Lightweight convolutional neural network with knowledge distillation for cervical cells classification

The task of segmenting cytoplasm in cytology images is one of the most challenging tasks in cervix cytological analysis due to the presence of fuzzy and highly overlapping cells. Deep learning-based diagnostic technology has proven to be effective in segmenting complex medical images. We present a two-stage framework based on Mask RCNN to automatically segment overlapping cells. In stage one, candidate cytoplasm bounding boxes are proposed. In stage two, pixel-to-pixel alignment is used to refine the boundary and category classification is also presented. The performance of the proposed method is evaluated on publicly available datasets from ISBI 2014 and 2015. The experimental results demonstrate that our method outperforms other state-of-the-art approaches with DSC 0.92 and FPRp 0.0008 at the DSC threshold of 0.8. Those results indicate that our Mask RCNN-based segmentation method could be effective in cytological analysis.

Download Full-text

A scalable convolutional neural network for task-specified scenarios via knowledge distillation

2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) ◽

10.1109/icassp.2017.7952600 ◽

2017 ◽

Cited By ~ 3

Author(s):

Mengnan Shi ◽

Fei Qin ◽

Qixiang Ye ◽

Zhenjun Han ◽

Jianbin Jiao

Keyword(s):

Neural Network ◽

Convolutional Neural Network ◽

Knowledge Distillation

Download Full-text

An Automatic Mass Screening System for Cervical Cancer Detection Based on Convolutional Neural Network

Mathematical Problems in Engineering ◽

10.1155/2020/4864835 ◽

2020 ◽

Vol 2020 ◽

pp. 1-14

Author(s):

Aziz-ur -Rehman ◽

Nabeel Ali ◽

Imtiaz.A. Taj ◽

Muhammad Sajid ◽

Khasan S. Karimov

Keyword(s):

Neural Network ◽

Cervical Cancer ◽

Convolutional Neural Network ◽

Fine Tuning ◽

Screening Tests ◽

Support Vector ◽

Screening System ◽

Cervical Cells ◽

Fully Connected ◽

Testing Protocols

Cervical cancer is the fourth most common type of cancer and is also a leading cause of mortality among women across the world. Various types of screening tests are used for its diagnosis, but the most popular one is the Papanicolaou smear test, in which cell cytology is carried out. It is a reliable tool for early identification of cervical cancer, but there is always a chance of misdiagnosis because of possible errors in human observations. In this paper, an auto-assisted cervical cancer screening system is proposed that uses a convolutional neural network trained on Cervical Cells database. The training of the network is accomplished through transfer learning, whereby initializing weights are obtained from the training on ImageNet dataset. After fine-tuning the network on the Cervical Cells database, the feature vector is extracted from the last fully connected layer of convolutional neural network. For final classification/screening of the cell samples, three different classifiers are proposed including Softmax regression (SR), Support vector machine (SVM), and GentleBoost ensemble of decision trees (GEDT). The performance of the proposed screening system is evaluated for two different testing protocols, namely, 2-class problem and 7-class problem, on the Herlev database. Classification accuracies of SR, SVM, and GEDT for the 2-class problem are found to be 98.8%, 99.5%, and 99.6%, respectively, while for the 7-class problem, they are 97.21%, 98.12%, and 98.85%, respectively. These results show that the proposed system provides better performance than its previous counterparts under various testing conditions.

Download Full-text

Convolutional neural network-based transfer learning and knowledge distillation using multi-subject data in motor imagery BCI

2017 8th International IEEE/EMBS Conference on Neural Engineering (NER) ◽

10.1109/ner.2017.8008420 ◽

2017 ◽

Cited By ~ 12

Author(s):

Siavash Sakhavi ◽

Cuntai Guan

Keyword(s):

Neural Network ◽

Convolutional Neural Network ◽

Transfer Learning ◽

Motor Imagery ◽

Knowledge Distillation

Download Full-text

CONVOLUTIONAL NEURAL NETWORK BASED ALGORITHM FOR CECUM ACHIEVEMENT CONFIRMATION

10.1055/s-0040-1705059 ◽

2020 ◽

Author(s):

S Kashin ◽

D Zavyalov ◽

A Rusakov ◽

V Khryashchev ◽

A Lebedev

Keyword(s):

Neural Network ◽

Convolutional Neural Network

Download Full-text

No-Reference Utility Estimation with a Convolutional Neural Network

Electronic Imaging ◽

10.2352/issn.2470-1173.2018.09.iriacv-202 ◽

2018 ◽

Vol 2018 (9) ◽

pp. 202-1-202-6 ◽

Cited By ~ 2

Author(s):

Edward T. Scott ◽

Sheila S. Hemami

Keyword(s):

Neural Network ◽

Convolutional Neural Network ◽

Utility Estimation

Download Full-text

Non-Blind Image Deconvolution Based on “Ringing” Removal Using Convolutional Neural Network

Electronic Imaging ◽

10.2352/issn.2470-1173.2020.10.ipas-180 ◽

2020 ◽

Vol 2020 (10) ◽

pp. 181-1-181-7

Author(s):

Takahiro Kudo ◽

Takanori Fujisawa ◽

Takuro Yamaguchi ◽

Masaaki Ikehara

Keyword(s):

Neural Network ◽

Convolutional Neural Network ◽

Network Architecture ◽

Large Scale ◽

Blind Deconvolution ◽

Training Dataset ◽

Image Deconvolution ◽

Classic Problem ◽

Key Points ◽

Blind Image

Image deconvolution has been an important issue recently. It has two kinds of approaches: non-blind and blind. Non-blind deconvolution is a classic problem of image deblurring, which assumes that the PSF is known and does not change universally in space. Recently, Convolutional Neural Network (CNN) has been used for non-blind deconvolution. Though CNNs can deal with complex changes for unknown images, some CNN-based conventional methods can only handle small PSFs and does not consider the use of large PSFs in the real world. In this paper we propose a non-blind deconvolution framework based on a CNN that can remove large scale ringing in a deblurred image. Our method has three key points. The first is that our network architecture is able to preserve both large and small features in the image. The second is that the training dataset is created to preserve the details. The third is that we extend the images to minimize the effects of large ringing on the image borders. In our experiments, we used three kinds of large PSFs and were able to observe high-precision results from our method both quantitatively and qualitatively.

Download Full-text