CNN Architecture Research Articles

A novel benign and malignant classification model for lung nodules based on multi-scale interleaved fusion integrated network

AbstractOne of the precursors of lung cancer is the presence of lung nodules, and accurate identification of their benign or malignant nature is important for the long-term survival of patients. With the development of artificial intelligence, deep learning has become the main method for lung nodule classification. However, successful deep learning models usually require large number of parameters and carefully annotated data. In the field of medical images, the availability of such data is usually limited, which makes deep networks often perform poorly on new test data. In addition, the model based on the linear stacked single branch structure hinders the extraction of multi-scale features and reduces the classification performance. In this paper, to address this problem, we propose a lightweight interleaved fusion integration network with multi-scale feature learning modules, called MIFNet. The MIFNet consists of a series of MIF blocks that efficiently combine multiple convolutional layers containing 1 × 1 and 3 × 3 convolutional kernels with shortcut links to extract multiscale features at different levels and preserving them throughout the block. The model has only 0.7 M parameters and requires low computational cost and memory space compared to many ImageNet pretrained CNN architectures. The proposed MIFNet conducted exhaustive experiments on the reconstructed LUNA16 dataset, achieving impressive results with 94.82% accuracy, 97.34% F1 value, 96.74% precision, 97.10% sensitivity, and 84.75% specificity. The results show that our proposed deep integrated network achieves higher performance than pre-trained deep networks and state-of-the-art methods. This provides an objective and efficient auxiliary method for accurately classifying the type of lung nodule in medical images.

Segmentation and classification of white blood SMEAR images using modified CNN architecture

The classification and recognition of leukocytes or WBCs in blood smear images presents a key role in the corresponding diagnosis of specific diseases, such as leukemia, tumor, hematological disorders, etc. The computerized framework for automated segmentation & classification of WBCs nucleus contributes an important role for the recognition of WBCs related disorders. Therefore, this work emphasizes WBCs nucleus segmentation using modified U-Net architecture and the segmented WBCs nucleus are further classified into their subcategory i.e., basophil, eosinophil, neutrophil, monocyte and lymphocyte. The classification and nucleus characterization task has been performed using VGGNet and MobileNet V2 architecture. Initially, collected instances are passed to the preprocessing phase for image rescaling and normalization. The rescaled and normalized instances are passed to the U-Net model for nucleus segmentation. Extracted nucleus are forwarded to the classification phase for their class identifications. Furthermore, the functioning of the intended design will be compared with other modern methods. By the end of this study a successful model classifying various nucleus morphologies such as Basophil, Eosinophil, Lymphocyte, Monocyte and Neutrophil was obtained where overall test accuracy achieved was 97.0% for VGGNet classifier and 94.0% for MobileNet V2 classifier.

Exploring the Efficacy of Various CNN Architectures in Diagnosing Oral Cancer from Squamous Cell Carcinoma

Oral cancer can result from mutations in cells located in the lips or mouth. Diagnosing oral cavity squamous cell carcinoma (OCSCC) is particularly challenging, often occurring at advanced stages. To address this, computer-aided diagnosis methods are increasingly being used. In this work, a deep learning-based approach utilizing models such as VGG16, ResNet50, LeNet-5, MobileNetV2, and Inception V3 is presented. NEOR and OCSCC datasets were used for feature extraction, with virtual slide images divided into tiles and classified as normal or squamous cell cancer. Performance metrics like accuracy, F1-score, AUC, precision, and recall were analyzed to determine the prerequisites for optimal CNN performance. The proposed CNN approaches were effective for classifying OCSCC and oral dysplasia, with the highest accuracy of 95.41% achieved using MobileNetV2. Key findingsDeep learning models, particularly MobileNetV2, achieved high classification accuracy (95.41%) for OCSCC.CNN-based methods show promise for early-stage OCSCC and oral dysplasia diagnosis. Performance parameters like precision, recall, and F1-score help optimize CNN model selection for this task.

Recovering Permuted Sequential Features for effective Reinforcement Learning.

When applying Reinforcement Learning (RL) to the real-world visual tasks, two major challenges necessitate consideration: sample inefficiency and limited generalization. To address the above two challenges, previous works focus primarily on learning semantic information from the visual state for improving sample efficiency, but they do not explicitly learn other valuable aspects, such as spatial information. Moreover, they improve generalization by learning representations that are invariant to alterations of task-irrelevant variables, without considering task-relevant variables. To enhance sample efficiency and generalization of the base RL algorithm in visual tasks, we propose an auxiliary task called Recovering Permuted Sequential Features (RPSF). Our method enhances generalization by learning the spatial structure information of the agent, which can mitigate the effects of changes in both task-relevant and task-irrelevant variables. Moreover, it explicitly learns both semantic and spatial information from the visual state by disordering and subsequently recovering a sequence of features to generate more holistic representations, thereby improving sample efficiency. Extensive experiments demonstrate that our method significantly improves the sample efficiency and generalization of the base RL algorithm and outperforms various state-of-the-art baselines across diverse tasks in unseen environments. Furthermore, our method exhibits compatibility with both CNN and Transformer architectures.

ACMamba: A State Space Model-Based Approach for Multi-Weather Degraded Image Restoration

In computer vision, eliminating the effects of adverse weather conditions such as rain, snow, and fog on images is a key research challenge. Existing studies primarily focus on image restoration for single weather types, while methods addressing image restoration under multiple combined weather conditions remain relatively scarce. Furthermore, current mainstream restoration networks, mostly based on Transformer and CNN architectures, struggle to achieve an effective balance between global receptive field and computational efficiency, limiting their performance in practical applications. This study proposes ACMamba, an end-to-end lightweight network based on selective state space models, aimed at achieving image restoration under multiple weather conditions using a unified set of parameters. Specifically, we design a novel Visual State Space Module (VSSM) and a Spatially Aware Feed-Forward Network (SAFN), which organically combine the local feature extraction capabilities of convolutions with the long-range dependency modeling capabilities of selective state space models (SSMs). This combination significantly improves computational efficiency while maintaining a global receptive field, enabling effective application of the Mamba architecture to multi-weather image restoration tasks. Comprehensive experiments demonstrate that our proposed approach significantly outperforms existing methods for both specific and multi-weather tasks across multiple benchmark datasets, showcasing its efficient long-range modeling potential in multi-weather image restoration tasks.

AN APPROACH FOR LUNG CANCER DETECTION AND CLASSIFICATION USING LENET-DENSENET

Lung cancer is a malicious disorder that affects millions of people around the globe. Premature detection and classification of malignant cells can gradually improve the rate of survival of the patients. The study proposed an innovative strategy for lung cancer identification and classification by utilizing the hybrid deep learning (DL) technique. The presented strategy contains two major parts, segmentation and classification. Attention gated (AG) network is used for segmenting the affected regions. AGs guide a method’s attention to significant regions while repressing the activation of features in non-related fields. This augments the model’s representative influence without significantly increasing computing cost or the number of model parameters. For the lung cancer classification, by combining the architectures of LeNet and Dense Convolutional Network (DenseNet), the proposed LeNet–DenseNet was developed. LeNet and DenseNet are two popular CNN architectures known for the classification of images. LeNet comprises alternating and convolutional layers, followed by fully connected layers. As per the idea of densely connected layers, each layer in DenseNet, which is a recent CNN architecture, is associated with other layers in a feed-forward fashion. This allows for better feature reuse and gradient flow and lessens the parameter count. The findings stated that the proposed method outperforms when compared to other traditional strategies corresponding to specificity, sensitivity and accuracy. The presented approach assists in early-stage detection of high-accuracy lung cancer, which is considered crucial for the treatment procedures in time. The findings stipulated that the presented LeNet–DenseNet model procured accuracy of 94.8%, sensitivity of 96.8% and specificity of 93.5%.

Chest x-ray images: transfer learning model in COVID-19 detection.

This research aims to develop an effective algorithm for diagnosing COVID-19 in chest X-rays using the transfer learning method and support vector machines. In total, data was collected from 10 clinics, including both large city hospitals and smaller medical institutions. This ensured a diverse range of geographical and demographic information in the sample. An extensive data set was collected, including 10,000 chest X-ray images. 5000 images represent normal cases, 3993 images represent pneumonia cases, and 1007 images represent COVID-19 cases. Machine learning methods were applied to develop a classification model, and the results were compared with seven state-of-the-art models and a lightweight CNN architecture. The results showed that the proposed method achieves high accuracy values (Accuracy): 0.95 for COVID-19, 0.89 for pneumonia, and 0.92 for normal images (p < 0.05). Comparison with other models demonstrates statistically significant superiority of our method in accuracy across all three classes. The EfficientNet-B0 model surpasses our method only in accuracy for normal images with p < 0.01, confirming the advantages of our method. Our method demonstrates high sensitivity values (Sensitivity): 0.96 for COVID-19, 0.88 for pneumonia, and 0.93 for normal images (p < 0.05), outperforming most of the compared models. Correlation analysis showed Pearson coefficients of 0.92, 0.89, and 0.94 for COVID-19, pneumonia, and normal images, respectively, confirming a high degree of consistency between predicted and true class labels. In addition, the model was validated on external datasets to assess its generalizability. This validation confirmed its high level of effectiveness in a variety of clinical settings. This study confirms the importance of applying machine learning methods in medical applications and opens new perspectives for early diagnosis of infectious diseases. The practical application of the obtained results can enhance the efficiency of diagnosis and control the spread of COVID-19, as well as contribute to the development of innovative methods in medical practice.

Deep learning approaches for classification of copper-containing metal scrap in recycling processes

Separating copper from iron scrap is a critical operation in metal recycling and achieving this with low cost sensoric equipment like RGB cameras instead of XRF/XRT is becoming increasingly attractive. In this article, the groundwork for creating an image classification model to separate copper from iron scrap has been performed. Twenty of the most common and most easily available CNN architectures were trained on 2200 metal scrap specimens and evaluated inline on a sensor-based sorting rig for their prediction accuracy and their inference latency to mimic real circumstances in an industrial setting. Out of these evaluated architectures, DenseNet-201 with 98% accuracy in inline tests is recommended if potent hardware is available. Otherwise AlexNet with 92% accuracy or MobileNet-V2 with 90% accuracy are recommended for further investigation and model creation if hardware restrictions apply. Based on the presented results in this article, the initial cumbersome surveyance of the most suitable network architecture can be substantially reduced and the creation of a sorting model can be streamlined. This article thus provides the basis for creating an inline applicable sorting method for scrap metal that uses low cost sensorics equipment and can provide reasonably high accuracy in its prediction.

Classification of protected grassland habitats using deep learning architectures on Sentinel-2 satellite imagery data

This study examines the effectiveness of five deep learning models—ViTb-19, SwinV2-t, VGG-16, ResNet-50, and DenseNet-121—in distinguishing different vegetation types in the protected grasslands of Castilla y León region, Spain, following the guidelines of the Natura 92/43/CEE directive. Among the models, ResNet-50 achieved the highest weighted overall accuracy (OA) of 0.95, closely followed by SwinV2-t with an OA of 0.94, demonstrating their strong ability to detect complex patterns in satellite imagery. DenseNet-121 also performed competitively with a weighted OA of 0.93, while ViTb-19 and VGG-16 showed slightly lower performance. SwinV2-t, a transformer-based model, outperformed traditional CNN architectures in data-rich classes but faced challenges in classifying habitats with limited representation. Consequently, this study identifies these challenges that conventional transformer architectures pose in classifying certain habitats with limited representation and intricate features. Highlighting the advantages of deep learning technologies for environmental monitoring and conservation, the study provides important insights for adjusting neural network architectures for effective habitat classification. This suggests the necessity of selecting appropriate architectures such as SwinV2-t and ResNet50 to to effectively address the intricate requirements of satellite imagery analysis.

Deep models for multi-view 3D object recognition: a review

This review paper focuses on the progress of deep learning-based methods for multi-view 3D object recognition. It covers the state-of-the-art techniques in this field, specifically those that utilize 3D multi-view data as input representation. The paper provides a comprehensive analysis of the pipeline for deep learning-based multi-view 3D object recognition, including the various techniques employed at each stage. It also presents the latest developments in CNN-based and transformer-based models for multi-view 3D object recognition. The review discusses existing models in detail, including the datasets, camera configurations, view selection strategies, pre-trained CNN architectures, fusion strategies, and recognition performance. Additionally, it examines various computer vision applications that use multi-view classification. Finally, it highlights future directions, factors impacting recognition performance, and trends for the development of multi-view 3D object recognition method.

A novel embedded kernel CNN-PCFF algorithm for breast cancer pathological image classification

Early screening of breast cancer through image recognition technology can significantly increase the survival rate of patients. Therefore, breast cancer pathological image is of great significance for medical diagnosis and clinical research. In recent years, numerous deep learning models have been applied to breast cancer image classification, with deep CNN being a typical representative. Due to the use of multi-depth small convolutional kernels in mainstream CNN architectures such as VGG and Inception, the obtained image features often have high dimensionality. Although high dimensionality can bring more fine-grained features, it also increases the computational complexity of subsequent classifiers and may even lead to the curse of dimensionality and overfitting. To address these issues, a novel embedded kernel CNN principal component feature fusion (CNN-PCFF) algorithm is proposed. The constructed kernel function is embedded in the principal component analysis to form the multi-kernel principal component. Multi-kernel principal component analysis is used to fuse the high dimensional features obtained from the convolution base into some representative comprehensive variables, which are called kernel principal components, so as to achieve the purpose of dimensionality reduction. Any type of classifier can be added based on multi-kernel principal components. Through experimental analysis on two public breast cancer image datasets, the results show that the proposed algorithm can improve the performance of the current mainstream CNN architecture and subsequent classifiers. Therefore, the proposed algorithm in this paper is an effective tool for the classification of breast cancer pathological images.

Rotation Equivariant Proximal Operator for Deep Unfolding Methods in Image Restoration.

The deep unfolding approach has attracted significant attention in computer vision tasks, which well connects conventional image processing modeling manners with more recent deep learning techniques. Specifically, by establishing a direct correspondence between algorithm operators at each implementation step and network modules within each layer, one can rationally construct an almost "white box" network architecture with high interpretability. In this architecture, only the predefined component of the proximal operator, known as a proximal network, needs manual configuration, enabling the network to automatically extract intrinsic image priors in a data-driven manner. In current deep unfolding methods, such a proximal network is generally designed as a CNN architecture, whose necessity has been proven by a recent theory. That is, CNN structure substantially delivers the translational symmetry image prior, which is the most universally possessed structural prior across various types of images. However, standard CNN-based proximal networks have essential limitations in capturing the rotation symmetry prior, another universal structural prior underlying general images. This leaves a large room for further performance improvement in deep unfolding approaches. To address this issue, this study makes efforts to suggest a high-accuracy rotation equivariant proximal network that effectively embeds rotation symmetry priors into the deep unfolding framework. Especially, we deduce, for the first time, the theoretical equivariant error for such a designed proximal network with arbitrary layers under arbitrary rotation degrees. This analysis should be the most refined theoretical conclusion for such error evaluation to date and is also indispensable for supporting the rationale behind such networks with intrinsic interpretability requirements. Through experimental validation on different vision tasks, including blind image super-resolution, medical image reconstruction, and image de-raining, the proposed method is validated to be capable of directly replacing the proximal network in current deep unfolding architecture and readily enhancing their state-of-the-art performance. This indicates its potential usability in general vision tasks.

A wavelet CNN with appropriate feed-allocation and PSO optimized activations for diabetic retinopathy grading

This work modifies the architecture of conventional CNN with the integration of Multi-resolution Analysis (MRA) in a CNN framework for Diabetic Retinopathy (DR) diagnosis and grading. Here, the HF sub-bands are subjected to optimized activations and are directly fed to the fully connected layers, as it encompasses edge features. Unlike FD-Relu, the proposed function preserves significant negative coefficients, compared to the S-Relu, the proposed third-order S-Relu is optimized such that it sustains the activations in the range suitable for the wavelet coefficients. The coefficients of higher-order terms of the proposed 3rd-order S-Relu are optimized with PSO, fitting the maximum energy of the wavelet sub-bands to ensure High Frequency (HF) edge preservation. The authors re-architecture 3 different CNNs published in the Retinal Image analysis field, with spatial and wavelet inputs with optimized activations. The highest accuracy of 96% is attained with the AlexNet re-architecture, with 35,126 fundus images secured from the Kaggle dataset. As we can infer the proposed re-architecture wavelet CNN outperformed the multiscale shallow CNNs, multiscale attention net, and stacked CNNs with a 6.6, 0.3, 0.7 per cent increase in accuracy. The entire implementation of the wavelet CNN is made available under source code.

Bayesian Optimization for Sparse Neural Networks With Trainable Activation Functions.

In the literature on deep neural networks, there is considerable interest in developing activation functions that can enhance neural network performance. In recent years, there has been renewed scientific interest in proposing activation functions that can be trained throughout the learning process, as they appear to improve network performance, especially by reducing overfitting. In this paper, we propose a trainable activation function whose parameters need to be estimated. A fully Bayesian model is developed to automatically estimate from the learning data both the model weights and activation function parameters. An MCMC-based optimization scheme is developed to build the inference. The proposed method aims to solve the aforementioned problems and improve convergence time by using an efficient sampling scheme that guarantees convergence to the global maximum. The proposed scheme has been tested across a diverse datasets, encompassing both classification and regression tasks, and implemented in various CNN architectures to demonstrate its versatility and effectiveness. Promising results demonstrate the usefulness of our proposed approach in improving models accuracy due to the proposed activation function and Bayesian estimation of the parameters.

Role of inter- and extra-lesion tissue, transfer learning, and fine-tuning in the robust classification of breast lesions

Accurate and unbiased classification of breast lesions is pivotal for early diagnosis and treatment, and a deep learning approach can effectively represent and utilize the digital content of images for more precise medical image analysis. Breast ultrasound imaging is useful for detecting and distinguishing benign masses from malignant masses. Based on the different ways in which benign and malignant tumors affect neighboring tissues, i.e., the pattern of growth and border irregularities, the penetration degree of the adjacent tissue, and tissue-level changes, we investigated the relationship between breast cancer imaging features and the roles of inter- and extra-lesional tissues and their impact on refining the performance of deep learning classification. The novelty of the proposed approach lies in considering the features extracted from the tissue inside the tumor (by performing an erosion operation) and from the lesion and surrounding tissue (by performing a dilation operation) for classification. This study uses these new features and three pre-trained deep neuronal networks to address the challenge of breast lesion classification in ultrasound images. To improve the classification accuracy and interpretability of the model, the proposed model leverages transfer learning to accelerate the training process. Three modern pre-trained CNN architectures (MobileNetV2, VGG16, and EfficientNetB7) are used for transfer learning and fine-tuning for optimization. There are concerns related to the neuronal networks producing erroneous outputs in the presence of noisy images, variations in input data, or adversarial attacks; thus, the proposed system uses the BUS-BRA database (two classes/benign and malignant) for training and testing and the unseen BUSI database (two classes/benign and malignant) for testing. Extensive experiments have recorded accuracy and AUC as performance parameters. The results indicate that the proposed system outperforms the existing breast cancer detection algorithms reported in the literature. AUC values of 1.00 are calculated for VGG16 and EfficientNet-B7 in the dilation cases. The proposed approach will facilitate this challenging and time-consuming classification task.

Examining Tongue Movement Intentions in EEG-Based BCI with Machine and Deep Learning: An Approach for Dysphagia Rehabilitation

Abstract Dysphagia, a common swallowing disorder particularly prevalent among older adults and often associated with neurological conditions, significantly affects individuals’ quality of life by negatively impacting their eating habits, physical health, and social interactions. This study investigates the potential of brain-computer interface (BCI) technologies in dysphagia rehabilitation, focusing specifically on motor imagery paradigms based on EEG signals and integration with machine learning and deep learning methods for tongue movement. Traditional machine learning classifiers, such as K-Nearest Neighbors (KNN), Support Vector Machine (SVM), Decision Tree, Naive Bayes, Random Forest, AdaBoost, Bagging, and Kernel were employed in discrimination of rest and imagination phases of EEG signals obtained from 30 healthy subjects. Scalogram images obtained using continuous wavelet transform of EEG signals corresponding to the rest and imagination phases of the experiment were used as the input images to the CNN architecture. As a result, KNN (79.4%) and SVM (63.4%) exhibited lower accuracy rates compared to ensemble methods like AdaBoost, Bagging, and Random Forest, all achieving high accuracy rates of 99.8%. These ensemble techniques proved to be highly effective in handling complex EEG datasets, particularly in distinguishing between rest and imagination phases. Furthermore, the deep learning approach, utilizing CNN and Continuous Wavelet Transform (CWT), achieved an accuracy of 83%, highlighting its potential in analyzing motor imagery data. Overall, this study demonstrates the promising role of BCI technologies and advanced machine learning techniques, especially ensemble and deep learning methods, in improving outcomes for dysphagia rehabilitation.

Deep Convolutional Tables: Deep Learning Without Convolutions.

We propose a novel formulation of deep networks that do not use dot-product neurons and rely on a hierarchy of voting tables instead, denoted as convolutional tables (CTs), to enable accelerated CPU-based inference. Convolutional layers are the most time-consuming bottleneck in contemporary deep learning techniques, severely limiting their use in the Internet of Things and CPU-based devices. The proposed CT performs a fern operation at each image location: it encodes the location environment into a binary index and uses the index to retrieve the desired local output from a table. The results of multiple tables are combined to derive the final output. The computational complexity of a CT transformation is independent of the patch (filter) size and grows gracefully with the number of channels, outperforming comparable convolutional layers. It is shown to have a better capacity:compute ratio than dot-product neurons, and that deep CT networks exhibit a universal approximation property similar to neural networks. As the transformation involves computing discrete indices, we derive a soft relaxation and gradient-based approach for training the CT hierarchy. Deep CT networks have been experimentally shown to have accuracy comparable to that of CNNs of similar architectures. In the low-compute regime, they enable an error:speed tradeoff superior to alternative efficient CNN architectures.

Lightweight CNN architecture design for rolling bearing fault diagnosis

Abstract Rolling bearing is a key component of rotating machinery, and its fault diagnosis technology is very important to ensure the safety of equipment. With the rapid development of deep learning, the convolutional neural network (CNN) is widely used in bearing fault diagnosis, showing obvious advantages in diagnostic accuracy. However, the deep CNN model generally requires a lot of computing resources and storage space, and it is not easy to apply in practical engineering. Aiming at this problem, a lightweight CNN model for rolling bearing fault diagnosis is designed. This model is combined with a network pruning algorithm and neural architecture search, which not only ensures the accuracy of diagnosis but also reduces computing resources. By constructing the search space of the complete Cell class unit, using the multi-objective reinforcement learning search strategy, and applying the deep learning pruning method to prune and search the network, the lightweight CNN model with higher accuracy is efficiently searched. The rolling bearing fault data set is utilized to validate the use of the lightweight CNN model for diagnosing rolling bearing faults. This model significantly enhances operational efficiency without compromising accuracy, achieving a fault diagnosis accuracy of up to 98.56%.

Hcnnxgboost: A Hybrid Cnn-Xgboost Approach for Effective Emotion Detection in Textual Data

in recent years, emotional analysis has become a key focus in computational studies, driven by the need to understand societal sentiments. this exploration is motivated by its many promising applications, such as community well-being evaluation, human-computer interaction, suicide prevention, and personalized recommendations. even with progress in other areas like identifying expressions from facial cues and speech, the study of text-based emotion recognition remains a fascinating field of research because machines struggle to interpret context, especially compared to human capabilities. our research addresses this by introducing a multiclass text-based emotion detection system that combines a cnn architecture with xgboost for improved classification in natural language processing. we pre-process publicly available datasets and use glove pre-trained word embeddings for better text representation. a major contribution of our work is enhancing the feature space by combining cnn probabilities with the original text data. the proposed hcnnxgboost model outperforms all other machine learning and deep learning algorithms across the emoint, isear, and crowdflower datasets, achieving f-scores of 90.1%, 87.4%, and 62.2%, respectively. experimental evaluations on benchmark datasets show better f-scores, confirming the effectiveness of our approach. comparisons with other classifiers highlight the enhanced performance and effectiveness of our hybrid cnnxgboost (hcnnxgboost) model, making it one of the best solutions for emotion classification in natural language processing tasks.

CNN Based Determination of Polycystic Ovarian Syndrome using Automatic Follicle Detection Methods

The aim of this study was to determine the best method for follicle detection using ovarian ultrasound images and to classify the ultrasound images as pcos or normal ovaries using the proposed CNN architecture. Two different methods for follicle detection have been proposed to evaluate pcos. For this purpose, the Median, the Mean, the Wiener, and the Gaussian filters were tested using standard and adaptive thresholds. Second, Gaussian filtering, Discrete Wavelet Transform, and k-means clustering algorithms were tested. The Canny operator separates follicles from the background in the segmentation phase. In this study, a CNN architecture that classifies limited ultrasound ovary images was developed, and its success in the best follicle detection method was presented. The highest follicle detection accuracy of 97.63% was achieved with adaptive thresholding using a Wiener filter. Besides, the ultrasound images of the ovaries were classified as "normal" or "polycystic ovary syndrome" using CNN architecture with classification accuracy of 65.81% for unsegmented ovarian images and 77.81% for segmented images. In addition to the proposed method, classification was performed using SqueezeNet-based transfer learning, which was successful in limited datasets, and 74.18% classification accuracy was achieved for the unsegmented images and 75.54 % for segmented images . The results show that the combination of the Wiener filter with adaptive thresholding was quite successful in follicle detection and that the CNN can better classify ovaries using preprocessed ultrasound images.