LipoDDx: a mobile application for identification of rare lipodystrophy syndromes
LipoDDx: a mobile application for identification of rare lipodystrophy syndromes
复制标题
DOI：
10.1186/s13023-020-01364-1
复制
发表时间：
2020-04-02
影响因子：
3.7
通讯作者：
Sanchez-Iglesias, Sofia
中科院分区：
文献类型：
--
作者：
Araujo-Vilar, David;Fernandez-Pombo, Antia;Sanchez-Iglesias, Sofia
背景脂肪营养不良综合征是一组以脂肪组织丢失为特征的疾病，一旦排除了其他营养缺乏或恶化的情况。除了HIV相关的脂肪营养不良外，它们的患病率非常低，再加上它们巨大的表型异质性，使得它们的识别变得困难，甚至对于内分泌学家和儿科医生来说也是如此。这会导致诊断的严重延误甚至误诊。我们的团队已经开发出一种算法，可以识别迄今为止描述的40多种罕见的脂肪代谢障碍亚型。该算法已在免费的移动的应用程序LipoDDx（R）中实现。我们的目的是确定LipoDDx（R）的有效性。分析了40例确诊为大多数脂肪营养不良亚型的患者的临床记录，包括无脂肪营养不良的受试者。由13名医生、1名生物化学家和1名牙医对诊断设盲的病历进行评价。每个评估员首先根据自己的标准给出自己的结果。然后，使用LipoDDx（R）进行第二次诊断。根据每个病例的复杂程度和疾病的流行程度，根据评分表对结果进行分析。结果LipoDDx（R）提供了一个用户友好的环境，基于通常的二分问题或从下拉菜单中选择临床体征。该应用程序为特定病例提供的最终结果可以是患有特定脂肪代谢障碍亚型的低/高概率。不使用LipoDDx（R）的成功率为17 +/-20%，而使用LipoDDx（R）的成功率为79 +/- 20%（p < 0.01）。结论LipoDDx（R）是一款免费的应用程序，可以识别罕见脂肪代谢障碍的亚型，在这个小型队列中，其有效性约为80%，这将有助于非该领域专家的医生。然而，为了获得更准确的效率值，将需要分析更多的情况。
Background Lipodystrophy syndromes are a group of disorders characterized by a loss of adipose tissue once other situations of nutritional deprivation or exacerbated catabolism have been ruled out. With the exception of the HIV-associated lipodystrophy, they have a very low prevalence, which together with their large phenotypic heterogeneity makes their identification difficult, even for endocrinologists and pediatricians. This leads to significant delays in diagnosis or even to misdiagnosis. Our group has developed an algorithm that identifies the more than 40 rare lipodystrophy subtypes described to date. This algorithm has been implemented in a free mobile application, LipoDDx (R). Our aim was to establish the effectiveness of LipoDDx (R). Forty clinical records of patients with a diagnosis of certainty of most lipodystrophy subtypes were analyzed, including subjects without lipodystrophy. The medical records, blinded for diagnosis, were evaluated by 13 physicians, 1 biochemist and 1 dentist. Each evaluator first gave his/her results based on his/her own criteria. Then, a second diagnosis was given using LipoDDx (R). The results were analysed based on a score table according to the complexity of each case and the prevalence of the disease. Results LipoDDx (R) provides a user-friendly environment, based on usually dichotomous questions or choice of clinical signs from drop-down menus. The final result provided by this app for a particular case can be a low/high probability of suffering a particular lipodystrophy subtype. Without using LipoDDx (R) the success rate was 17 +/- 20%, while with LipoDDx (R) the success rate was 79 +/- 20% (p < 0.01). Conclusions LipoDDx (R) is a free app that enables the identification of subtypes of rare lipodystrophies, which in this small cohort has around 80% effectiveness, which will be of help to doctors who are not experts in this field. However, it will be necessary to analyze more cases in order to obtain a more accurate efficiency value.