Automated Assessment of Tumors in Histopathology Images

Version	Summary	Created by	Modification	Content Size	Created at	Operation
1		Khursheed Aurangzeb	--	2504	2023-09-05 10:37:52	\|
2	layout & references	Sirius Huang	Meta information modification	2504	2023-09-06 02:53:47	\|

This entry is adapted from the peer-reviewed paper 10.3390/biomimetics8040370

The automated assessment of tumors in medical image analysis encounters challenges due to the resemblance of colon and lung tumors to non-mitotic nuclei and their heteromorphic characteristics. An accurate assessment of tumor nuclei presence is crucial for determining tumor aggressiveness and grading.

bioinspiration medical image analysis tumor assessment convolutional neural network (CNN) histopathology images

1. Introduction

Cancer, the second biggest cause of death, represents a major worldwide health issue. According to estimates, 609,820 Americans will die from cancer in 2023, which works out to almost 1670 fatalities every day. There may be roughly 153,020 new instances of colorectal cancer (CRC) in the US in 2023, made up of 106,970 tumors in the colon and 46,050 tumors in the rectum ^[1]^[2]. There are an enormous amount of cells in the human body, and these cells divide, develop, and multiply. When cells become damaged or reach a certain age, they typically either die naturally or are replaced by healthy counterparts ^[3]. However, if this replacement procedure does not take place, damaged cells start to multiply and become benign or malignant tumors. While malignant tumors are characterized by their abnormal, fast development and their propensity to infect surrounding tissues, benign tumors are slow-growing masses that do not affect the neighboring cells. Numerous areas of the human body can be affected by cancer cells, but colon and lung disease are among the most frequent, affecting both sexes equally ^[4].

About 238,340 new cases of lung cancer are anticipated to be diagnosed nationwide in 2023. From this, 117,550 male cases and 120,790 female cases are expected. In terms of new instances, it is predicted that there may be 127,070 new cases of lung cancer overall in 2023. There will be 59,910 new cases of cancer diagnosed in women and 67,160 new cases of cancer diagnosed in men among these ^[1]^[4]^[5]^[6]. Numerous behavioral variables, such as smoking, being overweight, abusing alcohol, or being exposed to UV radiation, ionizing radiation, or other biological agents, can have an impact on the development of cancer ^[6]. The co-occurrence of lung and colon cancers, which accounts for around 17% of cases ^[7], is noteworthy. Smoking, which is added to unhealthy eating habits, has been linked to the emergence of breast and colon cancer, according to research ^[7]. Early-stage lung and colon cancer sometimes exhibit negligible or no symptoms, delaying identification and confirmation until later stages, when treatment results may be adversely affected ^[8]. Comprehensive tests including computed tomography (CT-Scans), MRI scans, ultrasound imaging, and tissue samples are required in order to describe lung and colon cancer in its early stages ^[9].

Individuals who smoke, are overweight, or have a family history of cancer should consider having regular checkups with their doctor. It is important to note that screening procedures can be expensive, making it challenging for people with low incomes to pay them. According to the World Health Organization (WHO), developing low- and middle-income nations account for 70% of cancer-related fatalities. It is essential to help these nations in creating fully functional hospitals with free diagnostic labs for everyone in order to solve this issue. In addition, there is a large time lag and the possibility of divergent medical opinions, especially in the early stages of cancer diagnosis. Collaboration across disciplines within the healthcare industry is necessary to meet these issues. Potential remedies include the use of artificial intelligence techniques like biomedical imaging for illness early detection and emergency healthcare forecasting models ^[10]^[11]^[12].

2. Automated Assessment of Tumors in Histopathology Images

There are two primary causes that make it difficult to segment nuclei: (i) color changes in histopathological images and (ii) variances in morphological features. In the literature, a number of image processing methods have been put out to address nuclei segmentation from histopathology images. These methods include the watershed method ^[13], the multi-level thresholding examined by the watershed algorithm ^[14], hybrid segmentation using k-Means clustering and adaptive thresholding ^[15], the multi-scale and multimarker approaches ^[16], and the graph-cuts approach ^[17]. However, because they all rely on parameter-based techniques, these image processing-based algorithms are unable to manage changes in staining and morphological features.

Machine learning (ML) techniques for nucleus segmentation have drawn more and more attention in recent years. Multiple hand-crafted features, including color, texture, the Laplacian of Gaussian response, local binary patterns of the nuclei, the Hough transform, the Histogram of Oriented Gradients (HOGs), and the marker-controlled watershed approach, have been used to train machine learning models ^[18]^[19]^[20]. To analyze and segment nuclei, these models often use supervised or unsupervised learning approaches.

Traditional methodologies, in particular unsupervised learning approaches, have difficulties in nucleus segmentation since they rely heavily on feature engineering. These techniques frequently result in under-segmented nuclear areas when there is noticeable color and textural diversity ^[21]. It can be difficult to manually recognize and extract useful characteristics from images, and it is possible to not have all the necessary data for an effective categorization.

On the other hand, deep learning (DL) techniques use neural networks to automatically extract features. These systems have the capacity to learn from images and extract details that may not be immediately noticeable to human viewers. Convolutional neural networks (CNNs), U-Net, ResNet, and Masked RCNN are examples of DL models that have demonstrated notable gains in difficult biological tasks including segmentation and classification.

For instance, radioactive material was divided into several categories and nuclear sites were found using a multi-scale deep residual aggregation network ^[22]. Clustered nuclei were divided using the Feature Pyramid Network (FPN) ^[23] and U-Net architecture. Deep learning models ^[24] were used to identify nuclei outlines, and segmentation was conducted using an iterative region expanding technique.

Better performance in many biological activities has been made possible by the exponential expansion of DL architectures and computer vision techniques in recent years. Researchers can increase their performance in nucleus segmentation and other difficult biological image processing tasks by utilizing the characteristics of DL models.

In their research, Adu et al. ^[25] established a brand-new method for examining lung and colon cancer histology images termed DHSCapsNet. To improve efficiency, the network integrates DHSCaps with encoder characteristics. In particular, the convolutional layers that collect important features are where the encoder features are formed. Additionally, HSquash is used to glean important data from people with various backgrounds. A CNN Pre-Trained Diagnostic Network created in the study by Mangal et al. ^[26] was created particularly for the categorization of colon and lung cancer. The network analyzed histology slides using a basic CNN architecture. The network impressively attained high accuracy for detecting colon and lung cancer. The concept described by Ali and Ali ^[27] uses a capsule network with many sources to build a diagnostic architecture for aberrant cell cancer in the lung and colon. The convolutional layer block (CLB) and separable CLB are the two unique building components that make up the capsule network. While the separable CLB is in charge of processing histopathological images, the CLB is in charge of processing pathological images. This method uses the capsule network architecture to provide a thorough examination of various input data types. The research of Mehmood et al. ^[28] offers a useful framework for the precise finding of lung and colon cancer cells. The four key levels of the network were modified by the researchers using the AlexNet design. They achieved a noteworthy accuracy of 89% by fine-tuning the model using the changed layers and training it on a dataset. This method highlights the possibility of modifying current deep learning architectures for enhancing lung and colon cancer diagnosis. Toğaçar ^[29] established the DarkNet-19 model as the foundation for training a lung and colon cancer-specific dataset. By learning the DarkNet-19 architecture from scratch as opposed to utilizing pre-trained models, the researchers were able to teach it the specific features of the cancer dataset. They used the equilibrium method, which assisted in locating and choosing the most pertinent characteristics for the classification job, to increase the model’s efficacy. The system distinguished between ineffective traits and effective ones, giving greater weight to the latter. After that, a support vector machine (SVM) was utilized to categorize the lung and colon cancer samples using the chosen effective characteristics. The objective of this strategy was to ameliorate the model’s accomplishment and accuracy by concentrating on its most illuminating and discriminative properties. The work by Masud et al. ^[30] introduces a deep learning model for the categorization of five different types of lung and colon cancers. To amend the quality of the input images and guarantee the best possible performance of the model, the researchers used optimization techniques for them. Both 2D Fourier and 2D wavelet features, which are frequently used methods for analyzing signals and images, were recovered from the images in order to extract pertinent information. The authors produced a deep learning model that has a high accuracy by using these characteristics. This demonstrates how well their suggested method works for correctly identifying and categorizing the various forms of colon and lung cancer. In their work, Hamida et al. ^[31] used histological images from the AiCOLO dataset to categorize areas afflicted by colon cancer using four pre-trained deep learning architectures. They used SegNet and UNet for pixel segmentation operations, which allowed them to precisely identify the impacted areas. Additionally, they used a pre-trained ResNet model, which successfully classified the areas of colon cancer with an excellent accuracy. This study illustrates the effectiveness of using pre-trained architectures and deep learning architectures for the precise analysis and categorization of colon cancer in histological images. In order to detect aberrant cells and assess biomarkers for the diagnosis of colon cancer, Sarker et al. ^[32] introduced a deep learning strategy to handle segmentation issues. The suggested methodology accurately identifies and highlights aberrant cells using cutting-edge image segmentation techniques. With the help of these segmentation capabilities, it is possible to not only spot probable problem regions but also provide annotations that are very helpful to doctors throughout the diagnostic procedure. This research makes use of deep learning algorithms to create cutting-edge tools that assist medical practitioners in accurately diagnosing colon cancer and measuring pertinent biomarkers. A pretrained architecture, such as ResNet, for identifying colon tumor was introduced by Sarwinda et al. ^[33]. On two datasets that were split into 20% and 40% subsets, the model’s performance was assessed. ResNet50 fared better than ResNet18 among the studied models, obtaining a sensitivity of 87% and an accuracy of 80%. A technique was developed by Zhou et al. ^[34] to categorize whole slide imaging (WSI) images using labeled data. By incorporating characteristics from various magnification levels, the network graded colorectal cancer with an accuracy of 94.6%. A 1D CNN network was developed by Moitra and Mandal ^[35] to classify small cell lung tumors. To outperform more conventional machine learning methods, the network combined clinical characteristics with hybrid features from images.

An AI-based pre-screening tool that can distinguish between normal and malignant colon samples was developed by ^[36] with the goal of assisting pathologists throughout the diagnosing procedure. With just slide-level labels needed, the program uses weakly supervised deep learning to extract histological patterns from complete slide images. It demonstrated great accuracy in cross-validation and external validation, and it may be useful for clinical settings to help with colorectal biopsy pre-screening. The understanding of the model’s forecasting and the connection between neoplastic histology and genetic heterogeneity was improved through genetic analysis and route investigation.

The necessity to distinguish between benign and malignant colorectal adenomas, which are precursors to colorectal cancer, is addressed in this study. The suggested method, known as MIST, makes use of a multiple instance learning network based on the Swin Transformer and is capable of correctly classifying whole slide images (WSIs) of colorectal adenomas using just slide-level labels. The model demonstrated a high accuracy in external validation and was trained and validated on a dataset of 666 WSIs from patients with colorectal adenoma. The results of the interpretability study are congruent with the local pathologists’ areas of interest. MIST offers a viable and practical approach to colorectal cancer screening, supporting physicians’ judgment and perhaps lowering CRC patient mortality ^[37].

The domain shift issue that arises in machine learning models when training and testing data have distinct distributions and varying color intensities, addressed by ^[38]. The authors suggest a methodology to address this problem that combines stain normalization methods with a collection of precise, scalable, and reliable convolutional neural networks (CNNs) termed ConvNexts. The improvement brought about by combining five widely used stain normalization approaches is empirically investigated in the study. On three datasets including more than 10,000 images of colon histopathology, the suggested method’s classification performance is assessed.

This study focuses on Invasive Ductal Carcinoma Breast Cancer (IDC-BC), a common and often asymptomatic cancer type with high mortality rates. The research explores the potential of pre-trained convolutional neural networks (CNNs), including EfficientNetV2L, ResNet152V2, and DenseNet201, either individually or as an ensemble, for IDC-BC grade classification using the DataBiox dataset. Data augmentation is used to address data scarcity and imbalances. The proposed ensemble model outperforms existing state-of-the-art techniques, achieving a 94% classification accuracy and significant area under the ROC curves for grades 1, 2, and 3 (96%, 94%, and 96%, respectively) in the Databiox dataset ^[39].

The goal of this work was to create a computer-aided diagnostic (CAD) methodology that could automatically categorize lung tissue histopathology images. The CAD system was created and validated using two datasets: a private dataset and a public dataset. The public dataset comprised 15,000 images categorized into three groups, whereas the secret dataset contained 94 images divided into five categories. Machine learning was used to classify the images, along with traditional texture analysis (TA) and homology-based image processing (HI), two different methods of extracting image features. In all datasets, the CAD methodology with HI performed higher-up than the one with TA, proving the use of HI for precise lung tissue categorization ^[40].

Bhattacharya et al. ^[41] employs computer vision algorithms to identify cancer, especially lung and colon carcinomas. The paper offers a system that combines two deep learning models (ResNet-18 and EfficientNet-b4-widese) with AdBet-WOA, a hybrid meta-heuristic optimization technique. Deep features are retrieved when the deep learning networks are learned on the LC25000 dataset. The suggested technique uses a support vector machine (SVM) classifier to accurately classify lung cancer, colon cancer, and a combination of the two. The data demonstrate nearly flawless precision for colon, lung cancer, and combination categorization. The suggested method performs better in terms of feature reduction and classification performance than existing optimization techniques.

Diao et al. ^[42] presents a unique method for the accurate categorization of histopathology images termed deep multi-magnification similarity learning (DSML). The method focuses on the largely unexplored fusing of cone-shaped histopathological images at various magnifications. The difficulty of comprehending cross-magnification information transmission is solved by DSML, which also makes it simple to visualize feature representations. A similarity cross-entropy loss function is used to determine how similar bits of information are at different magnifications. Experiments were conducted using clinical nasopharyngeal carcinoma and public breast cancer datasets to demonstrate how well the DSML performed in terms of categorization when compared to other comparable methodologies. The report also discusses the reason behind the efficacy of multi-magnification approaches.

Due to the similarities in the early stages of lung and colon cancer tumors, the researchers intended to produce encouraging findings in the detection of both diseases. The key goal for researchers in this area continues to be promising accuracy.

References

Siegel, R.L.; Wagle, N.S.; Cercek, A.; Smith, R.A.; Jemal, A. Colorectal cancer statistics, 2023. CA A Cancer J. Clin. 2023, 73, 233–254.
Equils, O.; Bakaj, A.; Wilson-Mifsud, B.; Chatterjee, A. Restoring Trust: The Need for Precision Medicine in Infectious Diseases, Public Health and Vaccines. Hum. Vaccines Immunother. 2023, 19, 2234787.
Gabriel, J.A. The Biology of Cancer; John Wiley & Sons: Hoboken, NJ, USA, 2007.
Kumar, N.; Sharma, M.; Singh, V.P.; Madan, C.; Mehandia, S. An empirical study of handcrafted and dense feature extraction techniques for lung and colon cancer classification from histopathological images. Biomed. Signal Process. Control 2022, 75, 103596.
Siegel, R.L.; Miller, K.D.; Wagle, N.S.; Jemal, A. Cancer statistics, 2023. CA Cancer J. Clin. 2023, 73, 17–48.
Pettit, R.W.; Byun, J.; Han, Y.; Ostrom, Q.T.; Edelson, J.; Walsh, K.M.; Bondy, M.L.; Hung, R.J.; McKay, J.D.; Amos, C.I. The shared genetic architecture between epidemiological and behavioral traits with lung cancer. Sci. Rep. 2021, 11, 17559.
Kurishima, K.; Miyazaki, K.; Watanabe, H.; Shiozawa, T.; Ishikawa, H.; Satoh, H.; Hizawa, N. Lung cancer patients with synchronous colon cancer. Mol. Clin. Oncol. 2018, 8, 137–140.
Del Re, M.; Rofi, E.; Restante, G.; Crucitta, S.; Arrigoni, E.; Fogli, S.; Di Maio, M.; Petrini, I.; Danesi, R. Implications of KRAS mutations in acquired resistance to treatment in NSCLC. Oncotarget 2018, 9, 6630.
Vega, P.; Valentín, F.; Cubiella, J. Colorectal cancer diagnosis: Pitfalls and opportunities. World J. Gastrointest. Oncol. 2015, 7, 422.
Chakraborty, S.; Mali, K. An overview of biomedical image analysis from the deep learning perspective. Res. Anthol. Improv. Med. Imaging Tech. Anal. Interv. 2023, 43–59.
Iqbal, S.; Qureshi, A.N.; Li, J.; Mahmood, T. On the Analyses of Medical Images Using Traditional Machine Learning Techniques and Convolutional Neural Networks. Arch. Comput. Methods Eng. 2023, 3173–3233.
Van der Velden, B.H.; Kuijf, H.J.; Gilhuijs, K.G.; Viergever, M.A. Explainable artificial intelligence (XAI) in deep learning-based medical image analysis. Med. Image Anal. 2022, 102470.
Yang, X.; Li, H.; Zhou, X. Nuclei segmentation using marker-controlled watershed, tracking using mean-shift, and Kalman filter in time-lapse microscopy. IEEE Trans. Circuits Syst. I Regul. Pap. 2006, 53, 2405–2414.
Abdolhoseini, M.; Kluge, M.G.; Walker, F.R.; Johnson, S.J. Segmentation of heavily clustered nuclei from histopathological images. Sci. Rep. 2019, 9, 4551.
Filipczuk, P.; Kowal, M.; Obuchowicz, A. Automatic breast cancer diagnosis based on k-means clustering and adaptive thresholding hybrid segmentation. In Proceedings of the Image Processing and Communications Challenges 3; Springer: Berlin/Heidelberg, Germany, 2011; pp. 295–302.
Cui, Y.; Zhang, G.; Liu, Z.; Xiong, Z.; Hu, J. A deep learning algorithm for one-step contour aware nuclei segmentation of histopathology images. Med. Biol. Eng. Comput. 2019, 57, 2027–2043.
Veta, M.; Van Diest, P.J.; Kornegoor, R.; Huisman, A.; Viergever, M.A.; Pluim, J.P. Automatic nuclei segmentation in H&E stained breast cancer histopathology images. PLoS ONE 2013, 8, e70221.
Rajyalakshmi, U.; Rao, S.K.; Prasad, K.S. Supervised classification of breast cancer malignancy using integrated modified marker controlled watershed approach. In Proceedings of the 2017 IEEE 7th International Advance Computing Conference (IACC), Hyderabad, India, 5–7 January 2017; IEEE: Manhattan, NY, USA, 2017; pp. 584–589.
Phoulady, H.A.; Goldgof, D.B.; Hall, L.O.; Mouton, P.R. Nucleus segmentation in histology images with hierarchical multilevel thresholding. In Proceedings of the Medical Imaging 2016: Digital Pathology; SPIE: Bellingham, WA, USA, 2016; Volume 9791, pp. 280–285.
Hussein, I.J.; Burhanuddin, M.A.; Mohammed, M.A.; Benameur, N.; Maashi, M.S.; Maashi, M.S. Fully-automatic identification of gynaecological abnormality using a new adaptive frequency filter and histogram of oriented gradients (HOG). Expert Syst. 2022, 39, e12789.
Kong, H.; Gurcan, M.; Belkacem-Boussaid, K. Partitioning histopathological images: An integrated framework for supervised color-texture segmentation and cell splitting. IEEE Trans. Med. Imaging 2011, 30, 1661–1677.
Vu, Q.D.; Graham, S.; Kurc, T.; To, M.N.N.; Shaban, M.; Qaiser, T.; Koohbanani, N.A.; Khurram, S.A.; Kalpathy-Cramer, J.; Zhao, T.; et al. Methods for segmentation and classification of digital microscopy tissue images. Front. Bioeng. Biotechnol. 2019, 53.
Cheng, Z.; Qu, A. A fast and accurate algorithm for nuclei instance segmentation in microscopy images. IEEE Access 2020, 8, 158679–158689.
Xing, F.; Xie, Y.; Yang, L. An automatic learning-based framework for robust nucleus segmentation. IEEE Trans. Med. Imaging 2015, 35, 550–566.
Adu, K.; Yu, Y.; Cai, J.; Owusu-Agyemang, K.; Twumasi, B.A.; Wang, X. DHS-CapsNet: Dual horizontal squash capsule networks for lung and colon cancer classification from whole slide histopathological images. Int. J. Imaging Syst. Technol. 2021, 31, 2075–2092.
Mangal, S.; Chaurasia, A.; Khajanchi, A. Convolution neural networks for diagnosing colon and lung cancer histopathological images. arXiv 2020, arXiv:2009.03878.
Ali, M.; Ali, R. Multi-input dual-stream capsule network for improved lung and colon cancer classification. Diagnostics 2021, 11, 1485.
Mehmood, S.; Ghazal, T.M.; Khan, M.A.; Zubair, M.; Naseem, M.T.; Faiz, T.; Ahmad, M. Malignancy detection in lung and colon histopathology images using transfer learning with class selective image processing. IEEE Access 2022, 10, 25657–25668.
Toğaçar, M. Disease type detection in lung and colon cancer images using the complement approach of inefficient sets. Comput. Biol. Med. 2021, 137, 104827.
Masud, M.; Sikder, N.; Nahid, A.A.; Bairagi, A.K.; AlZain, M.A. A machine learning approach to diagnosing lung and colon cancer using a deep learning-based classification framework. Sensors 2021, 21, 748.
Hamida, A.B.; Devanne, M.; Weber, J.; Truntzer, C.; Derangère, V.; Ghiringhelli, F.; Forestier, G.; Wemmert, C. Deep learning for colon cancer histopathological images analysis. Comput. Biol. Med. 2021, 136, 104730.
Sarker, M.M.K.; Makhlouf, Y.; Craig, S.G.; Humphries, M.P.; Loughrey, M.; James, J.A.; Salto-Tellez, M.; O’Reilly, P.; Maxwell, P. A means of assessing deep learning-based detection of ICOS protein expression in colon cancer. Cancers 2021, 13, 3825.
Sarwinda, D.; Paradisa, R.H.; Bustamam, A.; Anggia, P. Deep learning in image classification using residual network (ResNet) variants for detection of colorectal cancer. Procedia Comput. Sci. 2021, 179, 423–431.
Zhou, C.; Jin, Y.; Chen, Y.; Huang, S.; Huang, R.; Wang, Y.; Zhao, Y.; Chen, Y.; Guo, L.; Liao, J. Histopathology classification and localization of colorectal cancer using global labels by weakly supervised deep learning. Comput. Med. Imaging Graph. 2021, 88, 101861.
Moitra, D.; Mandal, R.K. Classification of non-small cell lung cancer using one-dimensional convolutional neural network. Expert Syst. Appl. 2020, 159, 113564.
Bilal, M.; Tsang, Y.W.; Ali, M.; Graham, S.; Hero, E.; Wahab, N.; Dodd, K.; Sahota, H.; Lu, W.; Jahanifar, M.; et al. AI based pre-screening of large bowel cancer via weakly supervised learning of colorectal biopsy histology images. medRxiv 2022.
Cai, H.; Feng, X.; Yin, R.; Zhao, Y.; Guo, L.; Fan, X.; Liao, J. MIST: Multiple instance learning network based on Swin Transformer for whole slide image classification of colorectal adenomas. J. Pathol. 2023, 259, 125–135.
Yengec-Tasdemir, S.B.; Aydin, Z.; Akay, E.; Dogan, S.; Yilmaz, B. Improved classification of colorectal polyps on histopathological images with ensemble learning and stain normalization. Comput. Methods Programs Biomed. 2023, 232, 107441.
Kumaraswamy, E.; Kumar, S.; Sharma, M. An Invasive Ductal Carcinomas Breast Cancer Grade Classification Using an Ensemble of Convolutional Neural Networks. Diagnostics 2023, 13, 1977.
Nishio, M.; Nishio, M.; Jimbo, N.; Nakane, K. Homology-based image processing for automatic classification of histopathological images of lung tissue. Cancers 2021, 13, 1192.
Bhattacharya, A.; Saha, B.; Chattopadhyay, S.; Sarkar, R. Deep feature selection using adaptive β-Hill Climbing aided whale optimization algorithm for lung and colon cancer detection. Biomed. Signal Process. Control 2023, 83, 104692.
Diao, S.; Luo, W.; Hou, J.; Lambo, R.; AL-kuhali, H.A.; Zhao, H.; Tian, Y.; Xie, Y.; Zaki, N.; Qin, W. Deep Multi-Magnification Similarity Learning for Histopathological Image Classification. IEEE J. Biomed. Health Inform. 2023, 27, 1535–1545.

© Text is available under the terms and conditions of the Creative Commons Attribution (CC BY) license; additional terms may apply. By using this site, you agree to the Terms and Conditions and Privacy Policy.

Upload a video for this entry

Information

Subjects: Engineering, Biomedical

Contributors MDPI registered users' name will be linked to their SciProfiles pages. To register with us, please refer to https://encyclopedia.pub/register :

Saeed Iqbal

Adnan N. Qureshi

Musaed Alhussein

Khursheed Aurangzeb

Seifedine Kadry

View Times: 149

Update Date: 06 Sep 2023

Table of Contents

Video Upload Options

Confirm

1. Introduction

2. Automated Assessment of Tumors in Histopathology Images

References