CNN Explainer: Learning Convolutional Neural Networks with Interactive Visualization

Wang, Zijie; Turko, Robert; Shaikh, Omar; Park, Haekyu; Das, Nilaksh; Hohman, Fred; Kahng, Minsuk; Chau, Duen Horng

doi:10.1109/tvcg.2020.3030418

Cited by 193 publications

(71 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, in the broader ML field, there has been significant recent progress in explainable ML techniques, and it has been pointed out that these approaches may be preferred by the medical community and regulators [ 48 , 49 ]. Several explanation methods take specific, previously black-box methods, such as convolutional neural networks [ 50 ], and allow for post-hoc explanation of their decision-making process. Other explainability algorithms are model-agnostic, meaning they can be applied to any type of model, regardless of its mathematical basis [ 51 ].…”

Section: Discussionmentioning

confidence: 99%

Machine Learning–Based Early Warning Systems for Clinical Deterioration: Systematic Scoping Review

Muralitharan

Nelson

et al. 2021

J Med Internet Res

View full text Add to dashboard Cite

Background Timely identification of patients at a high risk of clinical deterioration is key to prioritizing care, allocating resources effectively, and preventing adverse outcomes. Vital signs–based, aggregate-weighted early warning systems are commonly used to predict the risk of outcomes related to cardiorespiratory instability and sepsis, which are strong predictors of poor outcomes and mortality. Machine learning models, which can incorporate trends and capture relationships among parameters that aggregate-weighted models cannot, have recently been showing promising results. Objective This study aimed to identify, summarize, and evaluate the available research, current state of utility, and challenges with machine learning–based early warning systems using vital signs to predict the risk of physiological deterioration in acutely ill patients, across acute and ambulatory care settings. Methods PubMed, CINAHL, Cochrane Library, Web of Science, Embase, and Google Scholar were searched for peer-reviewed, original studies with keywords related to “vital signs,” “clinical deterioration,” and “machine learning.” Included studies used patient vital signs along with demographics and described a machine learning model for predicting an outcome in acute and ambulatory care settings. Data were extracted following PRISMA, TRIPOD, and Cochrane Collaboration guidelines. Results We identified 24 peer-reviewed studies from 417 articles for inclusion; 23 studies were retrospective, while 1 was prospective in nature. Care settings included general wards, intensive care units, emergency departments, step-down units, medical assessment units, postanesthetic wards, and home care. Machine learning models including logistic regression, tree-based methods, kernel-based methods, and neural networks were most commonly used to predict the risk of deterioration. The area under the curve for models ranged from 0.57 to 0.97. Conclusions In studies that compared performance, reported results suggest that machine learning–based early warning systems can achieve greater accuracy than aggregate-weighted early warning systems but several areas for further research were identified. While these models have the potential to provide clinical decision support, there is a need for standardized outcome measures to allow for rigorous evaluation of performance across models. Further research needs to address the interpretability of model outputs by clinicians, clinical efficacy of these systems through prospective study design, and their potential impact in different clinical settings.

show abstract

Section: Discussionmentioning

confidence: 99%

Machine Learning–Based Early Warning Systems for Clinical Deterioration: Systematic Scoping Review

Muralitharan

Nelson

et al. 2021

J Med Internet Res

View full text Add to dashboard Cite

show abstract

“…In the next stage, the AC module consists of a single flatten layer, followed by a fully connected layer and an output layer. The flatten layer is responsible for transforming the features into a vector that can be forwarded into a fully connected [34]. ReLU and softmax activation functions are selected to be used in the fully-connected layer and output layer, respectively.…”

Section: Proposed Methodsology a Proposed Cnn Model Structurementioning

confidence: 99%

A Proof-of-Concept of Ultra-Edge Smart IoT Sensor: A Continuous and Lightweight Arrhythmia Monitoring Approach

et al. 2021

View full text Add to dashboard Cite

Due to the proliferation of the Internet of Things (IoT), the IoT devices are becoming utilized at the edge network at a much higher rate. Conventionally, the IoT devices lack the computation resources required for carrying out ultra-edge analytics. In this paper, we go beyond the typical edge analytics paradigm, which is mostly limited to user-smartphones, and investigate how to embed intelligence into the ultra-edge IoT sensors. To conceptualize the smart IoT sensors with enhanced intelligence, we select the arrhythmia detection task employing Electrocardiogram (ECG) trace as one of the mobile health (mHealth) cases. The existing approaches are not feasible for ultra-edge IoT sensors due to the extensive noise-filtering and manual feature extraction phase. Hence, in this paper, to facilitate the analytics, we propose a Deep Learning-based Lightweight Arrhythmia Classification (DL-LAC) method, which employs only singlelead ECG trace and does not require noise-filtering and manual feature extraction steps. As the proposed technique, we design a one-dimensional Convolutional Neural Network (CNN) architecture. Complying with the ANSI/AAMI EC57:1998 standard, four heartbeat types are taken into consideration as class labels. The efficiency and the generalization ability of the proposed model are evaluated, employing four different datasets from PhysioNet. The experimental results demonstrate that the proposed DL method outperforms traditional methods such as the Delay Differential Equation (DDE)-based optimization, K-Nearest Neighbor (KNN), and Random Forest (RF). The proposed DL-LAC illustrates encouraging performance in terms of time and memory requirement when the trained model is transferred to virtualized microcontrollers connected to IoT sensors. INDEX TERMSInternet of Things (IoT), Arrhythmia, Electrocardiogram (ECG), Deep Learning (DL), Convolutional Neural Network (CNN), smart health, smart sensor.

show abstract

“…An in-depth explanation of CNNs is beyond the scope of this study. The interested reader may refer to [65]. The section provides an architectural overview of the selected pretrained models.…”

Section: Pre-trained Models and Standard Neural Architecturesmentioning

confidence: 99%

Damage detection using in-domain and cross-domain transfer learning

Bukhsh

Jansen

Saeed

2021

Neural Comput & Applic

View full text Add to dashboard Cite

We investigate the capabilities of transfer learning in the area of structural health monitoring. In particular, we are interested in damage detection for concrete structures. Typical image datasets for such problems are relatively small, calling for the transfer of learned representation from a related large-scale dataset. Past efforts of damage detection using images have mainly considered cross-domain transfer learning approaches using pre-trained ImageNet models that are subsequently fine-tuned for the target task. However, there are rising concerns about the generalizability of ImageNet representations for specific target domains, such as for visual inspection and medical imaging. We, therefore, evaluate a combination of in-domain and cross-domain transfer learning strategies for damage detection in bridges. We perform comprehensive comparisons to study the impact of cross-domain and in-domain transfer, with various initialization strategies, using six publicly available visual inspection datasets. The pre-trained models are also evaluated for their ability to cope with the extremely low-data regime. We show that the combination of cross-domain and in-domain transfer persistently shows superior performance specially with tiny datasets. Likewise, we also provide visual explanations of predictive models to enable algorithmic transparency and provide insights to experts about the intrinsic decision logic of typically black-box deep models.

show abstract

CNN Explainer: Learning Convolutional Neural Networks with Interactive Visualization

Cited by 193 publications

References 38 publications

Machine Learning–Based Early Warning Systems for Clinical Deterioration: Systematic Scoping Review

Machine Learning–Based Early Warning Systems for Clinical Deterioration: Systematic Scoping Review

A Proof-of-Concept of Ultra-Edge Smart IoT Sensor: A Continuous and Lightweight Arrhythmia Monitoring Approach

Damage detection using in-domain and cross-domain transfer learning

Contact Info

Product

Resources

About