A lightweight hybrid model for the automatic recognition of uterine fibroid ultrasound images based on deep learning.

PURPOSE: Uterine fibroids (UF) are the most frequent tumors in ladies and can pose an enormous threat to complications, such as miscarriage. The accuracy of prognosis may also be affected by way of doctor inexperience and fatigue, underscoring the want for automatic classification fashions that can analyze UF from a giant wide variety of images. METHODS: A hybrid model has been proposed that combines the MobileNetV2 community and deep convolutional generative adversarial networks (DCGAN) into useful resources for medical practitioners in figuring out UF and evaluating its characteristics. Real-time automated classification of UF can aid in diagnosing the circumstance and minimizing subjective errors. The DCGAN science is utilized for superior statistics augmentation to create first-rate UF images, which are labeled into UF and non-uterine-fibroid (NUF) classes. The MobileNetV2 model then precisely classifies the photos based totally on this data. RESULTS: The overall performance of the hybrid model contrasts with different models. The hybrid model achieves a real-time classification velocity of 40 frames per second (FPS), an accuracy of 97.45%, and an F1 rating of 0.9741. CONCLUSION: By using this deep learning hybrid approach, we address the shortcomings of the current classification methods of uterine fibroid.

A Fault Diagnosis Strategy for Analog Circuits with Limited Samples Based on the Combination of the Transformer and Generative Models.

As a pivotal integral component within electronic systems, analog circuits are of paramount importance for the timely detection and precise diagnosis of their faults. However, the objective reality of limited fault samples in operational devices with analog circuitry poses challenges to the direct applicability of existing diagnostic methods. This study proposes an innovative approach for fault diagnosis in analog circuits by integrating deep convolutional generative adversarial networks (DCGANs) with the Transformer architecture, addressing the problem of insufficient fault samples affecting diagnostic performance. Firstly, the employment of the continuous wavelet transform in combination with Morlet wavelet basis functions serves as a means to derive time-frequency images, enhancing fault feature recognition while converting time-domain signals into time-frequency representations. Furthermore, the augmentation of datasets utilizing deep convolutional GANs is employed to generate synthetic time-frequency signals from existing fault data. The Transformer-based fault diagnosis model was trained using a mixture of original signals and generated signals, and the model was subsequently tested. Through experiments involving single and multiple fault scenarios in three simulated circuits, a comparative analysis of the proposed approach was conducted with a number of established benchmark methods, and its effectiveness in various scenarios was evaluated. In addition, the ability of the proposed fault diagnosis technique was investigated in the presence of limited fault data samples. The outcome reveals that the proposed diagnostic method exhibits a consistently high overall accuracy of over 96% in diverse test scenarios. Moreover, it delivers satisfactory performance even when real sample sizes are as small as 150 instances in various fault categories.

Geometry Sampling-Based Adaption to DCGAN for 3D Face Generation.

Despite progress in the past decades, 3D shape acquisition techniques are still a threshold for various 3D face-based applications and have therefore attracted extensive research. Moreover, advanced 2D data generation models based on deep networks may not be directly applicable to 3D objects because of the different dimensionality of 2D and 3D data. In this work, we propose two novel sampling methods to represent 3D faces as matrix-like structured data that can better fit deep networks, namely (1) a geometric sampling method for the structured representation of 3D faces based on the intersection of iso-geodesic curves and radial curves, and (2) a depth-like map sampling method using the average depth of grid cells on the front surface. The above sampling methods can bridge the gap between unstructured 3D face models and powerful deep networks for an unsupervised generative 3D face model. In particular, the above approaches can obtain the structured representation of 3D faces, which enables us to adapt the 3D faces to the Deep Convolution Generative Adversarial Network (DCGAN) for 3D face generation to obtain better 3D faces with different expressions. We demonstrated the effectiveness of our generative model by producing a large variety of 3D faces with different expressions using the two novel down-sampling methods mentioned above.

Waste Detection System Based on Data Augmentation and YOLO_EC.

The problem of waste classification has been a major concern for both the government and society, and whether waste can be effectively classified will affect the sustainable development of human society. To perform fast and efficient detection of waste targets in the sorting process, this paper proposes a data augmentation + YOLO_EC waste detection system. First of all, because of the current shortage of multi-objective waste classification datasets, the heavy workload of human data collection, and the limited improvement of data features by traditional data augmentation methods, DCGAN (deep convolution generative adversarial networks) was optimized by improving the loss function, and an image-generation model was established to realize the generation of multi-objective waste images; secondly, with YOLOv4 (You Only Look Once version 4) as the basic model, EfficientNet is used as the backbone feature extraction network to realize the light weight of the algorithm, and at the same time, the CA (coordinate attention) attention mechanism is introduced to reconstruct the MBConv module to filter out high-quality information and enhance the feature extraction ability of the model. Experimental results show that on the HPU_WASTE dataset, the proposed model outperforms other models in both data augmentation and waste detection.

Dynamic and Real-Time Object Detection Based on Deep Learning for Home Service Robots.

Home service robots operating indoors, such as inside houses and offices, require the real-time and accurate identification and location of target objects to perform service tasks efficiently. However, images captured by visual sensors while in motion states usually contain varying degrees of blurriness, presenting a significant challenge for object detection. In particular, daily life scenes contain small objects like fruits and tableware, which are often occluded, further complicating object recognition and positioning. A dynamic and real-time object detection algorithm is proposed for home service robots. This is composed of an image deblurring algorithm and an object detection algorithm. To improve the clarity of motion-blurred images, the DA-Multi-DCGAN algorithm is proposed. It comprises an embedded dynamic adjustment mechanism and a multimodal multiscale fusion structure based on robot motion and surrounding environmental information, enabling the deblurring processing of images that are captured under different motion states. Compared with DeblurGAN, DA-Multi-DCGAN had a 5.07 improvement in Peak Signal-to-Noise Ratio (PSNR) and a 0.022 improvement in Structural Similarity (SSIM). An AT-LI-YOLO method is proposed for small and occluded object detection. Based on depthwise separable convolution, this method highlights key areas and integrates salient features by embedding the attention module in the AT-Resblock to improve the sensitivity and detection precision of small objects and partially occluded objects. It also employs a lightweight network unit Lightblock to reduce the network's parameters and computational complexity, which improves its computational efficiency. Compared with YOLOv3, the mean average precision (mAP) of AT-LI-YOLO increased by 3.19%, and the detection precision of small objects, such as apples and oranges and partially occluded objects, increased by 19.12% and 29.52%, respectively. Moreover, the model inference efficiency had a 7 ms reduction in processing time. Based on the typical home activities of older people and children, the dataset Grasp-17 was established for the training and testing of the proposed method. Using the TensorRT neural network inference engine of the developed service robot prototype, the proposed dynamic and real-time object detection algorithm required 29 ms, which meets the real-time requirement of smooth vision.

Integration of Deep Learning Network and Robot Arm System for Rim Defect Inspection Application.

Automated inspection has proven to be the most effective approach to maintaining quality in industrial-scale manufacturing. This study employed the eye-in-hand architecture in conjunction with deep learning and convolutional neural networks to automate the detection of defects in forged aluminum rims for electric vehicles. RobotStudio software was used to simulate the environment and path trajectory for a camera installed on an ABB robot arm to capture 3D images of the rims. Four types of surface defects were examined: (1) dirt spots, (2) paint stains, (3) scratches, and (4) dents. Generative adversarial network (GAN) and deep convolutional generative adversarial networks (DCGAN) were used to generate additional images to expand the depth of the training dataset. We also developed a graphical user interface and software system to mark patterns associated with defects in the images. The defect detection algorithm based on YOLO algorithms made it possible to obtain results more quickly and with higher mean average precision (mAP) than that of existing methods. Experiment results demonstrated the accuracy and efficiency of the proposed system. Our developed system has been shown to be a helpful rim defective detection system for industrial applications.

Lesion Image Synthesis Using DCGANs for Metastatic Liver Cancer Detection.

This chapter proposes a method to detect metastatic liver cancer from X-ray CT images using a convolutional neural network (CNN). The proposed method generates various lesion images by the combination of three kinds of generation methods: (1) synthesis using Poisson Blending, (2) generation based on CT value distributions, and (3) generation using deep convolutional generative adversarial networks (DCGANs). The proposed method constructs two kinds of detectors by using synthetic (fake) lesion images generated by the methods as well as real ones. One of the detectors is a 2D CNN for detecting candidate regions in a CT image, and the other is a 3D CNN for validating the candidate regions. Experimental results showed that the proposed method gave 0.30 improvement from 0.65 to 0.95 in terms of the detection rate, and 0.70 improvement from 0.90 to 0.20 in terms of the number of false detections per case. From the results, we confirmed the effectiveness of the proposed method.

Data Augmentation for Motor Imagery Signal Classification Based on a Hybrid Neural Network.

As an important paradigm of spontaneous brain-computer interfaces (BCIs), motor imagery (MI) has been widely used in the fields of neurological rehabilitation and robot control. Recently, researchers have proposed various methods for feature extraction and classification based on MI signals. The decoding model based on deep neural networks (DNNs) has attracted significant attention in the field of MI signal processing. Due to the strict requirements for subjects and experimental environments, it is difficult to collect large-scale and high-quality electroencephalogram (EEG) data. However, the performance of a deep learning model depends directly on the size of the datasets. Therefore, the decoding of MI-EEG signals based on a DNN has proven highly challenging in practice. Based on this, we investigated the performance of different data augmentation (DA) methods for the classification of MI data using a DNN. First, we transformed the time series signals into spectrogram images using a short-time Fourier transform (STFT). Then, we evaluated and compared the performance of different DA methods for this spectrogram data. Next, we developed a convolutional neural network (CNN) to classify the MI signals and compared the classification performance of after DA. The Fréchet inception distance (FID) was used to evaluate the quality of the generated data (GD) and the classification accuracy, and mean kappa values were used to explore the best CNN-DA method. In addition, analysis of variance (ANOVA) and paired t-tests were used to assess the significance of the results. The results showed that the deep convolutional generative adversarial network (DCGAN) provided better augmentation performance than traditional DA methods: geometric transformation (GT), autoencoder (AE), and variational autoencoder (VAE) (p < 0.01). Public datasets of the BCI competition IV (datasets 1 and 2b) were used to verify the classification performance. Improvements in the classification accuracies of 17% and 21% (p < 0.01) were observed after DA for the two datasets. In addition, the hybrid network CNN-DCGAN outperformed the other classification methods, with average kappa values of 0.564 and 0.677 for the two datasets.

Parkinson's Disease EMG Data Augmentation and Simulation with DCGANs and Style Transfer.

This paper proposes two new data augmentation approaches based on Deep Convolutional Generative Adversarial Networks (DCGANs) and Style Transfer for augmenting Parkinson's Disease (PD) electromyography (EMG) signals. The experimental results indicate that the proposed models can adapt to different frequencies and amplitudes of tremor, simulating each patient's tremor patterns and extending them to different sets of movement protocols. Therefore, one could use these models for extending the existing patient dataset and generating tremor simulations for validating treatment approaches on different movement scenarios.

A new method for deep learning detection of defects in X-ray images of pressure vessel welds.

Given that defect detection in weld X-ray images is a critical aspect of pressure vessel manufacturing and inspection, accurate differentiation of the type, distribution, number, and area of defects in the images serves as the foundation for judging weld quality, and the segmentation method of defects in digital X-ray images is the core technology for differentiating defects. Based on the publicly available weld seam dataset GDX-ray, this paper proposes a complete technique for fault segmentation in X-ray pictures of pressure vessel welds. The key works are as follows: (1) To address the problem of a lack of defect samples and imbalanced distribution inside GDX-ray, a DA-DCGAN based on a two-channel attention mechanism is devised to increase sample data. (2) A convolutional block attention mechanism is incorporated into the coding layer to boost the accuracy of small-scale defect identification. The proposed MAU-Net defect semantic segmentation network uses multi-scale even convolution to enhance large-scale features. The proposed method can mask electrostatic interference and non-defect-class parts in the actual weld X-ray images, achieve an average segmentation accuracy of 84.75% for the GDX-ray dataset, segment and accurately rate the valid defects with a correct rating rate of 95%, and thus realize practical value in engineering.

A Comparative Analysis of the Novel Conditional Deep Convolutional Neural Network Model, Using Conditional Deep Convolutional Generative Adversarial Network-Generated Synthetic and Augmented Brain Tumor Datasets for Image Classification.

Disease prediction is greatly challenged by the scarcity of datasets and privacy concerns associated with real medical data. An approach that stands out to circumvent this hurdle is the use of synthetic data generated using Generative Adversarial Networks (GANs). GANs can increase data volume while generating synthetic datasets that have no direct link to personal information. This study pioneers the use of GANs to create synthetic datasets and datasets augmented using traditional augmentation techniques for our binary classification task. The primary aim of this research was to evaluate the performance of our novel Conditional Deep Convolutional Neural Network (C-DCNN) model in classifying brain tumors by leveraging these augmented and synthetic datasets. We utilized advanced GAN models, including Conditional Deep Convolutional Generative Adversarial Network (DCGAN), to produce synthetic data that retained essential characteristics of the original datasets while ensuring privacy protection. Our C-DCNN model was trained on both augmented and synthetic datasets, and its performance was benchmarked against state-of-the-art models such as ResNet50, VGG16, VGG19, and InceptionV3. The evaluation metrics demonstrated that our C-DCNN model achieved accuracy, precision, recall, and F1 scores of 99% on both synthetic and augmented images, outperforming the comparative models. The findings of this study highlight the potential of using GAN-generated synthetic data in enhancing the training of machine learning models for medical image classification, particularly in scenarios with limited data available. This approach not only improves model accuracy but also addresses privacy concerns, making it a viable solution for real-world clinical applications in disease prediction and diagnosis.

Synthesis of sEMG Signals for Hand Gestures Using a 1DDCGAN.

The emergence of modern prosthetics controlled by bio-signals has been facilitated by AI and microchip technology innovations. AI algorithms are trained using sEMG produced by muscles during contractions. The data acquisition procedure may result in discomfort and fatigue, particularly for amputees. Furthermore, prosthetic companies restrict sEMG signal exchange, limiting data-driven research and reproducibility. GANs present a viable solution to the aforementioned concerns. GANs can generate high-quality sEMG, which can be utilised for data augmentation, decrease the training time required by prosthetic users, enhance classification accuracy and ensure research reproducibility. This research proposes the utilisation of a one-dimensional deep convolutional GAN (1DDCGAN) to generate the sEMG of hand gestures. This approach involves the incorporation of dynamic time wrapping, fast Fourier transform and wavelets as discriminator inputs. Two datasets were utilised to validate the methodology, where five windows and increments were utilised to extract features to evaluate the synthesised sEMG quality. In addition to the traditional classification and augmentation metrics, two novel metrics-the Mantel test and the classifier two-sample test-were used for evaluation. The 1DDCGAN preserved the inter-feature correlations and generated high-quality signals, which resembled the original data. Additionally, the classification accuracy improved by an average of 1.21-5%.

Morphology and mechanical performance of dental crown designed by 3D-DCGAN.

OBJECTIVES: This study utilised an Artificial Intelligence (AI) method, namely 3D-Deep Convolutional Generative Adversarial Network (3D-DCGAN), which is one of the true 3D machine learning methods, as an automatic algorithm to design a dental crown. METHODS: Six hundred sets of digital casts containing mandibular second premolars and their adjacent and antagonist teeth obtained from healthy personnel were machine-learned using 3D-DCGAN. Additional 12 sets of data were used as the test dataset, whereas the natural second premolars in the test dataset were compared with the designs in (1) 3D-DCGAN, (2) CEREC Biogeneric, and (3) CAD for morphological parameters of 3D similarity, cusp angle, occlusal contact point number and area, and in silico fatigue simulations with finite element (FE) using lithium disilicate material. RESULTS: The 3D-DCGAN design and natural teeth had the lowest discrepancy in morphology compared with the other groups (root mean square value = 0.3611). The Biogeneric design showed a significantly (p < 0.05) higher cusp angle (67.11°) than that of the 3D-DCGAN design (49.43°) and natural tooth (54.05°). No significant difference was observed in the number and area of occlusal contact points among the four groups. FE analysis showed that the 3D-DCGAN design had the best match to the natural tooth regarding the stress distribution in the crown. The 3D-DCGAN design was subjected to 26.73 MPa and the natural tooth was subjected to 23.97 MPa stress at the central fossa area under physiological occlusal force (300 N); the two groups showed similar fatigue lifetimes (F-N curve) under simulated cyclic loading of 100-400 N. Designs with Biogeneric or technician would yield respectively higher or lower fatigue lifetime than natural teeth. SIGNIFICANCE: This study demonstrated that 3D-DCGAN could be utilised to design personalised dental crowns with high accuracy that can mimic both the morphology and biomechanics of natural teeth.

Staging study of single-channel sleep EEG signals based on data augmentation.

Introduction: Accurate sleep staging is an essential basis for sleep quality assessment and plays an important role in sleep quality research. However, the occupancy of different sleep stages is unbalanced throughout the sleep process, which makes the EEG datasets of different sleep stages have a class imbalance, which will eventually affect the automatic assessment of sleep stages. Method: In this paper, we propose a Residual Dense Block and Deep Convolutional Generative Adversarial Network (RDB-DCGAN) data augmentation model based on the DCGAN and RDB, which takes two-dimensional continuous wavelet time-frequency maps as input, expands the minority class of sleep EEG data and later performs sleep staging by Convolutional Neural Network (CNN). Results and discussion: The results of the CNN classification comparison test with the publicly available dataset Sleep-EDF show that the overall sleep staging accuracy of each stage after data augmentation is improved by 6%, especially the N1 stage, which has low classification accuracy due to less original data, also has a significant improvement of 19%. It is fully verified that data augmentation by improving the DCGAN model can effectively improve the classification problem of the class imbalance sleep dataset.

Improved Semantic Image Inpainting Method with Deep Convolution Generative Adversarial Networks.

With the development of generative adversarial networks (GANs), more and more researchers apply them to image inpainting technologies. However, many existing approaches caused some inpainting images to be unclear or even restore failures due to a failure to keep the consistency of the inpainted content and structures in line with the surroundings. In this article, we propose the Improved Semantic Image Inpainting Method with Deep Convolution GANs, which can resolve this inconsistency. In the proposed method, we design a patch discriminator and contextual loss to jointly perform the accuracy and effectiveness for image inpainting. In addition, we also designed a consistency loss based on deep convolutional neural networks to constrain the difference between the generated image and the original image in the feature space. Our proposed method improves the details and authenticity effectively for the inpainting images. We evaluate our proposed method on two different datasets, and the result shows that our proposed method achieves state-of-the-art results.

Deep Convolutional Generative Adversarial Network-Based EMG Data Enhancement for Hand Motion Classification.

The acquisition of bio-signal from the human body requires a strict experimental setup and ethical approvements, which leads to limited data for the training of classifiers in the era of big data. It will change the situation if synthetic data can be generated based on real data. This article proposes such a kind of multiple channel electromyography (EMG) data enhancement method using a deep convolutional generative adversarial network (DCGAN). The generation procedure is as follows: First, the multiple channels of EMG signals within sliding windows are converted to grayscale images through matrix transformation, normalization, and histogram equalization. Second, the grayscale images of each class are used to train DCGAN so that synthetic grayscale images of each class can be generated with the input of random noises. To evaluate whether the synthetic data own the similarity and diversity with the real data, the classification accuracy index is adopted in this article. A public EMG dataset (that is, ISR Myo-I) for hand motion recognition is used to prove the usability of the proposed method. The experimental results show that adding synthetic data to the training data has little effect on the classification performance, indicating the similarity between real data and synthetic data. Moreover, it is also noted that the average accuracy (five classes) is slightly increased by 1%-2% for support vector machine (SVM) and random forest (RF), respectively, with additional synthetic data for training. Although the improvement is not statistically significant, it implies that the generated data by DCGAN own its new characteristics, and it is possible to enrich the diversity of the training dataset. In addition, cross-validation analysis shows that the synthetic samples have large inter-class distance, reflected by higher cross-validation accuracy of pure synthetic sample classification. Furthermore, this article also demonstrates that histogram equalization can significantly improve the performance of EMG-based hand motion recognition.

Deep Convolution Generative Adversarial Network-Based Electroencephalogram Data Augmentation for Post-Stroke Rehabilitation with Motor Imagery.

The motor imagery brain-computer interface (MI-BCI) system is currently one of the most advanced rehabilitation technologies, and it can be used to restore the motor function of stroke patients. The deep learning algorithms in the MI-BCI system require lots of training samples, but the electroencephalogram (EEG) data of stroke patients is quite scarce. Therefore, the expansion of EEG data has become an important part of stroke clinical rehabilitation research. In this paper, a deep convolution generative adversarial network (DCGAN) model is proposed to generate artificial EEG data and further expand the scale of the stroke dataset. First, multichannel one-dimensional EEG data is converted into a two-dimensional EEG spectrogram using EEG2Image based on the modified S-transform. Then, DCGAN is used to artificially generate EEG data based on MI. Finally, the validity of the generated artificial EEG data is proved. This paper preliminarily indicates that generating artificial stroke data is a promising strategy, which contributes to the further development of stroke clinical rehabilitation.

Generative Adversarial Network Image Synthesis Method for Skin Lesion Generation and Classification.

BACKGROUND: One of the common limitations in the treatment of cancer is in the early detection of this disease. The customary medical practice of cancer examination is a visual examination by the dermatologist followed by an invasive biopsy. Nonetheless, this symptomatic approach is timeconsuming and prone to human errors. An automated machine learning model is essential to capacitate fast diagnoses and early treatment. OBJECTIVE: The key objective of this study is to establish a fully automatic model that helps Dermatologists in skin cancer handling process in a way that could improve skin lesion classification accuracy. METHOD: The work is conducted following an implementation of a Deep Convolutional Generative Adversarial Network (DCGAN) using the Python-based deep learning library Keras. We incorporated effective image filtering and enhancement algorithms such as bilateral filter to enhance feature detection and extraction during training. The Deep Convolutional Generative Adversarial Network (DCGAN) needed slightly more fine-tuning to ripe a better return. Hyperparameter optimization was utilized for selecting the best-performed hyperparameter combinations and several network hyperparameters. In this work, we decreased the learning rate from the default 0.001 to 0.0002, and the momentum for Adam optimization algorithm from 0.9 to 0.5, in trying to reduce the instability issues related to GAN models and at each iteration the weights of the discriminative and generative network were updated to balance the loss between them. We endeavour to address a binary classification which predicts two classes present in our dataset, namely benign and malignant. More so, some wellknown metrics such as the receiver operating characteristic -area under the curve and confusion matrix were incorporated for evaluating the results and classification accuracy. RESULTS: The model generated very conceivable lesions during the early stages of the experiment and we could easily visualise a smooth transition in resolution along the way. Thus, we have achieved an overall test accuracy of 93.5% after fine-tuning most parameters of our network. CONCLUSION: This classification model provides spatial intelligence that could be useful in the future for cancer risk prediction. Unfortunately, it is difficult to generate high quality images that are much like the synthetic real samples and to compare different classification methods given the fact that some methods use non-public datasets for training.

A Bi-fold Approach to Detect and Classify COVID-19 X-Ray Images and Symptom Auditor.

In this paper, we propose an ensemble-based transfer learning method to predict the X-ray image of a COVID-19 affected person. We have used a weighted Euclidean distance average as the parameter to ensemble the transfer learning model viz. ResNet50, VGG16, VGG19, Xception, and InceptionV3. Image augmentations have been carried out using generative adversarial network modelling. We took 784 training images, and 278 test images to validate our model accuracy, and the accuracy of our proposed model was around 98.67% for the training data set and 95.52% for the test data set. Along with that, we also propose a genetic algorithm optimized classification algorithm, to analyze the symptoms of COVID-19 for low, medium, and high-risk patients. The accuracy for the optimized set overshadowed the accuracy of un-optimized classification, and the optimized accuracy is as high as 88.96% for the optimized model. The novelty of this paper lies in the bi-sided model of the paper, i.e., we propose two major models, and one is the genetic algorithm optimized model to analyze the symptoms for a patient of varied risk and the other is to classify the X-ray image using an ensemble-based transfer learning model.

Combining DC-GAN with ResNet for blood cell image classification.

In medicine, white blood cells (WBCs) play an important role in the human immune system. The different types of WBC abnormalities are related to different diseases so that the total number and classification of WBCs are critical for clinical diagnosis and therapy. However, the traditional method of white blood cell classification is to segment the cells, extract features, and then classify them. Such method depends on the good segmentation, and the accuracy is not high. Moreover, the insufficient data or unbalanced samples can cause the low classification accuracy of model by using deep learning in medical diagnosis. To solve these problems, this paper proposes a new blood cell image classification framework which is based on a deep convolutional generative adversarial network (DC-GAN) and a residual neural network (ResNet). In particular, we introduce a new loss function which is improved the discriminative power of the deeply learned features. The experiments show that our model has a good performance on the classification of WBC images, and the accuracy reaches 91.7%. Graphical Abstract Overview of the proposed method, we use the deep convolution generative adversarial networks (DC-GAN) to generate new samples that are used as supplementary input to a ResNet, the transfer learning method is used to initialize the parameters of the network, the output of the DC-GAN and the parameters are applied the final classification network. In particular, we introduced a modified loss function for classification to increase inter-class variations and decrease intra-class differences.