Differential Diagnosis Model of Hypocellular Myelodysplastic Syndrome and Aplastic Anemia Based on the Medical Big Data Platform

Abstract

The arrival of the era of big data has brought new ideas to solve problems for all walks of life. Medical clinical data is collected and stored in the medical field by utilizing the medical big data platform. Based on medical information big data, new ideas and methods for the differential diagnosis of hypo-MDS and AA are studied. The basic information, peripheral blood classification counts, peripheral blood cell morphology, bone marrow cell morphology, and other information were collected from patients diagnosed with hypo-MDS and AA diagnosed in the first diagnosis. First, statistical analysis was performed. Then, the logistic regression model, decision tree model, BP neural network model, and support vector machine (SVM) model of hypo-MDS and AA were established. The sensitivity, specificity, Youden index, positive likelihood ratio (+LR), negative likelihood ratio (−LR), area under curve (AUC), accuracy, Kappa value, positive predictive value (+PV), negative predictive value (−PV) of the four model training set and test set were compared, respectively. Finally, with the support of medical big data, using logistic regression, decision tree, BP neural network, and SVM four classification algorithms, the decision tree algorithm is optimal for the classification of hypo-MDS and AA and analyzes the characteristics of the optimal model misjudgment data.

Research and Application of Acoustic Emission Signal Processing Technology

“…The parametric analysis method is to obtain the waveform characteristics of material damage by using acoustic emission sensors, and to analyze the waveform by using acoustic emission parameters, so that the damage degree of material can be estimated by using acoustic emission parameters [31]. Common parameters include the number of acoustic emission events, amplitude, ringing count, impact technology, rise time, duration and so on [32]. The parametric analysis method is a classical and constantly developing sound acoustic emission signal processing method, which has relatively perfect theory and is widely used.…”

Research on Intelligent Guidance Optimal Path of Shared Car Charging in the IOT Environment

“…The Basic Idea of PSO Algorithm. The particle swarm algorithm was proposed by Eberhart, an expert in computational intelligence research in the United States, and Dr. Kennedy, a psychologist research specialist in 1995 [23][24][25]. The algorithm is a random search algorithm based on the regularity of bird activity and the swarm intelligence.…”

Application of Clustering Analysis in Brain Gene Data Based on Deep Learning

“…Deep belief network is a neural network with multiple hidden layers published by Hinton et al [28]. It is stacked by a probability model called restricted Boltzmann machine (RBM).…”