department Name: Imaging & Biomedical Engineering

Abstract

Lung cancer is the second most common cancer in both males and females, and has a very poor prognosis, causing >35,000 of cancer-related deaths each year (nearly 100 every day). This is due to the mostly very late-stage diagnosis of cancer: nearly 50% of all lung cancer cases are only diagnosed at very late Stage IV where no curative treatment exists. The annual cost of lung cancer to the UK economy is estimated to be around £2.4 billion, taking into account the cost of treatment and premature death, the cost to business of sick leave and of unpaid care by friends and family. It eclipses the cost of any other cancer, and continues to present a significant economic and healthcare burden. There is currently no national lung cancer screening programme in the UK, as current tests are deemed to be inadequate, and not outweighing risks associated with screening that involves radiation exposure.

The vision for our Healthcare Impact Partnership is to pave the way for a fundamentally different approach in which lung cancer can be detected at an early stage in patients identified as being at high risk of developing lung cancer: In a close three-way partnership between clinicians specialising in lung disease, healthcare industry, and computational imaging/machine learning researchers, we will jointly develop, test and clinically evaluate new computational methods for detecting, classifying and monitoring lung nodules in low-dose longitudinal CT of patients identified of being at risk of developing lung cancer. Specifically, we will embed powerful machine learning methods within a lung registration framework that has been previous developed using EPSRC funding. Through the use of deep feature learning techniques as well as through learning complex respiratory motion patterns, we will explore transfer learning from large, annotated lung CT image databases to a high-risk patient cohort receiving low-dose CT imaging.

Through this partnership, which brings together expertise in computational imaging and deep learning, expert clinicians and a leading CT manufacturer, we aim to achieve both computational innovations in personalised diagnostics, as well as a strong translational focus for improved patient management, effectively leading the way in providing advanced healthcare technologies for a future UK lung cancer screening programme.

Planned Impact

Lung cancer is the second most common cancer in men and women in the UK, with >45,000 new diagnoses each year and a high morbidity (5-year survival is 10%, with 100 deaths per day). 50% of all cases are diagnosis only at late stage IV, when curative treatment is no longer possible. The annual cost of lung cancer to the UK economy is estimated to be around £2.4 billion, taking into account the cost of treatment and premature death, the cost to business of sick leave and of unpaid care by friends and family. It eclipses the cost of any other cancer, with each lung cancer patient costing the UK healthcare system alone over £9k p.a. Lung cancer therefore continues to present a significant economic and healthcare burden. There is currently no national lung cancer screening programme in the UK, as current tests are deemed to be inadequate, and not outweighing risks associated with screening that involves radiation exposure (Source: CRUK).

This Healthcare Impact Partnership will form a close partnership with clinicians and a leading computed tomography (CT) manufacturer, as well as scientists in computational imaging and machine learning, to develop and clinically evaluate new computational approaches for early detection of lung cancer, to improve patient management, reduce morbidity and reduce healthcare costs. Our goal of designing an intelligent computational tool for personalised risk stratification for patients identified as being at risk of developing lung cancer will pave the way for a future UK national lung cancer screening programme.

The main beneficiaries are patients with chronic obstructive pulmonary disease (COPD), who may be identified as high-risk patients due to smoking habits or other factors. These patients will benefit from early diagnosis of cancer, where it may still be treatable, and would be at least trebling their survival rate, and improve their quality of life and patient experience. Beneficiaries also include their families or carers, as well as their employers and through that the UK economy. Healthcare savings would potentially be substantial, e.g. by reducing major surgeries to minor interventions, which would be of benefit to healthcare providers.
Clinicians (radiologists, oncologists, physicians) would also benefit from the proposed research, being assisted by an intelligent tool for early clinical diagnosis, reducing the conventional radiological reading and providing automated, objective and predictive measures. The risk stratification would further aid in clinical decision making for optimal treatment selection, patient recall or referrals. Early diagnosis of lung cancer would widen the currently very small window of opportunity of developing effective cancer treatments, which would also be of benefit to pharmaceutical industries.
Healthcare industry, represented by a market leader in medical imaging in this Healthcare Impact Partnership, are also a beneficiary of the proposed research, as they would significantly benefit from advanced diagnostics tools that provide meaningful clinical information to the end user, aiding their product development and marketing.

Finally, other researchers would benefit from translating ideas generated by this research to other clinical applications in cancer, cardiology or neurology, where patient risk stratification is required. Through that, the societal and healthcare benefit would further increase as a wider range of patients would received optimised treatment based on more accurate and early diagnosis.

Publications

10 25 50
 
Description While this project is still at an early stage, it effectively links national and international lung image databases for early detection and characterisation of lung nodules for patients at risk of developing lung cancer. During the first year of this project, several large scale clinical trials (the North-American National Lung Screening Trial - NLST, and the Dutch-Belgian Randomized Lung Cancer Screening Trial - NELSON) supported the hypotheses of this project that annual or even more frequent lung screening can help detect cancer at an earlier, still treatable stage, which is now also trialed by the NHS.
Exploitation Route Findings in this work might be helping to pave the way for a UK lung screening programme.
Sectors Communities and Social Services/Policy,Digital/Communication/Information Technologies (including Software),Healthcare,Pharmaceuticals and Medical Biotechnology

 
Description European Commission Expert Group on Liability and New Technologies
Geographic Reach Europe 
Policy Influence Type Participation in a national consultation
Impact The expert consultation has directly influenced the report of the working group on liability of new technologies in healthcare.
URL https://ec.europa.eu/transparency/regexpert/index.cfm?do=groupDetail.groupDetail&groupID=3592&news=1
 
Description EPSRC Centre for Doctoral Training in Smart Medical Imaging at King's College London and Imperial College London
Amount £6,022,394 (GBP)
Funding ID EP/S022104/1 
Organisation Engineering and Physical Sciences Research Council (EPSRC) 
Sector Public
Country United Kingdom
Start 09/2019 
End 03/2028
 
Description Improving the early detection of lung cancer by combining exosomal analysis of hypoxia with standard of care imaging
Amount £531,701 (GBP)
Organisation Cancer Research UK 
Sector Charity/Non Profit
Country United Kingdom
Start 04/2019 
End 03/2023
 
Description London Medical Imaging & Artificial Intelligence Centre for Value-Based Healthcare
Amount £9,985,272 (GBP)
Funding ID 104691 
Organisation Innovate UK 
Sector Public
Country United Kingdom
Start 02/2019 
End 01/2022
 
Description Collaboration with Prof Kensaku Mori, Nagoya University 
Organisation Nagoya University
Country Japan 
Sector Academic/University 
PI Contribution We have hosted Prof Kensaku Mori and his PhD student, Hirohisa Oda, several times in our labs.
Collaborator Contribution Prof Mori is paying frequent lab visits, resulting in joint publications
Impact Publications as associated with grant
Start Year 2017
 
Description Collaboration with Royal Brompton Hospital 
Organisation Royal Brompton & Harefield NHS Foundation Trust
Country United Kingdom 
Sector Public 
PI Contribution We have engaged with regular joined project meetings, exchanging new research ideas.
Collaborator Contribution Dr Anant Devaraj from RBH has joined this project team, contributing with additional annotated lung CT data
Impact n/a
Start Year 2018
 
Description Collaboration with University College Hospital (2018 - Still Active) 
Organisation University College Hospital
Country United Kingdom 
Sector Hospitals 
PI Contribution See next point.
Collaborator Contribution Dr Arjun Nair, originally based at Guy's and St. Thomas' Trusts (GSTT), has taken on a new appointment at UCLH. As the lead of the lung cancer pathway of the GSTT TOHETI study (Transforming Healthcare Through imaging), Dr Nair as one of our key clinical collaborators continues to provide direct clinical input and expertise into this project.
Impact Ongoing
Start Year 2018
 
Description BIR/RCR Meeting: AI in Radiology in 2020 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Professional Practitioners
Results and Impact Invited talk at the BIR/RCR Annual Meeting on AI in radiology
Year(s) Of Engagement Activity 2020
URL https://www.mybir.org.uk/CPBase__event_detail?id=a173Y00000CwkXhQAJ&site=a0N2000000COvFsEAL
 
Description CRUK/EPSRC Early Detection Sandpit - 7 January 2019, Oxford 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Professional Practitioners
Results and Impact Invited panel talk & discussion on "AI for Early Detection" - Prof Julia Schnabel
Year(s) Of Engagement Activity 2019
 
Description Developments in Healthcare Imaging - Connecting with Academia 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Professional Practitioners
Results and Impact As part of the MedIAN network, we organised a session on Deep Learning in Medical Imaging with selected invited talks from early career researchers and an introduction the network.
Year(s) Of Engagement Activity 2018
URL https://gateway.newton.ac.uk/event/tgmw58/programme
 
Description Developments in Healthcare Imaging - Connecting with Industry 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Industry/Business
Results and Impact Presented the MedIAN network and invited participants to join the community.
Year(s) Of Engagement Activity 2017
URL http://www.turing-gateway.cam.ac.uk/event/tgmw48/programme
 
Description ESR Connect 
Form Of Engagement Activity A broadcast e.g. TV/radio/film/podcast (other than news/press)
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact Live-streamed show on AI in radiology
Year(s) Of Engagement Activity 2020
URL https://connect.myesr.org/course/use-cases-in-ai-reasons-to-do-ai-with-friends/
 
Description East Anglian Radiological Society Annual Meeting (EARS) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Professional Practitioners
Results and Impact Talk "Hopes and Hurdles for AI in Radiology"
Year(s) Of Engagement Activity 2019
URL https://radiology.medschl.cam.ac.uk/event/east-anglian-radiological-society-annual-meeting-ears-2019...
 
Description European Commission Expert Group on Liability and New Technologies 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Policymakers/politicians
Results and Impact Provide the European Commission with expertise on the applicability of the Product Liability Directive to traditional products, new technologies and new societal challenges (Product Liability Directive formation) and assist the Commission in developing principles that can serve as guidelines for possible adaptations of applicable laws at EU and national level relating to new technologies (New Technologies formation).
Year(s) Of Engagement Activity 2018
URL http://ec.europa.eu/transparency/regexpert/index.cfm?do=groupDetail.groupMeeting&meetingId=9390
 
Description Horizon Europe: New Parliament, new Commission, new agenda 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Policymakers/politicians
Results and Impact Panel discussion: Value-based healthcare: How technologies can improve care across the EU
Year(s) Of Engagement Activity 2019
URL https://sciencebusiness.net/events/horizon-europe-new-parliament-new-commission-new-agenda
 
Description Imperial Global Science Policy Forum 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Policymakers/politicians
Results and Impact The Imperial Global Science Policy Forum is a high-profile network connecting Imperial academics with senior international science and technology advisers and diplomats, UK government policymakers, industry experts and other relevant stakeholders.
Year(s) Of Engagement Activity 2018
URL https://www.imperial.ac.uk/news/188848/diplomats-industry-leaders-discuss-future-ai/
 
Description Machine Learning for Translational Medicine & Personalized Healthcare 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Professional Practitioners
Results and Impact Talk "Spot-the-Lesion: Image- based disease detection with deep learning"
Year(s) Of Engagement Activity 2019
URL https://www.eventbrite.com/e/imperial-college-itmat-dsg-workshop-2019-data-science-challenges-and-op...
 
Description Medical Imaging Summer School (MISS 2018): Medical Imaging meets Deep Learning - Director Prof Julia Schnabel 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Postgraduate students
Results and Impact The focus of this Medical Imaging Summer School (MISS) is to train a new generation of young scientists to bridge this gap, by providing insights into the various interfaces between medical imaging and deep learning, based on the shared broad categories of medical image computing, computer-aided image interpretation and disease classification. The course will contain a combination of in-depth tutorial-style lectures on fundamental state-of-the-art concepts, followed by accessible yet advanced research lectures using examples and applications. A broad overview of the field will be given, and guided reading groups will complement lectures. The course will be delivered by world renowned experts from both academia and industry, who are working closely at the interface of medical imaging/deep learning.

The school aims to provide a stimulating graduate training opportunity for young researchers and Ph.D. students. The participants will benefit from direct interaction with world leaders in medical image computing and deep learning(often working in both fields). Participants will also have the opportunity to present their own research, and to interact with their scientific peers, in a friendly and constructive setting.
Year(s) Of Engagement Activity 2018
URL http://iplab.dmi.unict.it/miss18/