Local minima found in the subparameter space can be effective for ensembles of deep convolutional neural networks

Yang, Yongquan; Lv, Haijun; Chen, Ning; Wu, Yang; Zheng, Jiayi; Zheng, Zhongxi

doi:10.1016/j.patcog.2020.107582

Cited by 15 publications

(7 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Then, based on the learned discriminative features, we optimized the prediction (classification) part of CNN II by using cross-entropy loss [ 36 ] and SGD [ 37 ]. Additionally, we leveraged a recently proposed fast ensemble deep learning strategy [ 39 – 41 ] to further boost the optimized CNN II. In the post-processing step, a pCR-score was calculated by averaging the pCR probabilities of TE tiles for each WSI, which was regarded as a novel biomarker for pCR prediction from histology (Fig.…”

Section: Methodsmentioning

confidence: 99%

Deep learning-based predictive biomarker of pathological complete response to neoadjuvant chemotherapy from histological images in breast cancer

Yang

Wei

et al. 2021

J Transl Med

Self Cite

View full text Add to dashboard Cite

Background Pathological complete response (pCR) is considered a surrogate endpoint for favorable survival in breast cancer patients treated with neoadjuvant chemotherapy (NAC). Predictive biomarkers of treatment response are crucial for guiding treatment decisions. With the hypothesis that histological information on tumor biopsy images could predict NAC response in breast cancer, we proposed a novel deep learning (DL)-based biomarker that predicts pCR from images of hematoxylin and eosin (H&E)-stained tissue and evaluated its predictive performance. Methods In total, 540 breast cancer patients receiving standard NAC were enrolled. Based on H&E-stained images, DL methods were employed to automatically identify tumor epithelium and predict pCR by scoring the identified tumor epithelium to produce a histopathological biomarker, the pCR-score. The predictive performance of the pCR-score was assessed and compared with that of conventional biomarkers including stromal tumor-infiltrating lymphocytes (sTILs) and subtype. Results The pCR-score derived from H&E staining achieved an area under the curve (AUC) of 0.847 in predicting pCR directly, and achieved accuracy, F1 score, and AUC of 0.853, 0.503, and 0.822 processed by the logistic regression method, respectively, higher than either sTILs or subtype; a prediction model of pCR constructed by integrating sTILs, subtype and pCR-score yielded a mean AUC of 0.890, outperforming the baseline sTIL-subtype model by 0.051 (0.839, P = 0.001). Conclusion The DL-based pCR-score from histological images is predictive of pCR better than sTILs and subtype, and holds the great potentials for a more accurate stratification of patients for NAC.

show abstract

Section: Methodsmentioning

confidence: 99%

Deep learning-based predictive biomarker of pathological complete response to neoadjuvant chemotherapy from histological images in breast cancer

Yang

Wei

et al. 2021

J Transl Med

Self Cite

View full text Add to dashboard Cite

show abstract

“…Recent fast ensemble deep learning techniques [60,61] can be applied in whole slide medical image analysis (f.e. predicting pCR from H&E stained whole slide images [62]) to reduce time and space overheads, at the expense of certain accuracy.…”

Section: Ensemble Learningmentioning

confidence: 99%

Application of Transfer Learning and Ensemble Learning in Image-level Classification for Breast Histopathology

Zheng¹,

Li²,

Zhou³

et al. 2022

Preprint

View full text Add to dashboard Cite

Background: Breast cancer has the highest prevalence in women globally. The classification and diagnosis of breast cancer and its histopathological images have always been a hot spot of clinical concern. In Computer-Aided Diagnosis (CAD), traditional classification models mostly use a single network to extract features, which has significant limitations. On the other hand, many networks are trained and optimized on patient-level datasets, ignoring the application of lower-level data labels.Method: This paper proposes a deep ensemble model based on image-level labels for the binary classification of benign and malignant lesions of breast histopathological images. First, the BreakHis dataset is randomly divided into a training, validation and test set. Then, data augmentation techniques are used to balance the number of benign and malignant samples. Thirdly, considering the performance of transfer learning and the complementarity between each network, VGG-16, Xception, Resnet-50, DenseNet-201 are selected as the base classifiers.Result: In the ensemble network model with accuracy as the weight, the image-level binary classification achieves an accuracy of 98.90%. In order to verify the capabilities of our method, the latest Transformer and Multilayer Perception (MLP) models have been experimentally compared on the same dataset. Our model wins with a 5% − 20% advantage, emphasizing the ensemble model's far-reaching significance in classification tasks.Conclusion: This research focuses on improving the model's classification performance with an ensemble algorithm. Transfer learning plays an essential role in small datasets, improving training speed and accuracy. Our model has outperformed many existing approaches in accuracy, providing a method for the field of auxiliary medical diagnosis.

show abstract

“…The final model that emerges as a result of ensemble deep learning combines the advantages of its sub-models and shows much better generalization performance (Ganaie, Hu, Tanveer, & Suganthan, 2021). High-performance results are obtained with these integrated network applications in large-scale challenges such as natural language processing, computer vision and speech recognition (Yang et al, 2021).…”

Section: Ensemble Deep Learningmentioning

confidence: 99%

Covid-19 Detection from Chest X-Ray Images and Hybrid Model Recommendation with Convolutional Neural Networks

Karacan

ERYILMAZ²

2021

Journal of Advanced Research in Natural and Applied Sciences

View full text Add to dashboard Cite

The coronavirus pandemic, which emerged at the end of 2019, continues to be effective. Although various vaccines have been developed, the diagnosis of the disease and the 14-day isolation method still remain valid. Purpose: To classify diseases such as COVID-19, tuberculosis, other pneumonias, and lung opacity. For this goal, it was aimed to automate disease diagnosis, which was done manually by expert radiologists with chest X-Ray images, with a hybrid model created with the four most commonly used convolutional neural networks. In addition, it is aimed to reach a more accurate result by combining four CNN models instead of one model during the epidemic when diagnosis is of critical importance. Method: In the binary classification, the images were categorised as COVID-19positive and negative, and multi-class classification was conducted according to labels including Lung Opacity, COVID-19, Normal, and Viral Pneumonia. The recommended model consisted of MobileNetV2, DenseNet121, InceptionResNetV2, and Xception networks. Transfer learning was used for initial weights. These models, were combined with ensemble learning to obtain better classification performance. Results: The best binary classification result was obtained from the MobileNetv2 with an accuracy rate of 98.84%. The hybrid model improved slightly, with a value 0.16. In the multi-class classification, the DenseNet121 accomplished the highest classification rate, with 93.17%. The hybrid model improvement in the multi-class classification was 0.81. According to these results, the proposed method will alleviate the burden of health personnel during the epidemic. It will also aid in the diagnosis of COVID-19 in areas where facilities are limited.

show abstract

Local minima found in the subparameter space can be effective for ensembles of deep convolutional neural networks

Cited by 15 publications

References 30 publications

Deep learning-based predictive biomarker of pathological complete response to neoadjuvant chemotherapy from histological images in breast cancer

Deep learning-based predictive biomarker of pathological complete response to neoadjuvant chemotherapy from histological images in breast cancer

Application of Transfer Learning and Ensemble Learning in Image-level Classification for Breast Histopathology

Covid-19 Detection from Chest X-Ray Images and Hybrid Model Recommendation with Convolutional Neural Networks

Contact Info

Product

Resources

About