AI-based Screening for Breast Cancer Treatment

人工智能在乳腺癌治疗中的应用 乳腺癌是影响女性的最常见癌症。在加拿大，2014年估计有24,400例新诊断的乳腺癌，报告的诊断和治疗费用范围为3.09亿至4.54亿加元。最近，个性化医疗进一步增加了这一成本。拟议研究的目标是找到基于人工智能的解决方案来取代昂贵的分子测试，并减少这些分子测试的复杂性所固有的显著周转时间(TAT)。在这个项目中，艾伯塔省卫生服务公司(AHS)和艾德蒙顿大学的DyaLIFE医学实验室正在与艾伯塔大学计算科学的Nilanjan Ray博士和他的团队合作。 加拿大的整体医疗质量排名仍然在世界上名列前茅，超过了美国的医疗体系，尽管它占国内生产总值(GDP)的10.3%，而占美国GDP的17.8%。维持这一级别在预算上是具有挑战性的。医学正在永久性地发展，一种新的范式包括通过人工智能(AI)和机器学习(ML)利用大量数据的能力，以更好地筛查、诊断、预测、治疗和跟踪疾病。这些工具可以显著降低成本。因此，加拿大不仅要做好采用这些工具的准备，而且还要开发它们，这可能至关重要。加拿大在ML领域设有英才中心。这个项目是实现上述两个目标的一个例子：1)优化(降低成本，改善TAT)重要的癌症生物标志物，以及2)加强当地ML的专业知识(研究、招生、潜在创业和改善当地经济)。

AI-based Screening for Breast Cancer Treatment Breast cancer is the commonest cancer affecting women. For Canada in 2014, an estimated number of 24,400 new diagnoses of breast cancer with a reported range cost of 309-454 million CAD for diagnosis and treatment. Recently, personalized medicine has increased this cost further. The goal of the proposed research is to find AI-based solutions to replace expensive molecular tests and to reduce the significant turn around time (TAT) inherent to the complexity of these molecular assays. For this project, Alberta Health Services (AHS) and DynaLIFE Medical Labs Edmonton are partnering with Dr. Nilanjan Ray and his team at Computing Science, University of Alberta. Canadian overall healthcare quality ranking is still among the best in the world and beats the U.S. healthcare system, even though it represents 10.3% of gross domestic product (GDP) compared to 17.8% of the U.S. GDP. It is budgetarily challenging to maintain this rank. Medicine is permanently evolving, and a new paradigm includes the capacity to exploit a large amount of data through Artificial Intelligence (AI) and Machine Learning (ML) to better screen, diagnose, predict, treat and follow-up diseases. These tools can reduce costs significantly. It might be therefore critical for Canada to not only be ready to adopt these tools but to develop them. Canada has centres of excellence in the field of ML. This project is an example where these two aforementioned goals are met: 1) optimization (reduction of cost, improved TAT) of important cancer biomarkers, and 2) strengthening local ML expertise (Research, student recruitment, potential startup, and improving local economy).