Deep learning architectures for multi-label classification of intelligent health risk prediction

Abstract

BackgroundMulti-label classification of data remains to be a challenging problem. Because of the complexity of the data, it is sometimes difficult to infer information about classes that are not mutually exclusive. For medical data, patients could have symptoms of multiple different diseases at the same time and it is important to develop tools that help to identify problems early. Intelligent health risk prediction models built with deep learning architectures offer a powerful tool for physicians to identify patterns in patient data that indicate risks associated with certain types of chronic diseases.ResultsPhysical examination records of 110,300 anonymous patients were used to predict diabetes, hypertension, fatty liver, a combination of these three chronic diseases, and the absence of disease (8 classes in total). The dataset was split into training (90%) and testing (10%) sub-datasets. Ten-fold cross validation was used to evaluate prediction accuracy with metrics such as precision, recall, and F-score. Deep Learning (DL) architectures were compared with standard and state-of-the-art multi-label classification methods. Preliminary results suggest that Deep Neural Networks (DNN), a DL architecture, when applied to multi-label classification of chronic diseases, produced accuracy that was comparable to that of common methods such as Support Vector Machines. We have implemented DNNs to handle both problem transformation and algorithm adaption type multi-label methods and compare both to see which is preferable.ConclusionsDeep Learning architectures have the potential of inferring more information about the patterns of physical examination data than common classification methods. The advanced techniques of Deep Learning can be used to identify the significance of different features from physical examination data as well as to learn the contributions of each feature that impact a patient’s risk for chronic diseases. However, accurate prediction of chronic disease risks remains a challenging problem that warrants further studies.

Biomarker Application for Precision Medicine in Stroke

“…With the amount of data gathered, machine-learning technology could foster the development of prognostic and diagnostic biomarkers and provide a solid mechanistic understanding of the dynamics relative to clinical course. Along that line, the U.S. Food and Drug Administration (FDA) announced an effort to develop a regulatory framework for medical artificial intelligence (AI) algorithms that reflects the fact that these tools are continuously learning and evolving from experience gained in real-world clinical use [68,69]. This biomarker-powered, self-learning engine may ultimately transform healthcare.…”

Multilabel segmentation of cancer cell culture on vascular structures with deep neural networks

“…15 It can be difficult to infer information about mutually not exclusive classes. 16,17 Segmenting this kind of multilabel images is less common. The class imbalance is also a problem.…”

Bangkok CCTV Image through a Road Environment Extraction System Using Multi-Label Convolutional Neural Network Classification

“…The framework outperformed other models in terms of accuracy, but the technique is still difficult to train end-to-end due to the objective function. Maxwell et al [31] presented health-risk prediction for multi-label problems with respect to chronic diseases. Their evaluation revealed more accuracy than traditional classification techniques such as decision trees, sequential minimal optimization, and multi-layer perception.…”

“…For example, if the image is calculated as a set of predicted labels {0.7, 0.3, 0.3, 0.6, 0.2, 0.7, 0.8, 0.3}, the image has an acceptable probability of showing rain, darkness, flow traffic, and wet road. To evaluate the model in terms of, for instance, exact match accuracy, we translated those events from the probabilistic value by defining the threshold value at 0.5, similar to [31]:…”