Development of a risk model towards personalised screening for radiation-induced breast cancer using the national Breast Screening After Radiotherapy

放疗使用辐射（x射线、质子）杀死癌细胞，对许多患者来说，这是癌症治疗的重要组成部分。然而，放射也可能导致在最初的癌症治疗后15年内发生新的癌症。对于儿童或年轻时被诊断患有癌症的女性来说，这可能会导致她们在治疗多年后患上乳腺癌的风险很高。在本研究中，我们将利用人工智能（机器学习）开发先进的图像处理和分析方法，以完善辐射诱发乳腺癌的风险预测。我们将使用在曼彻斯特开发和托管的国家放疗后乳房筛查数据集（BARD）的数据。BARD是首个此类国家登记处，是在曼彻斯特与英国公共卫生部门的同事广泛磋商和合作后开发的。它包括来自英国大约6500名女性的数据，这些女性由于之前的放疗而有患乳腺癌的风险。该项目将涉及处理2D和3D医学图像（CT、x光片、乳房x光片），并将这些成像特征与其他风险因素（乳房辐射剂量、乳房体积、年龄等）相结合，以开发个性化风险模型。该学生将在医学物理学、放射生物学、高级图像分析和流行病学的交叉领域工作。他们将受益于巴德研究小组内一个高度多学科的监督团队，包括医学物理学家、计算机科学家和肿瘤学家。他们将积极参与癌症幸存者/儿童和青年癌症领域的国家和国际合作。这项研究最终可能会为乳腺癌筛查提供个性化的指导方针，并改善癌症幸存者的生活质量和预后。

Radiotherapy uses radiation (x-rays, protons) to kill cancer cells, and is an important part of cancer treatment for many patients. However, radiation may also result in new cancers occurring over 15 years after the initial cancer treatment. For women diagnosed with cancer as children or young adults, this may lead to a high risk of developing breast cancer many years after treatment. In this study, we will use artificial intelligence (machine learning) to develop advanced image processing and analytical methods to refine risk prediction of radiation-induced breast cancer. We will use data from the national Breast Screening after Radiotherapy Dataset (BARD) developed and hosted in Manchester. BARD is a first-of-its-kind national registry, developed in Manchester following wide consultation and in collaboration with colleagues at Public Health England. It includes data from approximately 6,500 women across England at risk of breast cancer as a result of previous radiotherapy. The project will involve processing 2D and 3D medical images (CT, radiographs, mammograms) and combining these imaging features with other risk factors (radiation dose to the breast, breast volume, age, etc...) to develop a personalised risk model.The student will work at the intersection of medical physics, radiobiology, advance image analysis and epidemiology. They will benefit from a highly multidisciplinary supervisory team within the BARD research group, including medical physicists, computer scientists, and oncologists. They will be actively involved in collaborations at the national and international level in the field of cancer survivorship / childhood and young adult cancers. The study could ultimately lead to personalized guidelines for breast cancer screening and improved quality of life and outcomes for cancer survivors.