Dynamic Feature Acquisition Using Denoising Autoencoders

Kachuee, Mohammad; Darabi, Sajad; Moatamed, Babak; Sarrafzadeh, Majid

doi:10.1109/tnnls.2018.2880403

Cited by 19 publications

(30 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…ii) Opportunistic Learning (OL) [21] a method based on deep Q-learning with variations of model uncertainty as the reward function, iii) a method based on exhaustive measurements of the sensitivity [16], and iv) a method based on approximation of sensitivities using denoising autoencoders (FACT) [17].…”

Section: Methodsmentioning

confidence: 99%

“…This normalization permits using the value of zero for missing features during the prediction to act as mean imputation. Regarding the number of layers and hidden neurons, we used a similar number of trainable parameters for OL [21] and RL-Based [19], while due to the inherent differences, we had to use different architectures for FACT [17] and Exhustive [16]. Nonetheless, for each classification task, the compared models reach a similar baseline accuracy (i.e., average accuracy after acquiring all the features).…”

Section: Methodsmentioning

confidence: 99%

“…In an earlier work [17], the authors suggested an approximation of (3) using a binary representation layer in denoising autoencoder architectures:…”

Section: E Sensitivity-based Approachmentioning

confidence: 99%

“…Recently, sensitivity analysis of predictor models was suggested as a measure of feature importance [16], [17]. In this context, sensitivity is a feature value measure representing the influence of each feature on the predicted outcome by a predictor model.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Cost-Sensitive Diagnosis and Learning Leveraging Public Health Data

Kachuee,

Karkkainen,

Goldstein

et al. 2019

Preprint

Self Cite

View full text Add to dashboard Cite

Traditionally, machine learning algorithms rely on the assumption that all features of a given dataset are available for free. However, there are many concerns such as monetary data collection costs, patient discomfort in medical procedures, and privacy impacts of data collection that require careful consideration in any real-world health analytics system. An efficient solution would only acquire a subset of features based on the value it provides while considering acquisition costs. Moreover, datasets that provide feature costs are very limited, especially in healthcare. In this paper, we provide a health dataset as well as a method for assigning feature costs based on the total level of inconvenience asking for each feature entails. Furthermore, based on the suggested dataset, we provide a comparison of recent and state-of-the-art approaches to cost-sensitive feature acquisition and learning. Specifically, we analyze the performance of major sensitivity-based and reinforcement learning based methods in the literature on three different problems in the health domain, including diabetes, heart disease, and hypertension classification.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

“…In an earlier work [17], the authors suggested an approximation of (3) using a binary representation layer in denoising autoencoder architectures:…”

Section: E Sensitivity-based Approachmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Cost-Sensitive Diagnosis and Learning Leveraging Public Health Data

Kachuee,

Karkkainen,

Goldstein

et al. 2019

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Density Estimation via Autoencoders The work in Kachuee et al (2018) suggests to acquire the covariate which has the highest sensitivity to the output prediction y. In order to account for different covariate acquisition costs the sensitivity scores are re-scaled appropriately.…”

Section: Markov Decision Process (Mdp) Frameworkmentioning

confidence: 99%

Adaptive covariate acquisition for minimizing total cost of classification

Andrade

Okajima

2021

Mach Learn

View full text Add to dashboard Cite

In some applications, acquiring covariates comes at a cost which is not negligible. For example in the medical domain, in order to classify whether a patient has diabetes or not, measuring glucose tolerance can be expensive. Assuming that the cost of each covariate, and the cost of misclassification can be specified by the user, our goal is to minimize the (expected) total cost of classification, i.e. the cost of misclassification plus the cost of the acquired covariates. We formalize this optimization goal using the (conditional) Bayes risk and describe the optimal solution using a recursive procedure. Since the procedure is computationally infeasible, we consequently introduce two assumptions: (1) the optimal classifier can be represented by a generalized additive model, (2) the optimal sets of covariates are limited to a sequence of sets of increasing size. We show that under these two assumptions, a computationally efficient solution exists. Furthermore, on several medical datasets, we show that the proposed method achieves in most situations the lowest total costs when compared to various previous methods. Finally, we weaken the requirement on the user to specify all misclassification costs by allowing the user to specify the minimally acceptable recall (target recall). Our experiments confirm that the proposed method achieves the target recall while minimizing the false discovery rate and the covariate acquisition costs better than previous methods.

show abstract

Peri-Diagnostic Decision Support Through Cost-Efficient Feature Acquisition at Test-Time

Vivar

Mullakaeva

Zwergal

et al. 2020

Medical Image Computing and Computer Assisted Intervention – MICCAI 2020

View full text Add to dashboard Cite

Computer-aided diagnosis (CADx) algorithms in medicine provide patient-specific decision support for physicians. These algorithms are usually applied after full acquisition of high-dimensional multimodal examination data, and often assume feature-completeness. This, however, is rarely the case due to examination costs, invasiveness, or a lack of indication. A sub-problem in CADx, which to our knowledge has received very little attention among the CADx community so far, is to guide the physician during the entire peri-diagnostic workflow, including the acquisition stage. We model the following question, asked from a physician's perspective: "Given the evidence collected so far, which examination should I perform next, in order to achieve the most accurate and efficient diagnostic prediction?". In this work, we propose a novel approach which is enticingly simple: use dropout at the input layer, and integrated gradients of the trained network at test-time to attribute feature importance dynamically. We validate and explain the effectiveness of our proposed approach using two public medical and two synthetic datasets. Results show that our proposed approach is more cost-and feature-efficient than prior approaches and achieves a higher overall accuracy. This directly translates to less unnecessary examinations for patients, and a quicker, less costly and more accurate decision support for the physician.

show abstract

Dynamic Feature Acquisition Using Denoising Autoencoders

Cited by 19 publications

References 25 publications

Cost-Sensitive Diagnosis and Learning Leveraging Public Health Data

Cost-Sensitive Diagnosis and Learning Leveraging Public Health Data

Adaptive covariate acquisition for minimizing total cost of classification

Peri-Diagnostic Decision Support Through Cost-Efficient Feature Acquisition at Test-Time

Contact Info

Product

Resources

About