IBM Research, NY, USA.
IBM Research, MA, USA.
AMIA Annu Symp Proc. 2021 Jan 25;2020:727-736. eCollection 2020.
Type 1 diabetes (T1D) is a chronic autoimmune disease that affects about 1 in 300 children and up to 1 in 100 adults during their life-time. Improvements in early prediction of T1D onset may help prevent diagnosis for diabetic ketoacidosis, a serious complication often associated with a missed or delayed T1D diagnosis. In addition to genetic factors, progression to T1D is strongly associated with immunologic factors that can be measured during clinical visits. We developed a T1D-specific ontology that captures the dynamic patterns of these biomarkers and used it together with a survival model, RankSvx, proposed in our prior work. We applied this approach to a T1D dataset harmonized from three birth cohort studies from the United States, Finland, and Sweden. Results show that the dynamic biomarker patterns captured in the proposed ontology are able to improve prediction performance (in concordance index) by 5.3%, 3.3%, 2.8%, and 1.0% over baseline for 3, 6, 9, and 12 month duration windows, respectively.
1 型糖尿病（T1D）是一种慢性自身免疫性疾病，在儿童中的发病率约为每 300 人中 1 例，在成人中的发病率约为每 100 人中 1 例。提高 T1D 发病的早期预测能力可能有助于预防糖尿病酮症酸中毒的诊断，这是一种严重的并发症，常与 T1D 的漏诊或延迟诊断有关。除遗传因素外，T1D 的进展与免疫因素密切相关，这些免疫因素可以在临床就诊时进行测量。我们开发了一种针对 T1D 的本体，该本体捕获了这些生物标志物的动态模式，并将其与我们之前工作中提出的生存模型 RankSvx 一起使用。我们将该方法应用于从美国、芬兰和瑞典的三个出生队列研究中协调的 T1D 数据集。结果表明，所提出的本体中捕获的动态生物标志物模式能够分别将 3、6、9 和 12 个月持续时间窗口的预测性能（一致性指数）提高 5.3%、3.3%、2.8%和 1.0%。
