一种用于预测肥胖症静息能量消耗的人工神经网络。
An artificial neural network to predict resting energy expenditure in obesity.
机构信息
Centre Intégré de l'Obésité Rhône-Alpes, Fédération Hospitalo-Universitaire DO-iT, Department of Endocrinology and Nutrition, Groupement Hospitalier Sud, Hospices Civils de Lyon, Lyon, France; Centre de Recherche en Nutrition Humaine Rhône-Alpes (CRNH-RA), Centre Européen Nutrition et Santé (CENS), Lyon, France; Laboratoire CarMeN, Unité INSERM U1060 - INRA 1235 - INSA-Lyon, Université Claude Bernard Lyon 1, Lyon, France.
Centre Intégré Nord Francilien de l'Obésité (CINFO), Service des Explorations Fonctionnelles, Centre de référence de prise en charge de l'obésité, Hôpital Louis Mourier (AP-HP), Université Paris Diderot, Sorbonne Paris Cité, France.
出版信息
Clin Nutr. 2018 Oct;37(5):1661-1669. doi: 10.1016/j.clnu.2017.07.017. Epub 2017 Sep 1.
BACKGROUND & AIMS: The resting energy expenditure (REE) determination is important in nutrition for adequate dietary prescription. The gold standard i.e. indirect calorimetry is not available in clinical settings. Thus, several predictive equations have been developed, but they lack of accuracy in subjects with extreme weight including obese populations. Artificial neural networks (ANN) are useful predictive tools in the area of artificial intelligence, used in numerous clinical fields. The aim of this study was to determine the relevance of ANN in predicting REE in obesity.
METHODS
A Multi-Layer Perceptron (MLP) feed-forward neural network with a back propagation algorithm was created and cross-validated in a cohort of 565 obese subjects (BMI within 30-50 kg m) with weight, height, sex and age as clinical inputs and REE measured by indirect calorimetry as output. The predictive performances of ANN were compared to those of 23 predictive REE equations in the training set and in two independent sets of 100 and 237 obese subjects for external validation.
RESULTS
Among the 23 established prediction equations for REE evaluated, the Harris & Benedict equations recalculated by Roza were the most accurate for the obese population, followed by the USA DRI, Müller and the original Harris & Benedict equations. The final 5-fold cross-validated three-layer 4-3-1 feed-forward back propagation ANN model developed in that study improved precision and accuracy of REE prediction over linear equations (precision = 68.1%, MAPE = 8.6% and RMSPE = 210 kcal/d), independently from BMI subgroups within 30-50 kg m. External validation confirmed the better predictive performances of ANN model (precision = 73% and 65%, MAPE = 7.7% and 8.6%, RMSPE = 187 kcal/d and 200 kcal/d in the 2 independent datasets) for the prediction of REE in obese subjects.
CONCLUSIONS
We developed and validated an ANN model for the prediction of REE in obese subjects that is more precise and accurate than established REE predictive equations independent from BMI subgroups. For convenient use in clinical settings, we provide a simple ANN-REE calculator available at: https://www.crnh-rhone-alpes.fr/fr/ANN-REE-Calculator.
背景与目的
静息能量消耗（REE）的测定对于适当的饮食处方在营养方面很重要。间接测热法是金标准，但在临床环境中无法获得。因此，已经开发了多种预测方程，但它们在包括肥胖人群在内的体重极端的受试者中准确性不足。人工神经网络（ANN）是人工智能领域中有用的预测工具，已应用于许多临床领域。本研究的目的是确定 ANN 在肥胖人群中预测 REE 的相关性。
方法
创建了一个具有反向传播算法的多层感知器（MLP）前馈神经网络，并在 565 名肥胖受试者（BMI 在 30-50 kg/m 之间）队列中进行了交叉验证，体重、身高、性别和年龄作为临床输入，间接测热法测量的 REE 作为输出。ANN 的预测性能与训练集中的 23 个 REE 预测方程以及两个独立的 100 名和 237 名肥胖受试者的验证集进行了比较。
结果
在所评估的 23 种用于 REE 的既定预测方程中，重新计算的 Harris & Benedict 方程（Roza 计算）对于肥胖人群最准确，其次是美国膳食参考摄入量（DRI）、Müller 和原始的 Harris & Benedict 方程。在该研究中开发的最终 5 折交叉验证的三层 4-3-1 前馈反向传播 ANN 模型提高了线性方程对 REE 预测的精度和准确性（精度=68.1%，平均绝对百分比误差（MAPE）=8.6%，均方根误差（RMSPE）=210 千卡/d），与 BMI 亚组无关（BMI 在 30-50 kg/m 之间）。外部验证证实了 ANN 模型（在两个独立数据集的预测中，精度分别为 73%和 65%，平均绝对百分比误差（MAPE）分别为 7.7%和 8.6%，均方根误差（RMSPE）分别为 187 千卡/d 和 200 千卡/d）对肥胖受试者 REE 预测的更好的预测性能。
结论
我们开发并验证了一种用于预测肥胖受试者 REE 的 ANN 模型，该模型比基于 BMI 亚组的既定 REE 预测方程更精确、更准确。为了便于在临床环境中使用，我们提供了一个简单的 ANN-REE 计算器，网址为：https://www.crnh-rhone-alpes.fr/fr/ANN-REE-Calculator。
Zotero 插件