Causal importance of low-level feature selectivity for generalization in image recognition

Ukita, Jumpei

doi:10.1016/j.neunet.2020.02.009

Cited by 23 publications

(4 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…During the system design, multiple data security mechanisms are adopted to ensure reliable data transmission [27]. On the one hand, while the logistics information data is transmitted in real time, the data is stored locally for a short period of time.…”

Section: A Technical Test Environmentmentioning

confidence: 99%

Research on Information Management Based on Image Recognition and Virtual Reality

Zhou

2020

IEEE Access

View full text Add to dashboard Cite

Information management is an important part of national economic and social life. With the rapid rise of new technologies such as the Internet, more and more industries are seeking to integrate with informatization. At the same time, many problems have arisen due to the immaturity of information management technology. Therefore, improving the level of information management technology plays an important role in improving the efficiency of informatization. In view of the above problems, this paper builds a new information management model based on the original information management technology, combining image recognition and virtual reality technology. First, a text data collection model based on image recognition is designed in the information collection link. Secondly, based on the existing information management technology, virtual reality technology is used to visually display the image and text information. Finally, the case analysis and performance test results illustrate the superiority of the implemented information management technology. This method can provide scientific reference and basis for modern information management using image recognition technology.

show abstract

Section: A Technical Test Environmentmentioning

confidence: 99%

Research on Information Management Based on Image Recognition and Virtual Reality

Zhou

2020

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Deep learning has been widely used in recent medical research, such as automatic diagnosis from clinical images 1 – 4 , recognition of human genes 5 , and cognitive neuroscience 6 , 7 . This technique also aids in electrocardiogram (ECG) pattern recognition, such as predicting demographic features 8 and automatically identifying cardiovascular comorbidity 9 – 11 .…”

Section: Introductionmentioning

confidence: 99%

Predicting extremely low body weight from 12-lead electrocardiograms using a deep neural network

Kurisu,

Yamazaki,

Yoshiuchi

2024

Sci Rep

View full text Add to dashboard Cite

Previous studies have successfully predicted overweight status by applying deep learning to 12-lead electrocardiogram (ECG); however, models for predicting underweight status remain unexplored. Here, we assessed the feasibility of deep learning in predicting extremely low body weight using 12-lead ECGs, thereby investigating the prediction rationale for highlighting the parts of ECGs that are associated with extremely low body weight. Using records of inpatients predominantly with anorexia nervosa, we trained a convolutional neural network (CNN) that inputs a 12-lead ECG and outputs a binary prediction of whether body mass index is ≤ 12.6 kg/m2. This threshold was identified in a previous study as the optimal cutoff point for predicting the onset of refeeding syndrome. The CNN model achieved an area under the receiver operating characteristic curve of 0.807 (95% confidence interval, 0.745–0.869) on the test dataset. The gradient-weighted class activation map showed that the model focused on QRS waves. A negative correlation with the prediction scores was observed for QRS voltage. These results suggest that deep learning is feasible for predicting extremely low body weight using 12-lead ECGs, and several ECG features, such as lower QRS voltage, may be associated with extremely low body weight in patients with anorexia nervosa.

show abstract

“…e knowledge representation of neural network is unreadable to humans [8]. e exact reason why trained deep neural networks can implement recognition remains an open question [9]. e training algorithm does not specify the way to recognize.…”

Section: Introductionmentioning

confidence: 99%

An Implicit Memory-Based Method for Supervised Pattern Recognition

Wang

Yang

et al. 2021

Discrete Dynamics in Nature and Society

View full text Add to dashboard Cite

How the human brain does recognition is still an open question. No physical or biological experiment can fully reveal this process. Psychological evidence is more about describing phenomena and laws than explaining the physiological processes behind them. The need for interpretability is well recognized. This paper proposes a new method for supervised pattern recognition based on the working pattern of implicit memory. The artificial neural network (ANN) is trained to simulate implicit memory. When an input vector is not in the training set, the ANN can treat the input as a “do not care” term. The ANN may output any value when the input is a “do not care” term since the training process needs to use as few neurons as possible. The trained ANN can be expressed as a function to design a pattern recognition algorithm. Using the Mixed National Institute of Standards and Technology database, the experiments show the efficiency of the pattern recognition method.

show abstract

Causal importance of low-level feature selectivity for generalization in image recognition

Cited by 23 publications

References 29 publications

Research on Information Management Based on Image Recognition and Virtual Reality

Research on Information Management Based on Image Recognition and Virtual Reality

Predicting extremely low body weight from 12-lead electrocardiograms using a deep neural network

An Implicit Memory-Based Method for Supervised Pattern Recognition

Contact Info

Product

Resources

About