An improved joint non-negative matrix factorization for identifying surgical treatment timing of neonatal necrotizing enterocolitis

Abstract

Neonatal necrotizing enterocolitis is a severe neonatal intestinal disease. Timely identification of surgical indications is essential for newborns in order to seek the best time for treatment and improve prognosis. This paper attempts to establish an algorithm model based on multimodal clinical data to determine the features of surgical indications and construct an auxiliary diagnosis model. The proposed algorithm adds hypergraph constraints on the two modal data based on Joint Nonnegative Matrix Factorization (JNMF), aiming to mine the higher-order correlations of the two data features. In addition, the adjacency matrix of the two kinds of data is used as a network regularization constraint to prevent overfitting. Orthogonal and L1-norm regulations were introduced to avoid feature redundancy and perform feature selection, respectively, and confirmed 14 clinical features. Finally, we used three classifiers, random forest, support vector machine, and logistic regression, to perform binary classification of patients requiring surgery. The results show that when the features selected by the proposed algorithm model are classified by random forest, the area under the ROC curve is 0.8, which has high prediction accuracy.

A novel two-stage deep learning model used to assist in diagnosing neonatal necrotizing enterocolitis and determining the need for surgical treatment

“…It is estimated that one in a thousand live born infants develop NEC and one in four diagnosed with NEC deteriorate to require emergency life-saving surgery [8] . 46.5% of the patients with NEC who need surgery cannot survive. 25% of those survivors develop life altering comorbidities such as short bowel syndrome or impaired neuro-development [9,10] .…”

“…However, this study lacks a description of the data and does not quantify the results of the model, as well as testing and verifying the validity of the model on a large data set. Guoqiang proposed an improved joint non-negative matrix factorization for identifying surgical treatment timing of neonatal necrotizing enterocolitis [46] . This study used data from only 45 children for model training and lacked testing and validation of the model on a large data set.…”