Heuristics, Algorithms and Machine Learning: Evaluation & Testing in Radiation Therapy
ConditionCancer
Trial statusRecruiting
Trial phaseNot listed
Trial typeObservational
Biological sexAll
Age18+
SponsorCCTU- Cancer Theme
About this trial
The Hamlet.rt study is a prospective data collection and patient questionnaire study for patients undergoing image-guided radiotherapy with curative intent.
The aim of the study is to use novel machine learning and mathematical techniques to build a model that can predict the risk of significant side effects from radiotherapy treatment for an individual patient: using calculations of normal tissue dose from radiotherapy treatment planning and patient baseline characteristics derived from image and non-image data, continuously updated as the patient is reviewed both during and after treatment.
A secondary goal of the project is to facilitate research in machine learning and medical image processing for radiation therapy through the creation of a discoverable and shared data resource for research use.
Eligibility criteria
This trial does not accept healthy volunteersQualifiers
Participant is willing and able to give informed consent for participation in the study
Male or Female
Aged 18 years or older
Diagnosed with primary prostate cancer, head and neck cancer, lung cancer, or brain tumour
Disqualifiers
Participant is not willing or able to complete the protocol-stated requirements of the study, e.g. accessing & completing web-based long-term follow-up questionnaires.
Trial population
Adults suitable for radical image-guided radiotherapy with Prostate, Head \& Neck, Brain, or Lung Cancer. The variation in conditions is based on the requirements of Machine Learning algorithms requiring high levels of clinical applicability, which depends on the quality and quantity of the input data available. The input data set therefore should adequately encompass the variation in anatomy encountered in the population.
Trial design
Design model
Cohort
Time perspective
Prospective
Treatments tested in this trial
Radical Image-Guided Radiotherapy
RadiationQuestionnaires administered will monitor the clinical toxicity experienced by each patient up to 5 years post radiotherapy
Treatment groups
310 Participants
are divided into 4 treatment groupsGroup A: Prostate Cancer1 intervention
Group B: Head & Neck Cancer1 intervention
Group C: Central Nervous System Tumours1 intervention
Group D: Lung Cancer1 intervention
Trial outcomes
Primary outcomes
1
Machine Learning Modelling
Characterise machine learning models for the four disease sites. Developing machine learning algorithms for autosegmentation of normal tissue anatomy, and to extend machine learning algorithms to identify and segment normal tissue structures in cone beam CT images, and to utilise the ML segmentations to evaluate image signatures correlated with treatment toxicity
Time frame
8 years from FPFV2
Predictive Modelling
Predict performance matches with published techniques. Combining the machine learning models in outcome 1, with pre-treatment assessment data and on-treatment quantitative assessments in outcome 3 for the construction and evaluation of a predictive mathematical model
Time frame
8 years from FPFV3
Clinical Toxicity Evaluation
Evaluation of the clinical toxicity experienced by each patient up to 5 years post radiotherapy to inform the predictive models in outcome 2
Time frame
8 years from FPFVSponsors and contacts
Click on the lead sponsor to view all of their trials.
CCTU- Cancer Theme
Lead sponsor
Cambridge University Hospitals NHS Foundation Trust
Sponsor institution
University of Cambridge
Collaborator
Microsoft Research
Collaborator
Source ClinicalTrials.gov