|
| 骨质疏松人工智能技术研究进展及发展趋势 |
| Research progress and development trend of artificial intelligence technology for osteoporosis |
| |
| DOI：10.3969/j.issn.1006-7108.2019.02.028 |
| 中文关键词: 骨质疏松 人工智能 生物样本库 研究进展 |
| 英文关键词:osteoporosis, artificial intelligence, biological sample library, research progress |
| 基金项目: |
|
| 摘要点击次数: 1825 |
| 全文下载次数: 681 |
| 中文摘要: |
| 目的 探讨人工智能(artificial intelligence，AI)技术在骨质疏松早期诊断、早期预防、标准化治疗及科学化随访中的应用现状及发展前景。方法 查阅近十年人工智能在医学中的应用现状，分析探讨骨质疏松AI技术开发的可行性及其关键技术限制瓶颈。结果 开发优质AI技术的重要前提是大量准确知识的学习。骨质疏松筛查AI技术需要大量的骨密度数据和流行病学因素调查作为筛查系统的数据基础，诊治和随访AI技术需要大量的专业术语、影像学数据、血液、尿液生化指标的采集学习。因此，学习和验证过程中重要的是需要大量的骨密度数据、流行病学调查因素、血液尿液生化指标数据、骨密度测定影像资料、骨质疏松诊断中的专业术语资料、骨质疏松治疗过程中的用药及治疗效果资料等。这些相关资料的收集过程可以通过骨质疏松生物样本库的构建完成。结论 骨质疏松AI的开发离不开骨质疏松体检生物样本库的建设，高质量多中心大规模骨质疏松生物样本库构建过程中收集的大量可供机器人学习及再学习的资料是决定骨质疏松AI技术开发成败的关键。 |
| 英文摘要: |
| Objective To investigate the application status and development prospects of artificial intelligence (AI) in early diagnosis, early prevention, standardized treatment, and scientific follow-up of osteoporosis. Methods The application status of AI in medicine in the past 10 years was reviewed. The feasibility of osteoporosis AI technology development and its key technical bottlenecks were analyzed. Results An important prerequisite for the development of high-quality AI technology is the learning of a large amount of accurate knowledge. Osteoporosis screening AI technology requires many bone mineral density data and epidemiological factor survey as the data basis of the screening system. Diagnosis and follow-up AI technology requires many professional terms, imaging data, blood and urine biochemical indicator collection. Therefore, it is important in the learning and verification process to require a large amount of bone mineral density data, epidemiological investigation factors, blood and urine biochemical indicator data, bone mineral density imaging data, terminology data in osteoporosis diagnosis and treatment. The collection process of these related data can be completed through the construction of a database of osteoporosis biological samples. Conclusion The development of osteoporosis AI is inseparable from the construction of a biological sample bank for osteoporosis physical examination. The data collected during the construction of a high-quality, multi-center, and large-scale osteoporosis biological sample library for robot learning and re-learning are the key to the success of osteoporosis AI technology development. |
| 查看全文 查看/发表评论 下载PDF阅读器 |
| 关闭 |
|
|
|