Figure - available from: Multimedia Tools and Applications
This content is subject to copyright. Terms and conditions apply.
Some examples of pap smear cells (a) Normal cell (b) Normal cell (c) Abnormal cell (d) Abnormal cell
Source publication
![Classification accuracy (%) on herlev dataset [7] based on different...](profile/Shervan-Fekri-Ershad/publication/334381012/figure/tbl1/AS:841021940305921@1577526664758/Classification-accuracy-on-herlev-dataset-7-based-on-different-ratios-of-dataset_Q320.jpg)
Pap smear test is routinely used today to diagnose cervical cancer. In the last 50 years, this simple test has saved millions of women’s lives. Although successful, pap smears are not always perfectly analyzed and it is time-consuming work. Computer-assisted screening can be widely used for cervical cancer diagnosis today. So far, a variety of imag...
Similar publications
![FIGURE Methodology [Color figure can be viewed at wileyonlinelibrary.com]](publication/344328461/figure/fig1/AS:11431281175914756@1689945742963/FIGURE-Methodology-Color-figure-can-be-viewed-at-wileyonlinelibrarycom_Q320.jpg)
![Pap smear cell image [Color figure can be viewed at wileyonlinelibrary.com]](publication/344328461/figure/fig4/AS:11431281175907564@1689945747461/Pap-smear-cell-image-Color-figure-can-be-viewed-at-wileyonlinelibrarycom_Q320.jpg)
Cervical cancer is identified as the fourth most recurrent cancer among women across the globe. The cancer is treatable, if identified at the early stage. Pap smear test is the most common and the best tool for initial screening of cancer. Pap smear cell level image analysis is an open issue. The limitation of the analysis is due to the complexity...
Citations
... Therefore, distinguishing between the stages can be crucial for diagnosis. In most literature, the classification of Pap smear images consists of a binary separation between normal and abnormal cell (two classes), using different methodologies such as Support Vector Machines (SVM) [35,49,51,119], k-Nearest Neighbours (kNN) [35,49,73,149], Fuzzy c-Means Algorithm (FCM) [49,239], k-Means clustering [175], Artificial Neural Networks (ANN) [49], and, more recently, Convolutional Neural Networks (CNN) [126,138,257]. ...
Cervical cancer stands as the fourth leading cause of cancer-related fatalities among women, with the majority of cases occurring in low to middle-income countries. Despite the recent surge in scientific advances, a completely effective treatment, particularly for advanced-stage diagnoses, remains elusive. Screening methods, such as cytology, colposcopy, and histology, have played a pivotal role in significantly reducing cervical cancer-related deaths. The imperative now is to develop innovative decision support systems for cervical cancer screening that seamlessly integrate multiple screening exams. In this context, Computer-Aided Diagnosis (CAD) systems have assumed growing significance, primarily for the purposes of quality control and evaluating cancer risk. These sophisticated CAD systems harness the power of machine learning algorithms and advanced image analysis techniques to assist healthcare practitioners in the efficient screening and diagnosis of cervical cancer. By offering automated analysis and decision support, CAD systems elevate the precision, efficiency, and consistency of cervical cancer diagnoses. Their capabilities extend to assessing image quality, detecting abnormal regions, and categorizing cancer risk, thereby playing a pivotal role in improving patient outcomes and aiding in the early detection and treatment of cervical cancer. Nevertheless, despite the notable progress achieved, there remains a substantial distance to traverse and a multitude of challenges to confront in order to comprehensively address the complexities of this field. This thesis embarks on a journey to tackle various challenges associated with quality control and risk assessment, spanning both fundamental and applied research domains.
The first part of this thesis results in several contributions related to quality control issues in medical decision support systems. It begins by focusing on the evaluation of image quality, with a particular emphasis on the detection of blurriness in digital pathology slides. Detecting blurriness in medical images is of paramount importance for ensuring accurate clinical decisions. Blurry or distorted images can lead to misinterpretations and potentially incorrect diagnoses, which can have serious consequences for patient care. Subsequently, this part also explores methods for improving image quality in cost-effective cytology microscopy. Lastly, this part closes by addressing a crucial daily task in a digital pathology lab: the automated detection and quantification of fragments, which holds significant importance in routine diagnostics. Automating the process of fragment counting has the potential to alleviate the workload of clinicians while simultaneously mitigating the risk of errors.
In the second part of this thesis, there is a focus on making fundamental contributions to ordinal classification within the realm of machine learning. These advancements involve the introduction of various ordinal loss functions tailored for classifying cervical cancer risk. The primary objective is to harness the inherent ordinal characteristics of the data to improve the overall performance of classification models.
In the final part of this thesis, we presented novel contributions related to the integration of multimodal data within decision support systems. These contributions involve the exploration and application of various techniques aimed at effectively combining data from diverse sources and modalities. This work seeks to leverage the strengths of each modality and harness the synergistic potential of integrating them.
In summary, this thesis underscores the critical importance of innovative approaches to cervical cancer screening, especially in low to middle-income countries where it remains a leading cause of cancer-related deaths among women. While progress has been made through traditional screening methods, integrating CAD systems, powered by machine learning and advanced image analysis, is enhancing precision and efficiency in cervical cancer diagnosis. This thesis focuses on quality control, ordinal classification, and multimodal data integration as key areas of improvement. These efforts represent a step toward reducing the impact of cervical cancer and ultimately improving patient outcomes.
... Statistical features help to understand pattern and texture information from image. Statistical characteristics offer lower computational cost and rotation invariance due to processing being global [18]. Inspired by [3], in our work three first-order statistic features are used to characterize the texture features of laryngeal cancer images. ...
Squamous cell carcinoma (SCC) is one of the most common as well as deadliest kinds of laryngeal cancer. The precise and early identification of laryngeal cancer plays a pivotal role in reducing mortality and maintaining laryngeal structure and vocal fold function. But small variations in the laryngeal tissues may go undetected by the human eye, which leads to misdiagnosis. In this study, we devise an early laryngeal cancer classification framework using the hybridization of deep and handcrafted features. The deep features of the DenseNet 201 using transfer learning and handcrafted features using Local Binary Pattern (LBP) and First-order statistics (STAT)s are extracted from the endoscopic narrowband images of the larynx and fused together which resulted in more representative features. From these hybridized features, the optimal features are selected by the Recursive Feature Elimination with Random Forest (RFE- RF) method. Firstly, the selected hybrid features are classified with three effective Machine Learning classifiers like Random Forest (RF), Support Vector Machine (SVM), and k-Nearest Neighbor (k-NN), and the results are compared with a stacking-based ensemble learning classification method using (SVM), (RF) and (k-NN) in order to distinguish early-stage SCC tissues, healthy tissues and precancerous tissues. The combination of hybrid features, effective feature selection, and an Ensemble classifier produced a median categorization recall of 99.5% on a standard dataset, which surpasses the state of the art (recall = 98%).
... As a result, inspection manually is complicated, and specialists are vulnerable to mistakes. So, a better solution is required for this issue [7,8]. This issue is resolved by developing automatic computer-aided diagnosis (CAD) systems. ...
Among the main factors contributing to death globally is cervical cancer, regardless of whether it can be avoided and treated if the afflicted tissues are removed early. Cervical screening programs must be made accessible to everyone and effectively, which is a difficult task that necessitates, among other things, identifying the population's most vulnerable members. Therefore, we present an effective deep-learning method for classifying the multi-class cervical cancer disease using Pap smear images in this research. The transfer learning-based optimized SE-ResNet152 model is used for effective multi-class Pap smear image classification. The reliable significant image features are accurately extracted by the proposed network model. The network's hyper-parameters are optimized using the Deer Hunting Optimization (DHO) algorithm. Five SIPaKMeD dataset categories and six CRIC dataset categories constitute the 11 classes for cervical cancer diseases. A Pap smear image dataset with 8838 images and various class distributions is used to evaluate the proposed method. The introduction of the cost-sensitive loss function throughout the classifier's learning process rectifies the dataset's imbalance. When compared to prior existing approaches on multi-class Pap smear image classification, 99.68% accuracy, 98.82% precision, 97.86% recall, and 98.64% F1-Score are achieved by the proposed method on the test set. For automated preliminary diagnosis of cervical cancer diseases, the proposed method produces better identification results in hospitals and cervical cancer clinics due to the positive classification results.
... The biggest challenge in Pap smear analysis for cervical cancer detection is the complexity and accuracy of the exam. The morphology of cervical cells varies significantly in terms of color, size, and shape [8]. It is difficult to differentiate between different cell types with the naked eye [9]. ...
Image processing has enabled faster and more accurate image classification. It has been of great benefit to the health industry. Manually examining medical images like MRI and X-rays can be very time-consuming, more prone to human error, and way more costly. One such examination is the Pap smear exam, where the cervical cells are examined in laboratory settings to distinguish healthy cervical cells from abnormal cells, thus indicating early signs of cervical cancer. In this paper, we propose a convolutional neural network- (CNN-) based cervical cell classification using the publicly available SIPaKMeD dataset having five cell categories: superficial-intermediate, parabasal, koilocytotic, metaplastic, and dyskeratotic. CNN distinguishes between healthy cervical cells, cells with precancerous abnormalities, and benign cells. Pap smear images were segmented, and a deep CNN using four convolutional layers was applied to the augmented images of cervical cells obtained from Pap smear slides. A simple yet efficient CNN is proposed that yields an accuracy of 0.9113% and can be successfully used to classify cervical cells. A simple architecture that yields a reasonably good accuracy can increase the speed of diagnosis and decrease the response time, reducing the computation cost. Future researchers can build upon this model to improve the model’s accuracy to get a faster and more accurate prediction.
... It achieved an accuracy of 0.86 for 400 training and validation and 200 tests, better than single-channel detection (either acetic acid or Lugol's iodine cervigram). There is more related work available in clinical practices [42][43][44][45][46] thanks to advances in computer vision methods. Some of these are observed as competent in performing the same or even better as the pathologists on medical data. ...
Cervical cancer is a critical imperilment to a female’s health due to its malignancy and fatality rate. The disease can be thoroughly cured by locating and treating the infected tissues in the preliminary phase. The traditional practice for screening cervical cancer is the examination of cervix tissues using the Papanicolaou (Pap) test. Manual inspection of pap smears involves false-negative outcomes due to human error even in the presence of the infected sample. Automated computer vision diagnosis revamps this obstacle and plays a substantial role in screening abnormal tissues affected due to cervical cancer. Here, in this paper, we propose a hybrid deep feature concatenated network (HDFCN) following two-step data augmentation to detect cervical cancer for binary and multiclass classification on the Pap smear images. This network carries out the classification of malignant samples for whole slide images (WSI) of the openly accessible SIPaKMeD database by utilizing the concatenation of features extracted from the fine-tuning of the deep learning (DL) models, namely, VGG-16, ResNet-152, and DenseNet-169, pretrained on the ImageNet dataset. The performance outcomes of the proposed model are compared with the individual performances of the aforementioned DL networks using transfer learning (TL). Our proposed model achieved an accuracy of 97.45% and 99.29% for 5-class and 2-class classifications, respectively. Additionally, the experiment is performed to classify liquid-based cytology (LBC) WSI data containing pap smear images.
... Fekri-Ershad [7] proposed an approach for Pap smear classification based on the combination of different statistical and computational features. Time series features in joint of textural features, such as mean, entropy and correlation, are used in the feature extraction phase. ...
... Additionally, global significant value is added as an innovative feature to improve the final accuracy. The performance of the proposed approach in [7] is evaluated based on different linear classifiers, such as the Bayesian network, naïve Bayes and KNN in the Herlev dataset. Finally, the classification accuracy of about 88.47 percent is provided for the 2-classes classification case [7]. ...
... The performance of the proposed approach in [7] is evaluated based on different linear classifiers, such as the Bayesian network, naïve Bayes and KNN in the Herlev dataset. Finally, the classification accuracy of about 88.47 percent is provided for the 2-classes classification case [7]. A set of statistical and numerical features are suggested in [7], which can be used in some other related computer vision applications, such as the characterization of bacteriophages from sewage water [8] or silver nanoparticles detection [9]. ...
Cervical cancer is one of the most common types of cancer among women, which has higher death-rate than many other cancer types. The most common way to diagnose cervical cancer is to analyze images of cervical cells, which is performed using Pap smear imaging test. Early and accurate diagnosis can save the lives of many patients and increase the chance of success of treatment methods. Until now, various methods have been proposed to diagnose cervical cancer based on the analysis of Pap smear images. Most of the existing methods can be divided into two groups of methods based on deep learning techniques or machine learning algorithms. In this study, a combination method is presented, whose overall structure is based on a machine learning strategy, where the feature extraction stage is completely separate from the classification stage. However, in the feature extraction stage, deep networks are used. In this paper, a multi-layer perceptron (MLP) neural network fed with deep features is presented. The number of hidden layer neurons is tuned based on four innovative ideas. Additionally, ResNet-34, ResNet-50 and VGG-19 deep networks have been used to feed MLP. In the presented method, the layers related to the classification phase are removed in these two CNN networks, and the outputs feed the MLP after passing through a flatten layer. In order to improve performance, both CNNs are trained on related images using the Adam optimizer. The proposed method has been evaluated on the Herlev benchmark database and has provided 99.23 percent accuracy for the two-classes case and 97.65 percent accuracy for the 7-classes case. The results have shown that the presented method has provided higher accuracy than the baseline networks and many existing methods.
... Variations to the Multi-Layer Perceptron (MLP), Auto Encoder (AE), and Extreme Learning Machine (ELM) based classifiers were also researched [15]. Fekri-Ershad provided a method with high diagnostic accuracy, as shown in Figure 5, along with a sample of Pap smear images classification model employing various classifiers, such as K-Nearest Neighbor (KNN), J48 Tree, MLP, and Bayesian Network [24]. Likewise, Zhang classified precancerous cervical lesions via pre-trained, closely connected CNN in the form of a computer-aided diagnosis method, as shown in Figure 6 [25]. ...
... A flowchart based on the Fekri-Ershad method[24]. ...
Nowadays, deep learning (DL) is a popular tool used in various applications in different fields, including the medical domain. DL techniques can cope with several challenges, which are difficult to resolve via traditional artificial intelligence (AI) techniques. Cervical cancer (CC) is one of the leading reasons for death in females and ranks second after breast cancer, with more than 700 mortalities daily. This number is estimated to be 400,000 annually by 2030. However, if the cancer is detected in the early and precancerous stages, it is completely curable. Pap smear and colposcopy are the most widely used screening methods for the detection of cervical cancer. But manual screening approach suffers from a high false rate due to human errors. To overcome this challenge, machine learning (ML) and DL-based computer-aided diagnostic (CAD) techniques are being extensively expanded to automatically segment and categorize cervical cytology and colposcopy images. These methods increase the accuracy of detecting different stages of cervical cancer. Hence, there is an increased interest in creating computer-aided solutions for CC screening, especially in less-developed countries where the majority of cervical cancer-related fatalities occur. This review overviews state-of-the-art approaches that use DL techniques to analyze cervical cytology and screening images. It reviews and discusses relevant DL techniques, their architectures, classification methods, and the segmentation of cervical cytology and colposcopy images. Finally, it reviews the DL algorithms that are currently used in CC screening and offers useful insights, research opportunities and future directions in this field.
... This cancer has two main types named squamous cell carcinoma and adenocarcinoma, where former one is most frequent one & typically occurs in thin and flat cells forming outer layer of cervix and later one happens in glandular cells of cervix. Pap smear test is most commonly adopted test for detection & treatment of cervical cancer [86,87]. Although in the past this Pap smear test protected millions of lives world-wide, however it's been found time consuming job to perfectly analyse pep test. ...
... Various studies have done on cervical cancer detection using image processing-based methods so far, most of them observed to be struggling for accuracy due low quality of test images, moreover Pap smear testbased detection methods are susceptible to rotation and gray-scale variations in test images. In line with these challenges, Fekri-Ershad, S [87]. proposed an efficient method for cervical cancer detection using dataset of 917 Pap smear samples collected from Herlev University Hospital. ...
Being a second most cause of mortality worldwide, cancer has been identified as a perilous disease for human beings, where advance stage diagnosis may not help much in safeguarding patients from mortality. Thus, efforts to provide a sustainable architecture with proven cancer prevention estimate and provision for early diagnosis of cancer is the need of hours. Advent of machine learning methods enriched cancer diagnosis area with its overwhelmed efficiency & low error-rate then humans. A significant revolution has been witnessed in the development of machine learning & deep learning assisted system for segmentation & classification of various cancers during past decade. This research paper includes a review of various types of cancer detection via different data modalities using machine learning & deep learning-based methods along with different feature extraction techniques and benchmark datasets utilized in the recent six years studies. The focus of this study is to review, analyse, classify, and address the recent development in cancer detection and diagnosis of six types of cancers i.e., breast, lung, liver, skin, brain and pancreatic cancer, using machine learning & deep learning techniques. Various state-of-the-art technique are clustered into same group and results are examined through key performance indicators like accuracy, area under the curve, precision, sensitivity, dice score on benchmark datasets and concluded with future research work challenges.
... These selected features were then adopted for classifying cervical cell images. Additionally, a classification method that combines global significance value, texture statistical features, and time-series features was proposed for cervical cell classification [16]. Inspired by the great success of deep learning technology in various computer vision and pattern recognition tasks [17][18][19], the application of deep convolution neural networks in the field of biomedicine has increased [20][21][22]. ...
Artificial intelligence (AI) technologies have resulted in remarkable achievements and conferred massive benefits to computer-aided systems in medical imaging. However, the worldwide usage of AI-based automation-assisted cervical cancer screening systems is hindered by computational cost and resource limitations. Thus, a highly economical and efficient model with enhanced classification ability is much more desirable. This paper proposes a hybrid loss function with label smoothing to improve the distinguishing power of lightweight convolutional neural networks (CNNs) for cervical cell classification. The results strengthen our confidence in hybrid loss-constrained lightweight CNNs, which can achieve satisfactory accuracy with much lower computational cost for the SIPakMeD dataset. In particular, ShufflenetV2 obtained a comparable classification result (96.18% in accuracy, 96.30% in precision, 96.23% in recall, and 99.08% in specificity) with only one-seventh of the memory usage, one-sixth of the number of parameters, and one-fiftieth of total flops compared with Densenet-121 (96.79% in accuracy). GhostNet achieved an improved classification result (96.39% accuracy, 96.42% precision, 96.39% recall, and 99.09% specificity) with one-half of the memory usage, one-quarter of the number of parameters, and one-fiftieth of total flops compared with Densenet-121 (96.79% in accuracy). The proposed lightweight CNNs are likely to lead to an easily-applicable and cost-efficient automation-assisted system for cervical cancer diagnosis and prevention.
... By combining the Haralick features, global significant values and time series features are used for feature extraction. This study is efficient for detecting cervical cancer using pap-smear images [5]. ...
Iris has specific advantages, which can record all organ conditions, body construction, and psychological disorders. Traces related to the intensity or deviation of organs caused by the disease are recorded systematically and patterned on the iris and its surroundings. The pattern that appears on the iris can be recognized by using image processing techniques. Based on the pattern in the iris image, this paper aims to provide an alternative noninvasive method for the early detection of DM and HC. In this paper, we perform detection based on iris images for two diseases, DM and HC simultaneously, by developing the invariant Haralick feature on quantized images with 256, 128, 64, 32, and 16 gray levels. The feature extraction process does early detection based on iris images. Researchers and scientists have introduced many methods, one of which is the feature extraction of the gray-level co-occurrence matrix (GLCM). Early detection based on the iris is done using the volumetric GLCM development, namely, 3D-GLCM. Based on 3D-GLCM, which is formed at a distance of d = 1 and in the direction of 0°, 45°, 90°, 135°, 180°, 225°, 270°, and 315°, it is used to calculate Haralick features and develop Haralick features which are invariant to the number of quantization gray levels. The test results show that the invariant feature with a gray level of 256 has the best identification performance. In dataset I, the accuracy value is 97.92, precision is 96.88, and recall is 95.83, while in dataset II, the accuracy value is 95.83, precision is 89.69, and recall is 91.67. The identification of DM and HC trained on invariant features showed higher accuracy than the original features.
