Interpretable Deep Learning in Electronics, Computer Science and Medical Imaging

A special issue of Electronics (ISSN 2079-9292). This special issue belongs to the section "Artificial Intelligence".

Deadline for manuscript submissions: closed (31 August 2020) | Viewed by 21892

Special Issue Editor


E-Mail Website1 Website2 Website3
Guest Editor

Special Issue Information

Dear Colleagues,

Artificial intelligence (AI), particularly, deep learning (DL), which involves automated feature extraction using deep neural networks (DNNs), has been used increasingly by electronics engineer, computer scientist, and physicians. AI can analyze computer vision and medical images at a level not possible by a single person. However, the resulting parameters are difficult to interpret. This so-called “black box” problem causes opaqueness in DL.

The aim of the Special Issue is to help realize interpretable DL in electronics, computer science, and medical imaging. To achieve this aim, we should attempt to bring about a paradigm shift in electronics, computer science, and medical imaging in which diagnostic accuracy is surpassed to achieve explainability. DL in medical imaging has still considerable limitations. To interpret and apply DL to medical imaging tasks effectively, sufficient expertise in computer science is required. We should interpret elements of decision-making behind classification decisions. While DL algorithms can markedly enhance the quantitative performance, such as accuracy, interpretability is a vital component.

Moreover, establishing accountability is one of the most important issues in medical imaging to explain the classification results clearly. Although more interpretable algorithms seem likely to be more readily accepted by electronics, computers science, and medical professionals, it remains necessary to determine whether this could increase effectiveness in electronics, computer science, and medical imaging. For the acceptance of AI by electronics engineers, computer scientists, and physicians, not only quantitative, but also qualitative algorithmic performance should be improved.

Topics of interest of this Special Issue include, but are not limited to:

  • Interpretable DL in electronics
  • Interpretable DL in computer science
  • Interpretable DL in medical imaging
  • Non-black-box machine learning
  • Interpretable large decision trees and random forests
  • Interpretable machine learning
  • Converting deep neural network to decision trees
  • Interpretable decision trees

Prof. Dr. Yoichi Hayashi
Guest Editor

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Electronics is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2400 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Benefits of Publishing in a Special Issue

  • Ease of navigation: Grouping papers by topic helps scholars navigate broad scope journals more efficiently.
  • Greater discoverability: Special Issues support the reach and impact of scientific research. Articles in Special Issues are more discoverable and cited more frequently.
  • Expansion of research network: Special Issues facilitate connections among authors, fostering scientific collaborations.
  • External promotion: Articles in Special Issues are often promoted through the journal's social media, increasing their visibility.
  • e-Book format: Special Issues with more than 10 articles can be published as dedicated e-books, ensuring wide and rapid dissemination.

Further information on MDPI's Special Issue polices can be found here.

Published Papers (6 papers)

Order results
Result details
Select all
Export citation of selected articles as:

Research

10 pages, 607 KiB  
Communication
Does Deep Learning Work Well for Categorical Datasets with Mainly Nominal Attributes?
by Yoichi Hayashi
Electronics 2020, 9(11), 1966; https://doi.org/10.3390/electronics9111966 - 21 Nov 2020
Cited by 4 | Viewed by 3012
Abstract
Given the complexity of real-world datasets, it is difficult to present data structures using existing deep learning (DL) models. Most research to date has concentrated on datasets with only one type of attribute: categorical or numerical. Categorical data are common in datasets such [...] Read more.
Given the complexity of real-world datasets, it is difficult to present data structures using existing deep learning (DL) models. Most research to date has concentrated on datasets with only one type of attribute: categorical or numerical. Categorical data are common in datasets such as the German (-categorical) credit scoring dataset, which contains numerical, ordinal, and nominal attributes. The heterogeneous structure of this dataset makes very high accuracy difficult to achieve. DL-based methods have achieved high accuracy (99.68%) for the Wisconsin Breast Cancer Dataset, whereas DL-inspired methods have achieved high accuracy (97.39%) for the Australian credit dataset. However, to our knowledge, no such method has been proposed to classify the German credit dataset. This study aimed to provide new insights into the reasons why DL-based and DL-inspired classifiers do not work well for categorical datasets, mainly consisting of nominal attributes. We also discuss the problems associated with using nominal attributes to design high-performance classifiers. Considering the expanded utility of DL, this study's findings should aid in the development of a new type of DL that can handle categorical datasets consisting of mainly nominal attributes, which are commonly used in risk evaluation, finance, banking, and marketing. Full article
Show Figures

Figure 1

15 pages, 3200 KiB  
Article
Guided Networks for Few-Shot Image Segmentation and Fully Connected CRFs
by Kun Zhang, Yuanjie Zheng, Xiaobo Deng, Weikuan Jia, Jian Lian and Xin Chen
Electronics 2020, 9(9), 1508; https://doi.org/10.3390/electronics9091508 - 14 Sep 2020
Viewed by 2693
Abstract
The goal of the few-shot learning method is to learn quickly from a low-data regime. Structured output tasks like segmentation are challenging for few-shot learning, due to their being high-dimensional and statistically dependent. For this problem, we propose improved guided networks and combine [...] Read more.
The goal of the few-shot learning method is to learn quickly from a low-data regime. Structured output tasks like segmentation are challenging for few-shot learning, due to their being high-dimensional and statistically dependent. For this problem, we propose improved guided networks and combine them with a fully connected conditional random field (CRF). The guided network extracts task representations from annotated support images through feature fusion to do fast, accurate inference on new unannotated query images. By bringing together few-shot learning methods and fully connected CRFs, our method can do accurate object segmentation by overcoming poor localization properties of deep convolutional neural networks and can quickly updating tasks, without further optimization, when faced with new data. Our guided network is at the forefront of accuracy for the terms of annotation volume and time. Full article
Show Figures

Figure 1

15 pages, 3600 KiB  
Article
One-Dimensional Convolutional Neural Networks with Feature Selection for Highly Concise Rule Extraction from Credit Scoring Datasets with Heterogeneous Attributes
by Yoichi Hayashi and Naoki Takano
Electronics 2020, 9(8), 1318; https://doi.org/10.3390/electronics9081318 - 16 Aug 2020
Cited by 11 | Viewed by 4714
Abstract
Convolution neural networks (CNNs) have proven effectiveness, but they are not applicable to all datasets, such as those with heterogeneous attributes, which are often used in the finance and banking industries. Such datasets are difficult to classify, and to date, existing high-accuracy classifiers [...] Read more.
Convolution neural networks (CNNs) have proven effectiveness, but they are not applicable to all datasets, such as those with heterogeneous attributes, which are often used in the finance and banking industries. Such datasets are difficult to classify, and to date, existing high-accuracy classifiers and rule-extraction methods have not been able to achieve sufficiently high classification accuracies or concise classification rules. This study aims to provide a new approach for achieving transparency and conciseness in credit scoring datasets with heterogeneous attributes by using a one-dimensional (1D) fully-connected layer first CNN combined with the Recursive-Rule Extraction (Re-RX) algorithm with a J48graft decision tree (hereafter 1D FCLF-CNN). Based on a comparison between the proposed 1D FCLF-CNN and existing rule extraction methods, our architecture enabled the extraction of the most concise rules (6.2) and achieved the best accuracy (73.10%), i.e., the highest interpretability–priority rule extraction. These results suggest that the 1D FCLF-CNN with Re-RX with J48graft is very effective for extracting highly concise rules for heterogeneous credit scoring datasets. Although it does not completely overcome the accuracy–interpretability dilemma for deep learning, it does appear to resolve this issue for credit scoring datasets with heterogeneous attributes, and thus, could lead to a new era in the financial industry. Full article
Show Figures

Figure 1

21 pages, 928 KiB  
Article
A Two-Step Rule-Extraction Technique for a CNN
by Guido Bologna and Silvio Fossati
Electronics 2020, 9(6), 990; https://doi.org/10.3390/electronics9060990 - 13 Jun 2020
Cited by 13 | Viewed by 3476
Abstract
The explanation of the decisions provided by a model are crucial in a domain such as medical diagnosis. With the advent of deep learning, it is very important to explain why a classification is reached by a model. This work tackles the transparency [...] Read more.
The explanation of the decisions provided by a model are crucial in a domain such as medical diagnosis. With the advent of deep learning, it is very important to explain why a classification is reached by a model. This work tackles the transparency problem of convolutional neural networks(CNNs). We propose to generate propositional rules from CNNs, because they are intuitive to the way humans reason. Our method considers that a CNN is the union of two subnetworks: a multi-layer erceptron (MLP) in the fully connected layers; and a subnetwork including several 2D convolutional layers and max-pooling layers. Rule extraction exhibits two main steps, with each step generating rules from each subnetwork of the CNN. In practice, we approximate the two subnetworks by two particular MLP models that makes it possible to generate propositional rules. We performed the experiments with two datasets involving images: MNISTdigit recognition; and skin-cancer diagnosis. With high fidelity, the extracted rules designated the location of discriminant pixels, as well as the conditions that had to be met to achieve the classification. We illustrated several examples of rules by their centroids and their discriminant pixels. Full article
Show Figures

Figure 1

16 pages, 2480 KiB  
Article
Channel and Spatial Attention Regression Network for Cup-to-Disc Ratio Estimation
by Shuo Li, Chiru Ge, Xiaodan Sui, Yuanjie Zheng and Weikuan Jia
Electronics 2020, 9(6), 909; https://doi.org/10.3390/electronics9060909 - 29 May 2020
Cited by 4 | Viewed by 2860
Abstract
Cup-to-disc ratio (CDR) is of great importance during assessing structural changes at the optic nerve head (ONH) and diagnosis of glaucoma. While most efforts have been put on acquiring the CDR number through CNN-based segmentation algorithms followed by the calculation of CDR, these [...] Read more.
Cup-to-disc ratio (CDR) is of great importance during assessing structural changes at the optic nerve head (ONH) and diagnosis of glaucoma. While most efforts have been put on acquiring the CDR number through CNN-based segmentation algorithms followed by the calculation of CDR, these methods usually only focus on the features in the convolution kernel, which is, after all, the operation of the local region, ignoring the contribution of rich global features (such as distant pixels) to the current features. In this paper, a new end-to-end channel and spatial attention regression deep learning network is proposed to deduces CDR number from the regression perspective and combine the self-attention mechanism with the regression network. Our network consists of four modules: the feature extraction module to extract deep features expressing the complicated pattern of optic disc (OD) and optic cup (OC), the attention module including the channel attention block (CAB) and the spatial attention block (SAB) to improve feature representation by aggregating long-range contextual information, the regression module to deduce CDR number directly, and the segmentation-auxiliary module to focus the model’s attention on the relevant features instead of the background region. Especially, the CAB selects relatively important feature maps in channel dimension, shifting the emphasis on the OD and OC region; meanwhile, the SAB learns the discriminative ability of feature representation at pixel level by capturing the relationship of intra-feature map. The experimental results of ORIGA dataset show that our method obtains absolute CDR error of 0.067 and the Pearson’s correlation coefficient of 0.694 in estimating CDR and our method has a great potential in predicting the CDR number. Full article
Show Figures

Figure 1

15 pages, 1559 KiB  
Article
On the Interpretability of Machine Learning Models and Experimental Feature Selection in Case of Multicollinear Data
by Franc Drobnič, Andrej Kos and Matevž Pustišek
Electronics 2020, 9(5), 761; https://doi.org/10.3390/electronics9050761 - 6 May 2020
Cited by 22 | Viewed by 3894
Abstract
In the field of machine learning, a considerable amount of research is involved in the interpretability of models and their decisions. The interpretability contradicts the model quality. Random Forests are among the best quality technologies of machine learning, but their operation is of [...] Read more.
In the field of machine learning, a considerable amount of research is involved in the interpretability of models and their decisions. The interpretability contradicts the model quality. Random Forests are among the best quality technologies of machine learning, but their operation is of “black box” character. Among the quantifiable approaches to the model interpretation, there are measures of association of predictors and response. In case of the Random Forests, this approach usually consists of calculating the model’s feature importances. Known methods, including the built-in one, are less suitable in settings with strong multicollinearity of features. Therefore, we propose an experimental approach to the feature selection task, a greedy forward feature selection method with least-trees-used criterion. It yields a set of most informative features that can be used in a machine learning (ML) training process with similar prediction quality as the original feature set. We verify the results of the proposed method on two known datasets, one with small feature multicollinearity and another with large feature multicollinearity. The proposed method also allows for a domain expert help with selecting among equally important features, which is known as the human-in-the-loop approach. Full article
Show Figures

Figure 1

Back to TopTop