Fine-grained Patient Similarity Measuring using Deep Metric Learning

Ni, Jiazhi; Liu, Jie; Zhang, Chenxin; Ye, Dan; Ma, Zhen

doi:10.1145/3132847.3133022

Cited by 36 publications

(19 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Samples to be presented to the network and the relationship between them are related to the metric loss function. The metric loss functions such as contrastive loss [29], triplet loss [85], quadruple loss [100], n-pair loss [70], and so on allow for us to increase the data sample size (n), such as n 2 (paired samples), n 3 (triplet samples), and n 4 (quadruple samples). Inefficient paired samples or triple samples cause time consumption and too much memory space in the network training.…”

Section: Discussionmentioning

confidence: 99%

“…The time complexity of network training may exponentially increase, depending on this situation. To overcome these problems, hard negative mining [89,90] and semi-hard negative mining [32,100] offers informative samples for training. The correct sampling strategy plays very important role for fast convergence [32].…”

Section: Discussionmentioning

confidence: 99%

“…The angular loss pushes the negative point away from the center of the positive cluster and then brings positive points closer to each other while using an angle that is a rotation and scale-invariant metric (Equation (8)). While the Triplet loss only considers positive and negative samples to compute the distance between samples, it does not exploit any similarity degree information [100]. The authors in [100] obtained a better degree of closeness between objects while using quadruple samples in each training batch.…”

Section: Loss Functions For Deep Metric Learningmentioning

confidence: 99%

“…While the Triplet loss only considers positive and negative samples to compute the distance between samples, it does not exploit any similarity degree information [100]. The authors in [100] obtained a better degree of closeness between objects while using quadruple samples in each training batch. As shown in Equation (7), a new input sample similar to the sample X was added to the quadruple loss metric as an alternative approach to triplet loss.…”

Section: Loss Functions For Deep Metric Learningmentioning

confidence: 99%

See 3 more Smart Citations

Deep Metric Learning: A Survey

2019

View full text Add to dashboard Cite

Metric learning aims to measure the similarity among samples while using an optimal distance metric for learning tasks. Metric learning methods, which generally use a linear projection, are limited in solving real-world problems demonstrating non-linear characteristics. Kernel approaches are utilized in metric learning to address this problem. In recent years, deep metric learning, which provides a better solution for nonlinear data through activation functions, has attracted researchers' attention in many different areas. This article aims to reveal the importance of deep metric learning and the problems dealt with in this field in the light of recent studies. As far as the research conducted in this field are concerned, most existing studies that are inspired by Siamese and Triplet networks are commonly used to correlate among samples while using shared weights in deep metric learning. The success of these networks is based on their capacity to understand the similarity relationship among samples. Moreover, sampling strategy, appropriate distance metric, and the structure of the network are the challenging factors for researchers to improve the performance of the network model. This article is considered to be important, as it is the first comprehensive study in which these factors are systematically analyzed and evaluated as a whole and supported by comparing the quantitative results of the methods.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Section: Loss Functions For Deep Metric Learningmentioning

confidence: 99%

Section: Loss Functions For Deep Metric Learningmentioning

confidence: 99%

See 2 more Smart Citations

Deep Metric Learning: A Survey

2019

View full text Add to dashboard Cite

show abstract

“…These standardized variables are capable of being widely used in a variety of predictive models. Another way to utilize the new representation is to define distances between patients/encounters so that a comparable control group can be easily extracted from the data [31,32].…”

Section: Discussionmentioning

confidence: 99%

The application of unsupervised deep learning in predictive models using electronic health records

Wang

Tong

Davis

et al. 2020

BMC Med Res Methodol

View full text Add to dashboard Cite

Background:The main goal of this study is to explore the use of features representing patient-level electronic health record (EHR) data, generated by the unsupervised deep learning algorithm autoencoder, in predictive modeling. Since autoencoder features are unsupervised, this paper focuses on their general lower-dimensional representation of EHR information in a wide variety of predictive tasks. Methods: We compare the model with autoencoder features to traditional models: logistic model with least absolute shrinkage and selection operator (LASSO) and Random Forest algorithm. In addition, we include a predictive model using a small subset of response-specific variables (Simple Reg) and a model combining these variables with features from autoencoder (Enhanced Reg). We performed the study first on simulated data that mimics real world EHR data and then on actual EHR data from eight Advocate hospitals. Results: On simulated data with incorrect categories and missing data, the precision for autoencoder is 24.16% when fixing recall at 0.7, which is higher than Random Forest (23.61%) and lower than LASSO (25.32%). The precision is 20.92% in Simple Reg and improves to 24.89% in Enhanced Reg. When using real EHR data to predict the 30-day readmission rate, the precision of autoencoder is 19.04%, which again is higher than Random Forest (18.48%) and lower than LASSO (19.70%). The precisions for Simple Reg and Enhanced Reg are 18.70 and 19.69% respectively. That is, Enhanced Reg can have competitive prediction performance compared to LASSO. In addition, results show that Enhanced Reg usually relies on fewer features under the setting of simulations of this paper. Conclusions: We conclude that autoencoder can create useful features representing the entire space of EHR data and which are applicable to a wide array of predictive tasks. Together with important response-specific predictors, we can derive efficient and robust predictive models with less labor in data extraction and model training.

show abstract

Learning Fine-Grained Patient Similarity with Dynamic Bayesian Network Embedded RNNs

Wang

Chen

2019

Database Systems for Advanced Applications

View full text Add to dashboard Cite

Fine-grained Patient Similarity Measuring using Deep Metric Learning

Cited by 36 publications

References 16 publications

Deep Metric Learning: A Survey

Deep Metric Learning: A Survey

The application of unsupervised deep learning in predictive models using electronic health records

Learning Fine-Grained Patient Similarity with Dynamic Bayesian Network Embedded RNNs

Contact Info

Product

Resources

About