暴殄天物啊,让玩数据的看到多少年来的真实医疗数据就这么变成了废纸,痛心
A paper published in Nature in 2020
A local hospital participated in a big data learning project targeting a local urinary tract infection. The initial traditional treatment regimen was a combination of two antibiotics (A and B), which was later improved to a combination of three (A, B, and C) based on research findings and clinical trials. During this project, when clinicians prescribed treatment protocols, they attempted to reduce the use of one antibiotic at a time, tracking and recording patient follow-ups and treatment outcomes. Based on the results of over 3,000 cases, it was determined that antibiotic C no longer had any therapeutic effect on the bacteria causing this urinary tract infection, as the local bacteria had mutated and developed resistance. The study also estimated the timeline of when this resistance emerged.
The findings helped the local hospital refine its treatment protocols, reduce financial expenditure and economic burden for patients, and lower antibiotic usage, which contributes to overall public health. Furthermore, the results assist in developing evolutionary tracing research on antibiotic resistance in pathogens.
For a long time, hospital case resources have remained underutilized. In the era before big data, these cases eventually became obsolete paper records stored in archives. Yet, these detailed and valuable resources often contain keys to predicting the future and revealing the truths of life. Today, newly hired doctors generally possess the skills to conduct meta-analyses of cases, electronic medical record entry is now widespread, and physicians worldwide have an obligation and mission to share local cases into global databases. With the rapid growth and development of the computer industry, its influence and impact will ultimately benefit every industry below it, shining and emitting heat in these interdisciplinary fields. Future scientific development is almost certainly going to be networked and intersecting, requiring specialized skills and knowledge from all walks of life and all fields.
20年Nature上的一篇文章
当地一家医院参与了一项大数据学习的项目,针对当地发生的一种尿道感染疾病,最开始的传统治疗方案是两种抗生素混用(A和B),后来依据科研成果和临床试验改良成三种抗生素混用(A,B和C)。这个项目是在医师开出治疗方案时,分别尝试减少一种抗生素的使用,记录追踪患者的复诊情况和治疗情况,最后根据3000+个病例结果,得出了抗生素C已经对这种诱发尿道感染的细菌没有治疗效果了,当地这种细菌已经发生了突变产生了抗性,并且将抗性产生时间也进行了估算。
成果将帮助当地医院改良治疗方案,减少患者财务支出和经济负担,减少抗生素的使用也有助于人群整体健康,其成果也有助于展开病原菌对抗生素耐药性产生的演化示踪研究。
长期以来医院的病例资源都是得不到利用的,在缺乏对大数据利用的时代,这些病例最终变成积压在资料室的废纸,这些记录详细的宝贵资源中往往存在着预言未来和揭示生命真理的钥匙。如今新入职的医生基本都掌握着meta分析病例的本领,电子录入病例也早已普及,全球的医师都有着将本地的病例共享进全球数据库中的义务和使命。如今计算机行业蓬勃高速发展,其影响力和作用力终究会惠及到下面的每一个行业,并且在这些交叉学科里发光发热。未来的科学发展几乎一定是网状的交叉的,需要各行各业各个领域的专业技能和知识。