Next Article in Journal
An Optimization-Based Technology Applied for Face Skin Symptom Detection
Next Article in Special Issue
A Curriculum Batching Strategy for Automatic ICD Coding with Deep Multi-Label Classification Models
Previous Article in Journal
Topics Mentioned by Stroke Patients during Supportive Phone Calls—Implications for Individualized Aftercare Programs
Previous Article in Special Issue
Virtual Reality Combined with Artificial Intelligence (VR-AI) Reduces Hot Flashes and Improves Psychological Well-Being in Women with Breast and Ovarian Cancer: A Pilot Study
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Review

Breast Cancer Dataset, Classification and Detection Using Deep Learning

1
Department of Computer Science and Information Technology, Women University AJK, Bagh 12500, Pakistan
2
Higher Education Department Govt, AJK, Mirpur 10250, Pakistan
3
Institute for Intelligent Systems Research and Innovation (IISRI), Deakin University, Geelong, VIC 3216, Australia
4
Examination Branch, Dibrugarh University, Dibrugarh 786004, India
5
Centre for Computer Science and Applications, Dibrugarh University, Dibrugarh 786004, India
*
Author to whom correspondence should be addressed.
Healthcare 2022, 10(12), 2395; https://doi.org/10.3390/healthcare10122395
Submission received: 13 November 2022 / Revised: 24 November 2022 / Accepted: 25 November 2022 / Published: 29 November 2022
(This article belongs to the Special Issue Artificial Intelligence Applications in Medicine)

Abstract

:
Incorporating scientific research into clinical practice via clinical informatics, which includes genomics, proteomics, bioinformatics, and biostatistics, improves patients’ treatment. Computational pathology is a growing subspecialty with the potential to integrate whole slide images, multi-omics data, and health informatics. Pathology and laboratory medicine are critical to diagnosing cancer. This work will review existing computational and digital pathology methods for breast cancer diagnosis with a special focus on deep learning. The paper starts by reviewing public datasets related to breast cancer diagnosis. Additionally, existing deep learning methods for breast cancer diagnosis are reviewed. The publicly available code repositories are introduced as well. The paper is closed by highlighting challenges and future works for deep learning-based diagnosis.

1. Introduction

Computational pathology (CP) has the potential to improve clinical workflow efficiency and diagnostic quality thanks to information integration and advanced digital communication networks [1]. CP is accompanied by several challenges, such as efficient data fusion, limited processing capabilities, and compliance with ethical practices [2].
Over 2 million women were examined for breast cancer in 2018, among whom approximately 0.6 million died worldwide. Most intrusive breast cancer diseases are chemical receptor-positive [3]. Chemical therapies targeting the trauma center flagging pathway often help patients with chemical receptor-positive tumors [4]. After delicately segmenting a patient’s example onto magnifying instrument slides for staining, a pathologist draws a visual conclusion based on hematoxylin and eosin (H&E) staining, and subatomic marker-explicit stains are used for confirmation and subtyping. Trauma centers are identified using atomic ImmunoHistoChemistry (IHC). However, IHC staining is both time-consuming and expensive [5,6]. Moreover, test quality can vary significantly due to differences in tissue, the skill level of the expert taking the tissue sample, and specialist ability levels [7,8]. Finally, pathologists’ decisions are prone to error [9]. These factors contribute to misdiagnosis. About 20% of current IHC-based trauma center and PR test results are incorrect [9,10], putting patients at risk of receiving subpar treatment. Recent research has shown that emergency room tests can be resolved using morphological stains. However, these studies rely on single-focus tissue microarray datasets (TMAs) [11].
This review examines the application of deep learning (DL) in understanding breast cancer images. We start by pointing out the significance of imaging in nervous system science and its clinical advantages. The review is continued by discussing DL advancements in breast cancer diagnosis. The capabilities of such frameworks, their challenges and possible solutions, and related datasets are investigated. The primary contributions of this paper are:
  • Recent articles (from 2018 to 2022) regarding the application of DL in breast cancer diagnosis are reviewed.
  • Open datasets related to breast cancer diagnosis are introduced, and their web links are given.
  • Publicly available source codes related to existing papers are listed with their web links.
  • Current challenges and possible future direction are given regarding the application of DL in breast cancer diagnosis.
The rest of the paper is organized as follows: a brief introduction of digital pathology, breast cancer, and the potential of artificial intelligence (AI) to automate the diagnosis process is given in Section 2. A set of datasets, existing literature, and challenges in breast cancer diagnosis using DL are given in Section 3. Discussion is presented in Section 4 followed by the conclusion in Section 5.

2. Digital Pathology and Deep Learning

Pathology is represented by a variety of terms, including “computerized pathology”, “AI”, and “computational pathology”. With the advancement of fluorescent slide scanners, entire glass slides can be virtualized and digitized [12]. The data from the slides can be saved in cloud storage, allowing pathologists to analyze the data with ease and the benefit of assistance from AI-based diagnosis tools [13,14]. To this end, researchers have already developed a variety of AI methods for medical diagnosis [15].
Breast cancer is the most widely recognized malignant growth in women, accounting for nearly half of cancer cases diagnosed in women [16,17]. HR-positive and lymph hub-negative infections also account for nearly half of all cases [18,19,20]. Following widespread clinical approval, multigene tests such as the Oncotype DX 21-gene test, PAM50, and Mamma Print are used to examine patients and guide ACTx in HR-positive and lymph node-negative breast cancer [21,22]. The clinical benefit of the 21-gene test is debatable in patients with HR-positive, lymph hub-negative, and early-stage breast cancer [23,24]. Furthermore, the fragility of RNA extracted from formalin-fixed paraffin-inserted (FFPE) tissue may jeopardize its precision and prevent proper interpretation of recurrence score (RS) results [24]. As a result, a simpler and more effective strategy for determining the risk of repetition based on super-durable tissue is required. Considering that the RS from the 21-gene test is not entirely determined by the expansion qualities bunch score (MKI67, STK15, BIRC5, CCNB1, and MYBL2) and that the mitotic count is linked to the RS7, a careful obsessive evaluation of mitosis and other cell-cell collaborations includes the RS7. Recently, the Lunit Extension has been demonstrated to predict mitosis accurately in every cell in bosom malignant growth [25], as well as recognized cancer cells and other cells in a microenvironment.
Breast carcinoma is the most common malignant growth in women worldwide, and it encompasses a wide range of diseases with varying histological, prognostic, and clinical outcomes [26,27]. Metastatic infections, such as liver and cellular breakdowns in the lungs, affect a majority of patients with malignant bosom growth [28]. A comprehensive genomic analysis of bosom disease patients identified key drivers of hereditary transformations responsible for therapeutic ramifications and outcome prediction [29].

3. Automated Breast Cancer Diagnosis

Inspired by the working mechanism of the human brain, artificial neural networks (ANNs) exploit multi-layer complex neuron structures to achieve high representation power [30]. Promising results of ANNs encouraged researchers to develop convolutional neural networks (CNNs) to handle high dimensional data such as images [31,32]. Thanks to automatic feature extraction using convolutional and max pooling layers, CNNs are able to learn challenging tasks [33,34].

3.1. Search Strategy

In this section, the search strategy for gathering existing papers related to breast cancer diagnosis is explained. To conduct our search, an AND/OR combination of multiple keywords was used: (breast cancer diagnosis OR malignant growth OR tumor) AND (deep learning OR machine learning). A total of 514 papers were gathered. Inclusion/exclusion of the gathered papers was performed based on authors’ voting. Papers with at least three votes were considered for inclusion in this survey. The number of selected papers categorized by their publishers were 10, 15, 28, and 19, corresponding to Elsevier, Springer, IEEE, and other publishers. These statistics correspond to the first blue row of Figure 1. We repeated our search among the references of the selected papers. Among the selected papers, 9, 9, 16, and 13 belonged to Elsevier, Springer, IEEE, and other publishers, which have been added to the statistics in the first blue row of Figure 1 to yield the values in the second blue row of the same figure.

3.2. Breast Cancer Datasets

There are multiple publicly available datasets for breast cancer diagnosis. To aid cancer detection, some datasets contain viewpoint, malignant growth box, impediment, and other characteristics [35,36]. We undertook extensive research to identify notable breast cancer datasets, which are summarized in Table 1.

3.3. DL Application in Breast Cancer Diagnosis

AI has recently demonstrated promising results in terms of precision and accuracy for the automated diagnosis of diseases such as breast cancer [37,38]. Among AI methods, DL stands out for processing high-dimensional data such as medical images [39,40]. An extensive search has been conducted to gather articles related to breast cancer diagnosis. The majority of these articles were gathered from the Nature database, bosom malignant growth. Significant effort has been put into covering recently published articles, especially the ones with publicly available source codes. The remainder of this section is devoted to the overview of the investigated papers.
Wang et al. (the winning team in the CAMELYON16 challenge) created various models using 256 × 256-pixel patches from positive and negative areas of whole slide images of bosom sentinel lymph hubs [41]. Pathologists reported that having a profound learning framework as an assistant decreases the human error rate by 85% [42]. Other studies reported that estrogen receptor status (trauma centers) is a fundamental atomic marker used to diagnose and select treatment options [43,44,45].
During clinical administration, pathologists examine biopsied tissue for the designated receptor with immunohistochemistry (IHC) to detect cell surface antigens [46,47]. Due to the importance of tissue analysis, attempts have been made to automate it using DL. For example, two deep neural networks (DNNs) were attached end-to-end for local and global feature extraction from microscopy images [48]. The first network acts as an autoencoder for efficient dimensionality reduction, and the second network takes the job of classification. The steps of this approach are shown in Figure 2.
Determining the factor with a high impact on cancer patients’ survival is vital for slowing down the cancer progression and increasing the life expectancy of the patients. To this end, Cho et al. [49] investigated the correlation between HE-stained tissue slides and adjuvant chemotherapy benefits for cancer patients. A CNN was trained on 1343 patients to identify histological parameters based on HE-stained whole slide images. The resulting method was called Lunit SCOPE, the steps of which are shown in Figure 3.
Another examination approach is a mammogram, which is an X-ray picture of the breast. This approach is even useful for regular examinations of women with no signs of breast cancer. This is particularly important for early diagnosis and taking preventive actions to reduce the potential threat of breast cancer. To this end, Shen et al. [50] utilized DL to diagnose breast cancer based on mammograms. To reduce the cost of preparing a sufficient amount of training data, two sets of training data with different annotations were considered. A limited set of samples with lesion-level annotation was used in the first phase of training. In the second phase, only samples with image-level annotation were used. The cost of image-level annotation is much less than lesion-level annotation, which is appealing. The high-level steps of the aforementioned method are depicted in Figure 4.
Given that mammography is a reliable approach for breast cancer diagnosis, Petrini et al. [51] have utilized two mammography images (bilateral craniocaudal and mediolateral oblique views) to enhance the diagnosis performance. Their method is based on EfficientNet and has two major components, which are the patch classifier and the whole-image classifier. The patch classifier inspects small sub-images, and the whole classifier uses the patch classifier to scan the whole mammogram. The high-level schematic of this method is depicted in Figure 5. As can be seen, the two mammograms are processed in parallel.
In addition to mammography, the detection of small tumors helps with the early diagnosis of breast cancer. To this end, the STAN method [52] has been proposed, which utilizes multiple convolution operations with different kernel sizes to capture breast tumors of various sizes (including small ones). The architecture of STAN is illustrated in Figure 6, in which convolutions with different sizes have been marked with different colors.
Researchers have observed that nuclear protein Ki-67 and tumor-infiltrating lymphocytes (TILs) are important factors for breast cancer diagnosis. Due to the lack of publicly available datasets for Ki-67 stained cell detection, Negahbani et al. [53] gathered such a dataset for public use. Additionally, a DNN named PathoNet was proposed which is a light backbone for cancer diagnosis. To facilitate experimenting with different DL models, a generic pipeline for cancerous cell detection was proposed that is compatible with a variety of DL models.
Although achieving state-of-the-art diagnosis performance is important, the ability to interpret the decision-making of DL models should not be overlooked. Being able to reason about the decision-making process is useful to gain better insight into the strengths and weaknesses of DL models. To this end, Patil et al. [54] took a multi-instance learning approach in a weakly supervised manner for the classification of breast cancer histology images. As shown in Figure 7, each input image is partitioned into multiple smaller patches. Feeding these patches to the feature extractor module, attention scores are computed, which are used to compute bag-level features. The classification is performed based on the bag-level features.
Graph neural networks have also been used to achieve interpretable results from DL models [55]. To this end, a set of quantitative metrics has been proposed to provide pathologists with understandable output. Four graph explainability methods have been used, which are based on graph pruning, gradient-based saliency, and layer-wise relevance propagation. The joint process of classification and explainability data preparation is shown in Figure 8.
Despite the considerable potential of DL in the medical domain, medical experts do not fully trust DL. To gain the experts’ trust, the output of DL models must be human-readable (i.e., interpretable). Chauhan et al. [56] have used DL for the prediction of genomic biomarkers such as TP53 mutation, PIK3CA mutation, ER status, etc. The motivation is that classification of genomic biomarkers based on gene expression data is costly and may not be available or sometimes even not feasible. On the other hand, genomic biomarker prediction using DL is an affordable and accessible alternative that is helpful for planning effective treatments. The overall schema of this method is illustrated in Figure 9.
It is also crucial to investigate the effect of using different CNN architectures and hardware processing platforms for breast cancer diagnosis. Such investigation has been undertaken for microscopic images of sentinel lymph tissue [15,57,58]. In particular, Bonnet [59] has conducted careful experiments to evaluate diagnostic performance using different CNN architectures and processing hardware platforms. Moving forward, Bonnet has investigated the effect of using transfer learning, hyperparameter tuning, and data augmentation on the diagnostic performance of DL models.
Considering that cancer is a chronic disease, monitoring the patient’s status during treatment is critical. Wang et al. [60] have proposed a TopoTxR pipeline for predicting the response to breast cancer treatment. To this end, 1D and 2D topological structures were extracted from breast MRI. Based on these 1/2D structures, new images were created in which voxels corresponding to the extracted structures were set to values in the breast MRI, and the rest were set to zero. The created images were fed to a simple CNN for pathological complete response prediction. The high-level steps of the TopoTxR method are depicted in Figure 10. To facilitate the comparison of existing methods, some of them are summarized in Table 2. Moreover, the set of articles that have accompanying public source codes are gathered in Table 3.

3.4. DL Challenges

Despite achieving remarkable results, DL also has its drawbacks [69,70]. To reach acceptable performance, DL methods need a tremendous amount of training data which is hard to come by in the medical domain [71,72]. Preparing training data requires manual labeling which must be carried out by pathologists. This process is costly and requires a considerable amount of time. Moreover, accessing a sufficient number of pathologists may not always be possible [7,73]. Another critical limitation of DL methods is their deterministic nature [74,75]. A well-trained DL model performs well on samples similar to the ones observed during training but fails miserably upon encountering out-of-distribution (OOD) samples. Providing the wrong diagnosis is not acceptable in safety-critical applications such as medical diagnosis. Therefore, it is crucial to develop uncertainty-aware DL models which can estimate how confident they are about their predictions [76,77]. Uncertainty-aware DL has already been investigated in multiple studies [78,79,80], but the field is still an active area of research.
Despite the drawbacks mentioned above, DL has excellent potential for handling challenging tasks [81,82]. For example, in the Camelyon Amazing Test 2016, DL-based approaches were evaluated for disease diagnosis in hematoxylin and eosin (H&E)- stained whole slide imaging (WSI) [83]. Promising outcomes were achieved with a cancer location pace of 92.4%, where a pathologist could accomplish 73.2% responsiveness [10]. Through worldwide joint efforts, computational pathology aims to work on symptomatic exactness, better patient treatment, and treatment cost reduction. Developing better breast cancer diagnosis systems using DL is a crucial part of this objective.
In the last 10 to 15 years, many articles in light of DL have been published [84,85]. Despite significant progress in the field of breast cancer diagnosis, there is still room for improvement. Explainable AI [86] is a research topic that strives to shed light on the complex working mechanism of DL models. Considering that medical diagnosis is safety-critical, careful analysis of DL-based diagnosis systems is an important future aspect [87,88]. Such analysis demands a sufficient amount of annotated data, which is still limited. Therefore, preparing more labeled data is also important for future work [89,90].

4. Discussion

Early diagnosis and treatment of breast cancer heavily contribute to increasing life expectancy [91]. In developed countries, age-normalized breast cancer mortality fell by 40% between the 1980s and 2020 [92]. Breast cancer mortality has been reduced by 2 to 4 percent per year in nations that have taken effective treatment strategies [93,94]. Assuming that the breast cancer mortality rate is decreased by 2.5 percent per year, it is anticipated that 2.5 million more patients will stay alive from 2020 to 2040 [95,96].
As a worldwide issue, breast cancer took more than 600,000 lives in 2018. Screening mammography is very effective at reducing bosom disease mortality by 20–40%, and it is recommended by health organizations worldwide for early detection of malignant growth locations [97,98]. Information obtained from our provincial disease reconnaissance framework revealed the status of breast cancer growth endurance and mortality rate in northwestern Iran [99]. Generally, Iran has better breast cancer explicit endurance and a lower mortality rate compared to the country’s general breast cancer growth explicit endurance. However, breast cancer endurance is still lower than in developed nations [100,101].
Breast cancer was reported as the third most common malignant growth in the studies carried out in Iran [102]. The US and Western Europe have reported the highest breast cancer rate, while East Asia has reported the lowest [8,103]. Iran is one of the countries with a rising cancer rate and mortality rate. On the other hand, in agricultural nations, these rates are lower [30]. The aging population, variation to the Western way of life, no full-term pregnancy, late age at first pregnancy, lack of bosom healthcare services, hormonal pregnancy control, and being overweight have contributed to these patterns [104,105,106].
Over the last decade, early diagnosis and efficient treatment have increased the age-normalized life expectancy of patients in developed countries. However, patients in some low-income nations in Africa and Asia still suffer lower life expectancy [87,107]. The Harmony study indicates that the 5-year net endurance rate for bosom malignant growth has consistently increased to almost 80% in numerous nations [108,109]. Breast cancer disease explicit endurance paces of 81–86% have been reported for Britain, Belgium, Canada, the US, and Italy, while comparable figures are much lower in Malaysia (68%), India (60%), Mongolia (57%), and South Africa (53%) [108]. These significant differences might be due to the absence of oncology administrations and medicines, similar to the absence of early diagnosis and screening offices [110]. As indicated by a new Iranian review, the one-, three-, and five-year bosom malignant growth explicit endurance rates were 95.6%, 80.8%, and 69.5%, respectively [111]. Nevertheless, when compared with developed nations, endurance to bosom malignant growth is much lower in Iran, which is partly due to improper treatment modalities [112,113,114,115,116]. It has also been reported that growth size, lymph hub contribution, growth grade, financial status, and genetic inheritance are among the primary factors related to bosom disease explicit endurance [117,118,119]. Disease libraries give basic data to local area-wide anticipation endeavors.
Identifying the major risk factors contributing to breast cancer is crucial for diagnosing breast cancer and controlling its progress. Several studies have been devoted to risk factor identification. For example, Zhang et al. [120] have identified 17 immune genes that were considered prognostic biomarkers for breast cancer. Using these genes and AI, a survival prediction system for breast cancer patients was proposed. Predicting cancer risk as accurately as possible is highly desirable. To this end, Behravan et al. [121] utilized XGBoost [122] to develop an approach for determining the combination of interacting genetic variants and demographic risk factors leading to maximum accuracy in breast cancer risk prediction. Liu et al. [123] have also utilized the XGBoost method to identify risk factors contributing to breast cancer in menopausal women. Given the importance of risk factors contributing to breast cancer, Sharma et al. [124] have devoted a full survey on risk factors and assessment models for breast cancer and pointed out that patients at high risk must receive more frequent examinations.
Automated diagnostic tools not only increase the efficiency of the examination process but also reduce the workload of radiologists. To this end, a commercial AI diagnostic tool was used for breast cancer detection. Based on the AI tool output, the mammograms of patients were triaged in order to reduce the number of patients that undergo radiology [125]. It is possible to prepare models utilizing H&E stains as information and IHC explanations as info marks. This is suitable for multi-instance learning (MIL) [126,127]. Recently, MIL has been utilized to predict ML-driven histopathology [128].
ML approaches for medical diagnosis need to be interpretable, i.e., they must be able to specify the regions of interest in the image. Interpretability is fundamental to gaining medical experts’ trust in using automated diagnosis systems based on ML [129,130]. The field of interpretable AI is itself a major research area that is crucial to gaining a better understanding of black box ML models such as DNNs. Based on the nature of the ML model, available data, and interpretation strategy, interpretable AI methods have been categorized [131]. In future work, it is imperative to determine interpretable AI methods best suited for the medical diagnosis domain. The progress toward incorporating interpretability in AI models for medical applications has already started. For example, Karatza et al. [132] have proposed an ensemble of neural networks for breast cancer diagnosis and evaluated its interpretability using individual conditional expectation (ICE) [133] plots. Some other metrics to evaluate the interpretability of AI models are the global surrogate (GS) [134,135] method and the Shapley values (a method borrowed from game theory) [136,137].

5. Conclusions

In this review, we looked at the most recent research on breast cancer diagnosis using DL in image modalities. Various well-known DL methods such as CNN, RNN, GoogLeNet, ResNet, and ANN have been used in the literature for breast cancer diagnosis. In addition to reviewing existing DL-based diagnosis methods, the publicly available datasets and source code repositories were introduced as well. Inspection of the existing approaches reveals the significant progress toward automated diagnosis using DL. However, the reliability of these automated systems is yet to be improved before full deployment in real-world applications.
Over the years, the field of DL has made significant progress to the point that model representation power is rarely the limiting factor. However, without having a sufficient number of training samples, these powerful models will be of no use. Dealing with limited training data is an ongoing research field and can be tackled using different approaches. The most obvious way of addressing data shortage is gathering high-quality datasets that are publicly available. However, data collation is not always possible. Image composition is an alternative promising approach that can be used to create new samples by merging two images [138]. In this technique, several background and foreground images are combined in different ways to generate new training samples. Transfer learning is another strategy to deal with data scarcity. It is highly desirable to make transfer learning domain-aware [139]. Oftentimes, existing pre-trained models have been trained on general-purpose datasets such as ImageNet, which bears little resemblance to medical images. To address this issue, it is better to pre-train models on datasets that share common features with our target dataset.
While DL models are general-purpose learners, relying solely on image data is a short-sighted strategy. Investigating the possibility of performance improvement via fusing multiple sources of data [140] is worth investigating. A different but related approach is utilizing an ensemble of DL models for more robust decision making. The challenge is reducing the complexity of ensemble DL models in order to achieve better performance with manageable computational complexity. Knowledge distillation approaches [141,142] may be useful in making ensemble methods computationally efficient without losing much performance.

Author Contributions

Conceptualization, M.S.I. and W.A.; methodology, R.A., S.H. and R.R.; software, M.S.I. and W.A.; validation, R.A., S.H. and R.R.; formal analysis, M.S.I. and W.A.; investigation, R.A., S.H. and R.R.; resources, M.S.I. and W.A.; data curation, R.A., S.H. and R.R.; writing—original draft preparation, M.S.I. and W.A.; writing—review and editing, R.A., S.H. and R.R.; visualization, M.S.I. and W.A.; supervision, R.A., S.H. and R.R.; project administration, M.S.I. and W.A.; funding acquisition, R.A., S.H. and R.R. All authors have read and agreed to the published version of the manuscript.

Funding

This work was supported by the Science and Technology Ph.D. Research Startup Project under No. Grant SZIIT2022KJ001 and the funding of the Guangdong Provincial Research Platform and Project (2022KQNCX233).

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Not applicable.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Otálora, S.; Atzori, M.; Andrearczyk, V.; Khan, A.; Müller, H. Staining invariant features for improving generalization of deep convolutional neural networks in computational pathology. Front. Bioeng. Biotechnol. 2019, 7, 198. [Google Scholar] [CrossRef] [PubMed]
  2. Duggento, A.; Conti, A.; Mauriello, A.; Guerrisi, M.; Toschi, N. Deep computational pathology in breast cancer. In Seminars in Cancer Biology; Elsevier: Amsterdam, The Netherlands, 2021; pp. 226–237. [Google Scholar]
  3. Bray, F.; Ferlay, J.; Soerjomataram, I.; Siegel, R.L.; Torre, L.A.; Jemal, A. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA A Cancer J. Clin. 2018, 68, 394–424. [Google Scholar] [CrossRef] [Green Version]
  4. Allred, D.C.; Carlson, R.W.; Berry, D.A.; Burstein, H.J.; Edge, S.B.; Goldstein, L.J.; Gown, A.; Hammond, M.E.; Iglehart, J.D.; Moench, S. NCCN task force report: Estrogen receptor and progesterone receptor testing in breast cancer by immunohistochemistry. J. Natl. Compr. Cancer Netw. 2009, 7, S-1–S-21. [Google Scholar] [CrossRef] [PubMed]
  5. Couture, H.D.; Williams, L.A.; Geradts, J.; Nyante, S.J.; Butler, E.N.; Marron, J.; Perou, C.M.; Troester, M.A.; Niethammer, M. Image analysis with deep learning to predict breast cancer grade, ER status, histologic subtype, and intrinsic subtype. NPJ Breast Cancer 2018, 4, 30. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  6. Rawat, R.R.; Ruderman, D.; Macklin, P.; Rimm, D.L.; Agus, D.B. Correlating nuclear morphometric patterns with estrogen receptor status in breast cancer pathologic specimens. NPJ Breast Cancer 2018, 4, 32. [Google Scholar] [CrossRef] [Green Version]
  7. Tang, P.; Tse, G.M. Immunohistochemical surrogates for molecular classification of breast carcinoma: A 2015 update. Arch. Pathol. Lab. Med. 2016, 140, 806–814. [Google Scholar] [CrossRef] [Green Version]
  8. Gown, A.M. Current issues in ER and HER2 testing by IHC in breast cancer. Mod. Pathol. 2008, 21, S8–S15. [Google Scholar] [CrossRef] [Green Version]
  9. Hammond, M.E.H.; Hayes, D.F.; Dowsett, M.; Allred, D.C.; Hagerty, K.L.; Badve, S.; Fitzgibbons, P.L.; Francis, G.; Goldstein, N.S.; Hayes, M. American Society of Clinical Oncology/College of American Pathologists guideline recommendations for immunohistochemical testing of estrogen and progesterone receptors in breast cancer (unabridged version). Arch. Pathol. Lab. Med. 2010, 134, e48–e72. [Google Scholar] [CrossRef]
  10. Shamai, G.; Binenbaum, Y.; Slossberg, R.; Duek, I.; Gil, Z.; Kimmel, R. Artificial intelligence algorithms to assess hormonal status from tissue microarrays in patients with breast cancer. JAMA Netw. Open 2019, 2, e197700. [Google Scholar] [CrossRef] [Green Version]
  11. Khouja, M.H.; Baekelandt, M.; Sarab, A.; Nesland, J.M.; Holm, R. Limitations of tissue microarrays compared with whole tissue sections in survival analysis. Oncol. Lett. 2010, 1, 827–831. [Google Scholar] [CrossRef]
  12. Griffin, M.C.; Robinson, R.A.; Trask, D.K. Validation of tissue microarrays using p53 immunohistochemical studies of squamous cell carcinoma of the larynx. Mod. Pathol. 2003, 16, 1181–1188. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  13. Ehteshami Bejnordi, B.; Mullooly, M.; Pfeiffer, R.M.; Fan, S.; Vacek, P.M.; Weaver, D.L.; Herschorn, S.; Brinton, L.A.; van Ginneken, B.; Karssemeijer, N. Using deep convolutional neural networks to identify and classify tumor-associated stroma in diagnostic breast biopsies. Mod. Pathol. 2018, 31, 1502–1512. [Google Scholar] [CrossRef] [PubMed]
  14. Deebak, B.D.; Al-Turjman, F. Smart mutual authentication protocol for cloud based medical healthcare systems using internet of medical things. IEEE J. Sel. Areas Commun. 2020, 39, 346–360. [Google Scholar] [CrossRef]
  15. Coudray, N.; Ocampo, P.S.; Sakellaropoulos, T.; Narula, N.; Snuderl, M.; Fenyö, D.; Moreira, A.L.; Razavian, N.; Tsirigos, A. Classification and mutation prediction from non–small cell lung cancer histopathology images using deep learning. Nat. Med. 2018, 24, 1559–1567. [Google Scholar] [CrossRef] [PubMed]
  16. Jemal, A.; Center, M.M.; DeSantis, C.; Ward, E.M. Global Patterns of Cancer Incidence and Mortality Rates and TrendsGlobal Patterns of Cancer. Cancer Epidemiol. Biomark. Prev. 2010, 19, 1893–1907. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  17. Howlader, N.; Altekruse, S.F.; Li, C.I.; Chen, V.W.; Clarke, C.A.; Ries, L.A.; Cronin, K.A. US incidence of breast cancer subtypes defined by joint hormone receptor and HER2 status. JNCI J. Natl. Cancer Inst. 2014, 106, 1–8. [Google Scholar] [CrossRef] [Green Version]
  18. Paik, S.; Shak, S.; Tang, G.; Kim, C.; Baker, J.; Cronin, M.; Baehner, F.L.; Walker, M.G.; Watson, D.; Park, T. A multigene assay to predict recurrence of tamoxifen-treated, node-negative breast cancer. N. Engl. J. Med. 2004, 351, 2817–2826. [Google Scholar] [CrossRef] [Green Version]
  19. Van De Vijver, M.J.; He, Y.D.; Van’t Veer, L.J.; Dai, H.; Hart, A.A.; Voskuil, D.W.; Schreiber, G.J.; Peterse, J.L.; Roberts, C.; Marton, M.J. A gene-expression signature as a predictor of survival in breast cancer. N. Engl. J. Med. 2002, 347, 1999–2009. [Google Scholar] [CrossRef] [Green Version]
  20. Lænkholm, A.-V.; Jensen, M.-B.; Eriksen, J.O.; Rasmussen, B.B.; Knoop, A.S.; Buckingham, W.; Ferree, S.; Schaper, C.; Nielsen, T.O.; Haffner, T. PAM50 risk of recurrence score predicts 10-year distant recurrence in a comprehensive Danish cohort of postmenopausal women allocated to 5 years of endocrine therapy for hormone receptor–positive early breast cancer. J. Clin. Oncol. 2018, 36, 735–740. [Google Scholar] [CrossRef]
  21. Sestak, I.; Buus, R.; Cuzick, J.; Dubsky, P.; Kronenwett, R.; Denkert, C.; Ferree, S.; Sgroi, D.; Schnabel, C.; Baehner, F.L. Comparison of the performance of 6 prognostic signatures for estrogen receptor–positive breast cancer: A secondary analysis of a randomized clinical trial. JAMA Oncol. 2018, 4, 545–553. [Google Scholar] [CrossRef]
  22. Paik, S.; Tang, G.; Shak, S.; Kim, C.; Baker, J.; Kim, W.; Cronin, M.; Baehner, F.L.; Watson, D.; Bryant, J. Gene expression and benefit of chemotherapy in women with node-negative, estrogen receptor–positive breast cancer. J. Clin. Oncol. 2006, 24, 3726–3734. [Google Scholar] [CrossRef]
  23. Sparano, J.A.; Gray, R.J.; Makower, D.F.; Pritchard, K.I.; Albain, K.S.; Hayes, D.F.; Geyer, C.E., Jr.; Dees, E.C.; Goetz, M.P.; Olson, J.A., Jr. Adjuvant chemotherapy guided by a 21-gene expression assay in breast cancer. N. Engl. J. Med. 2018, 379, 111–121. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  24. Wang, S.-Y.; Dang, W.; Richman, I.; Mougalian, S.S.; Evans, S.B.; Gross, C.P. Cost-effectiveness analyses of the 21-gene assay in breast cancer: Systematic review and critical appraisal. J. Clin. Oncol. 2018, 36, 1619. [Google Scholar] [CrossRef] [PubMed]
  25. Macabeo-Ong, M.; Ginzinger, D.G.; Dekker, N.; McMillan, A.; Regezi, J.A.; Wong, D.T.; Jordan, R.C. Effect of duration of fixation on quantitative reverse transcription polymerase chain reaction analyses. Mod. Pathol. 2002, 15, 979–987. [Google Scholar] [CrossRef] [PubMed]
  26. Guinney, J.; Dienstmann, R.; Wang, X.; De Reynies, A.; Schlicker, A.; Soneson, C.; Marisa, L.; Roepman, P.; Nyamundanda, G.; Angelino, P. The consensus molecular subtypes of colorectal cancer. Nat. Med. 2015, 21, 1350–1356. [Google Scholar] [CrossRef] [PubMed]
  27. Paeng, K.; Jung, G.; Lee, S.; Cho, S.Y.; Cho, E.Y.; Song, S.Y. Pan-cancer analysis of tumor microenvironment using deep learning-based cancer stroma and immune profiling in H&E images. Cancer Res. 2019, 79, 2445. [Google Scholar]
  28. Bale, R.; Putzer, D.; Schullian, P. Local treatment of breast cancer liver metastasis. Cancers 2019, 11, 1341. [Google Scholar] [CrossRef] [Green Version]
  29. Børresen-Dale, A.L. TP53 and breast cancer. Hum. Mutat. 2003, 21, 292–300. [Google Scholar] [CrossRef]
  30. Litjens, G.; Sánchez, C.I.; Timofeeva, N.; Hermsen, M.; Nagtegaal, I.; Kovacs, I.; Hulsbergen-Van De Kaa, C.; Bult, P.; Van Ginneken, B.; Van Der Laak, J. Deep learning as a tool for increased accuracy and efficiency of histopathological diagnosis. Sci. Rep. 2016, 6, 26286. [Google Scholar] [CrossRef] [Green Version]
  31. Paeng, K.; Hwang, S.; Park, S.; Kim, M. A unified framework for tumor proliferation score prediction in breast histopathology. In Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support; Springer: Berlin/Heidelberg, Germany, 2017; pp. 231–239. [Google Scholar]
  32. Eskandarian, R.; Sani, Z.A.; Behjati, M.; Zahmatkesh, M.; Haddadi, A.; Kakhi, K.; Roshanzamir, M.; Shoeibi, A.; Alizadehsani, R.; Hussain, S. Identification of clinical features associated with mortality in COVID-19 patients. medRxiv 2021, 1–12. [Google Scholar] [CrossRef]
  33. Dietterich, T.G.; Lathrop, R.H.; Lozano-Pérez, T. Solving the multiple instance problem with axis-parallel rectangles. Artif. Intell. 1997, 89, 31–71. [Google Scholar] [CrossRef]
  34. Campanella, G.; Hanna, M.G.; Geneslaw, L.; Miraflor, A.; Werneck Krauss Silva, V.; Busam, K.J.; Brogi, E.; Reuter, V.E.; Klimstra, D.S.; Fuchs, T.J. Clinical-grade computational pathology using weakly supervised deep learning on whole slide images. Nat. Med. 2019, 25, 1301–1309. [Google Scholar] [CrossRef] [PubMed]
  35. Yu, K.-H.; Wang, F.; Berry, G.J.; Re, C.; Altman, R.B.; Snyder, M.; Kohane, I.S. Classifying non-small cell lung cancer types and transcriptomic subtypes using convolutional neural networks. J. Am. Med. Inform. Assoc. 2020, 27, 757–769. [Google Scholar] [CrossRef] [PubMed]
  36. Feng, Y.; Spezia, M.; Huang, S.; Yuan, C.; Zeng, Z.; Zhang, L.; Ji, X.; Liu, W.; Huang, B.; Luo, W. Breast cancer development and progression: Risk factors, cancer stem cells, signaling pathways, genomics, and molecular pathogenesis. Genes Dis. 2018, 5, 77–106. [Google Scholar] [CrossRef] [PubMed]
  37. Sharifrazi, D.; Alizadehsani, R.; Joloudari, J.H.; Shamshirband, S.; Hussain, S.; Sani, Z.A.; Hasanzadeh, F.; Shoaibi, A.; Dehzangi, A.; Alinejad-Rokny, H. CNN-KCL: Automatic Myocarditis Diagnosis Using Convolutional Neural Network Combined with K-Means Clustering. 2020. Available online: https://www.preprints.org/manuscript/202007.0650/v1 (accessed on 11 November 2022).
  38. Koohestani, A.; Nahavandi, D.; Asadi, H.; Kebria, P.M.; Khosravi, A.; Alizadehsani, R.; Nahavandi, S. A knowledge discovery in motion sickness: A comprehensive literature review. IEEE Access 2019, 7, 85755–85770. [Google Scholar] [CrossRef]
  39. Shoeibi, A.; Moridian, P.; Khodatars, M.; Ghassemi, N.; Jafari, M.; Alizadehsani, R.; Kong, Y.; Gorriz, J.M.; Ramírez, J.; Khosravi, A. An overview of deep learning techniques for epileptic seizures detection and prediction based on neuroimaging modalities: Methods, challenges, and future works. Comput. Biol. Med. 2022, 149, 106053. [Google Scholar] [CrossRef]
  40. Nahavandi, D.; Alizadehsani, R.; Khosravi, A.; Acharya, U.R. Application of artificial intelligence in wearable devices: Opportunities and challenges. Comput. Methods Programs Biomed. 2022, 213, 106541. [Google Scholar] [CrossRef]
  41. Cui, M.; Zhang, D.Y. Artificial intelligence and computational pathology. Lab. Investig. 2021, 101, 412–422. [Google Scholar] [CrossRef]
  42. Lotter, W.; Diab, A.R.; Haslam, B.; Kim, J.G.; Grisot, G.; Wu, E.; Wu, K.; Onieva, J.O.; Boyer, Y.; Boxerman, J.L. Robust breast cancer detection in mammography and digital breast tomosynthesis using an annotation-efficient deep learning approach. Nat. Med. 2021, 27, 244–249. [Google Scholar] [CrossRef]
  43. Szegedy, C.; Vanhoucke, V.; Ioffe, S.; Shlens, J.; Wojna, Z. Rethinking the inception architecture for computer vision. In Proceedings of the IEEE Conference on Computer Vision and Pattern RecognitionLas, Vegas, NV, USA, 27–30 June 2016; pp. 2818–2826. [Google Scholar]
  44. Litjens, G.; Bandi, P.; Ehteshami Bejnordi, B.; Geessink, O.; Balkenhol, M.; Bult, P.; Halilovic, A.; Hermsen, M.; van de Loo, R.; Vogels, R. 1399 H&E-stained sentinel lymph node sections of breast cancer patients: The CAMELYON dataset. GigaScience 2018, 7, giy065. [Google Scholar]
  45. Liu, Y.; Gadepalli, K.; Norouzi, M.; Dahl, G.E.; Kohlberger, T.; Boyko, A.; Venugopalan, S.; Timofeev, A.; Nelson, P.Q.; Corrado, G.S. Detecting cancer metastases on gigapixel pathology images. arXiv 2017, arXiv:1703.02442. [Google Scholar]
  46. Wang, D.; Khosla, A.; Gargeya, R.; Irshad, H.; Beck, A.H. Deep learning for identifying metastatic breast cancer. arXiv 2016, arXiv:1606.05718. [Google Scholar]
  47. Khosravi, P.; Kazemi, E.; Imielinski, M.; Elemento, O.; Hajirasouliha, I. Deep convolutional neural networks enable discrimination of heterogeneous digital pathology images. eBioMedicine 2018, 27, 317–328. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  48. Nazeri, K.; Aminpour, A.; Ebrahimi, M. Two-stage convolutional neural network for breast cancer histology image classification. In Proceedings of the International Conference Image Analysis and Recognition, Póvoa de Varzim, Portugal, 27–29 June 2018; pp. 717–726. [Google Scholar]
  49. Cho, S.Y.; Lee, J.H.; Ryu, J.M.; Lee, J.E.; Cho, E.Y.; Ahn, C.H.; Paeng, K.; Yoo, I.; Ock, C.-Y.; Song, S.Y. Deep learning from HE slides predicts the clinical benefit from adjuvant chemotherapy in hormone receptor-positive breast cancer patients. Sci. Rep. 2021, 11, 17363. [Google Scholar] [CrossRef] [PubMed]
  50. Shen, L.; Margolies, L.R.; Rothstein, J.H.; Fluder, E.; McBride, R.; Sieh, W. Deep learning to improve breast cancer detection on screening mammography. Sci. Rep. 2019, 9, 12495. [Google Scholar] [CrossRef] [Green Version]
  51. Petrini, D.G.; Shimizu, C.; Roela, R.A.; Valente, G.V.; Folgueira, M.A.A.K.; Kim, H.Y. Breast Cancer Diagnosis in Two-View Mammography Using End-to-End Trained EfficientNet-Based Convolutional Network. IEEE Access 2022, 10, 77723–77731. [Google Scholar] [CrossRef]
  52. Shareef, B.; Xian, M.; Vakanski, A. Stan: Small tumor-aware network for breast ultrasound image segmentation. In Proceedings of the 2020 IEEE 17th International Symposium on Biomedical Imaging (ISBI), Iowa City, IA, USA, 3–7 April 2020; pp. 1–5. [Google Scholar]
  53. Negahbani, F.; Sabzi, R.; Jahromi, B.P.; Movahedi, F.; Shirazi, M.K.; Majidi, S.; Firouzabadi, D.; Dehghanian, A. PathoNet: Deep learning assisted evaluation of Ki-67 and tumor infiltrating lymphocytes (TILs) as prognostic factors in breast cancer; A large dataset and baseline. arXiv 2020, arXiv:2010.04713. [Google Scholar]
  54. Patil, A.; Tamboli, D.; Meena, S.; Anand, D.; Sethi, A. Breast Cancer histopathology image classification and localization using multiple instance learning. In Proceedings of the 2019 IEEE International WIE Conference on Electrical and Computer Engineering (WIECON-ECE), Bengaluru, India, 15–16 November 2019; pp. 1–4. [Google Scholar]
  55. Jaume, G.; Pati, P.; Bozorgtabar, B.; Foncubierta, A.; Anniciello, A.M.; Feroce, F.; Rau, T.; Thiran, J.-P.; Gabrani, M.; Goksel, O. Quantifying explainers of graph neural networks in computational pathology. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Nashville, TN, USA, 19–25 June 2021; pp. 8106–8116. [Google Scholar]
  56. Chauhan, R.; Vinod, P.; Jawahar, C. Exploring Genetic-histologic Relationships in Breast Cancer. In Proceedings of the 2021 IEEE 18th International Symposium on Biomedical Imaging (ISBI), Nice, France, 13–16 April 2021; pp. 1187–1190. [Google Scholar]
  57. Blackwell, K.L.; Burstein, H.J.; Storniolo, A.M.; Rugo, H.; Sledge, G.; Koehler, M.; Ellis, C.; Casey, M.; Vukelja, S.; Bischoff, J. Randomized study of Lapatinib alone or in combination with trastuzumab in women with ErbB2-positive, trastuzumab-refractory metastatic breast cancer. J. Clin. Oncol. 2010, 28, 1124–1130. [Google Scholar] [CrossRef]
  58. Hou, L.; Samaras, D.; Kurc, T.M.; Gao, Y.; Davis, J.E.; Saltz, J.H. Patch-based convolutional neural network for whole slide tissue image classification. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, NV, USA, 27–30 June 2016; pp. 2424–2433. [Google Scholar]
  59. Bonnet, E. Using convolutional neural networks for the classification of breast cancer images. arXiv 2021, arXiv:2108.13661. [Google Scholar]
  60. Wang, F.; Kapse, S.; Liu, S.; Prasanna, P.; Chen, C. TopoTxR: A topological biomarker for predicting treatment response in breast cancer. In Proceedings of the International Conference on Information Processing in Medical Imaging, Virtual Event, 28–30 June 2021; pp. 386–397. [Google Scholar]
  61. Naik, N.; Madani, A.; Esteva, A.; Keskar, N.S.; Press, M.F.; Ruderman, D.; Agus, D.B.; Socher, R. Deep learning-enabled breast cancer hormonal receptor status determination from base-level H&E stains. Nat. Commun. 2020, 11, 5727. [Google Scholar]
  62. He, B.; Bergenstråhle, L.; Stenbeck, L.; Abid, A.; Andersson, A.; Borg, Å.; Maaskola, J.; Lundeberg, J.; Zou, J. Integrating spatial gene expression and breast tumour morphology via deep learning. Nat. Biomed. Eng. 2020, 4, 827–834. [Google Scholar] [CrossRef] [PubMed]
  63. Senousy, Z.; Abdelsamea, M.M.; Gaber, M.M.; Abdar, M.; Acharya, U.R.; Khosravi, A.; Nahavandi, S. MCUa: Multi-level context and uncertainty aware dynamic deep ensemble for breast cancer histology image classification. IEEE Trans. Biomed. Eng. 2021, 69, 818–829. [Google Scholar] [CrossRef] [PubMed]
  64. Gamble, P.; Jaroensri, R.; Wang, H.; Tan, F.; Moran, M.; Brown, T.; Flament-Auvigne, I.; Rakha, E.A.; Toss, M.; Dabbs, D.J. Determining breast cancer biomarker status and associated morphological features using deep learning. Commun. Med. 2021, 1, 14. [Google Scholar] [CrossRef] [PubMed]
  65. Karthik, R.; Menaka, R.; Siddharth, M. Classification of breast cancer from histopathology images using an ensemble of deep multiscale networks. Biocybern. Biomed. Eng. 2022, 42, 963–976. [Google Scholar] [CrossRef]
  66. Hao, Y.; Zhang, L.; Qiao, S.; Bai, Y.; Cheng, R.; Xue, H.; Hou, Y.; Zhang, W.; Zhang, G. Breast cancer histopathological images classification based on deep semantic features and gray level co-occurrence matrix. PLoS ONE 2022, 17, e0267955. [Google Scholar] [CrossRef] [PubMed]
  67. Allugunti, V.R. Breast cancer detection based on thermographic images using machine learning and deep learning algorithms. Int. J. Eng. Comput. Sci. 2022, 4, 49–56. [Google Scholar]
  68. Wang, X.; Ahmad, I.; Javeed, D.; Zaidi, S.A.; Alotaibi, F.M.; Ghoneim, M.E.; Daradkeh, Y.I.; Asghar, J.; Eldin, E.T. Intelligent Hybrid Deep Learning Model for Breast Cancer Detection. Electronics 2022, 11, 2767. [Google Scholar] [CrossRef]
  69. Alizadehsani, R.; Roshanzamir, M.; Abdar, M.; Beykikhoshk, A.; Zangooei, M.H.; Khosravi, A.; Nahavandi, S.; Tan, R.S.; Acharya, U.R. Model uncertainty quantification for diagnosis of each main coronary artery stenosis. Soft Comput. 2020, 24, 10149–10160. [Google Scholar] [CrossRef]
  70. Alizadehsani, R.; Roshanzamir, M.; Abdar, M.; Beykikhoshk, A.; Khosravi, A.; Nahavandi, S.; Plawiak, P.; Tan, R.S.; Acharya, U.R. Hybrid genetic-discretized algorithm to handle data uncertainty in diagnosing stenosis of coronary arteries. Expert Syst. 2022, 39, e12573. [Google Scholar] [CrossRef]
  71. Joloudari, J.H.; Mojrian, S.; Nodehi, I.; Mashmool, A.; Zadegan, Z.K.; Shirkharkolaie, S.K.; Alizadehsani, R.; Tamadon, T.; Khosravi, S.; Kohnehshari, M.A. Application of Artificial Intelligence Techniques for Automated Detection of Myocardial Infarction: A Review. Physiol. Meas. 2022, 43, 08TR01. [Google Scholar] [CrossRef]
  72. Nasab, R.Z.; Ghamsari, M.R.E.; Argha, A.; Macphillamy, C.; Beheshti, A.; Alizadehsani, R.; Lovell, N.H.; Alinejad-Rokny, H. Deep Learning in Spatially Resolved Transcriptomics: A Comprehensive Technical View. arXiv 2022, arXiv:2210.04453. [Google Scholar]
  73. Kakhi, K.; Alizadehsani, R.; Kabir, H.D.; Khosravi, A.; Nahavandi, S.; Acharya, U.R. The internet of medical things and artificial intelligence: Trends, challenges, and opportunities. Biocybern. Biomed. Eng. 2022, 42, 749–771. [Google Scholar] [CrossRef]
  74. Alizadehsani, R.; Roshanzamir, M.; Hussain, S.; Khosravi, A.; Koohestani, A.; Zangooei, M.H.; Abdar, M.; Beykikhoshk, A.; Shoeibi, A.; Zare, A. Handling of uncertainty in medical data using machine learning and probability theory techniques: A review of 30 years (1991–2020). Ann. Oper. Res. 2021, 1–42. [Google Scholar] [CrossRef]
  75. Ayoobi, N.; Sharifrazi, D.; Alizadehsani, R.; Shoeibi, A.; Gorriz, J.M.; Moosaei, H.; Khosravi, A.; Nahavandi, S.; Chofreh, A.G.; Goni, F.A. Time series forecasting of new cases and new deaths rate for COVID-19 using deep learning methods. Results Phys. 2021, 27, 104495. [Google Scholar] [CrossRef]
  76. Asgharnezhad, H.; Shamsi, A.; Alizadehsani, R.; Khosravi, A.; Nahavandi, S.; Sani, Z.A.; Srinivasan, D.; Islam, S.M.S. Objective evaluation of deep uncertainty predictions for COVID-19 detection. Sci. Rep. 2022, 12, 815. [Google Scholar] [CrossRef]
  77. Alizadehsani, R.; Hosseini, M.J.; Boghrati, R.; Ghandeharioun, A.; Khozeimeh, F.; Sani, Z.A. Exerting cost-sensitive and feature creation algorithms for coronary artery disease diagnosis. Int. J. Knowl. Discov. Bioinform. 2012, 3, 59–79. [Google Scholar] [CrossRef]
  78. Liu, J.; Lin, Z.; Padhy, S.; Tran, D.; Bedrax Weiss, T.; Lakshminarayanan, B. Simple and principled uncertainty estimation with deterministic deep learning via distance awareness. Adv. Neural Inf. Process. Syst. 2020, 33, 7498–7512. [Google Scholar]
  79. Chua, K.; Calandra, R.; McAllister, R.; Levine, S. Deep reinforcement learning in a handful of trials using probabilistic dynamics models. Adv. Neural Inf. Process. Syst. 2018, 31, 1–12. [Google Scholar]
  80. Wickstrøm, K.; Mikalsen, K.Ø.; Kampffmeyer, M.; Revhaug, A.; Jenssen, R. Uncertainty-aware deep ensembles for reliable and explainable predictions of clinical time series. IEEE J. Biomed. Health Inform. 2020, 25, 2435–2444. [Google Scholar] [CrossRef]
  81. World Health Organization. World Health Statistics 2008; World Health Organization: Geneva, Switzerland, 2008. [Google Scholar]
  82. Bennett, J.E.; Stevens, G.A.; Mathers, C.D.; Bonita, R.; Rehm, J.; Kruk, M.E.; Riley, L.M.; Dain, K.; Kengne, A.P.; Chalkidou, K. NCD Countdown 2030: Worldwide trends in non-communicable disease mortality and progress towards Sustainable Development Goal target 3.4. Lancet 2018, 392, 1072–1088. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  83. Simonyan, K.; Zisserman, A. Very deep convolutional networks for large-scale image recognition. arXiv 2014, arXiv:1409.1556. [Google Scholar]
  84. Landmark, B.T.; Wahl, A. Living with newly diagnosed breast cancer: A qualitative study of 10 women with newly diagnosed breast cancer. J. Adv. Nurs. 2002, 40, 112–121. [Google Scholar] [CrossRef] [PubMed]
  85. Muthoni, A.; Miller, A.N. An exploration of rural and urban Kenyan women’s knowledge and attitudes regarding breast cancer and breast cancer early detection measures. Health Care Women Int. 2010, 31, 801–816. [Google Scholar] [CrossRef] [PubMed]
  86. Tjoa, E.; Guan, C. A survey on explainable artificial intelligence (xai): Toward medical xai. IEEE Trans. Neural Netw. Learn. Syst. 2020, 32, 4793–4813. [Google Scholar] [CrossRef] [PubMed]
  87. Ilse, M.; Tomczak, J.; Welling, M. Attention-based deep multiple instance learning. In Proceedings of the International Conference on Machine Learning, Stockholm, Sweden, 10–15 July 2018; pp. 2127–2136. [Google Scholar]
  88. Xin, M.; Qiao, Z.; Li, J.; Liu, J.; Song, S.; Zhao, X.; Miao, P.; Tang, T.; Wang, L.; Liu, W. miR-22 inhibits tumor growth and metastasis by targeting ATP citrate lyase: Evidence in osteosarcoma, prostate cancer, cervical cancer and lung cancer. Oncotarget 2016, 7, 44252. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  89. Shalamzari, S.A.; Agha-Alinejad, H.; Alizadeh, S.; Shahbazi, S.; Khatib, Z.K.; Kazemi, A.; Saei, M.A.; Minayi, N. The effect of exercise training on the level of tissue IL-6 and vascular endothelial growth factor in breast cancer bearing mice. Iran. J. Basic Med. Sci. 2014, 17, 231. [Google Scholar]
  90. Adraskela, K.; Veisaki, E.; Koutsilieris, M.; Philippou, A. Physical exercise positively influences breast cancer evolution. Clin. Breast Cancer 2017, 17, 408–417. [Google Scholar] [CrossRef]
  91. Akinyemiju, T.F. Socio-economic and health access determinants of breast and cervical cancer screening in low-income countries: Analysis of the World Health Survey. PLoS ONE 2012, 7, e48834. [Google Scholar] [CrossRef]
  92. Kakileti, S.T.; Madhu, H.J.; Manjunath, G.; Wee, L.; Dekker, A.; Sampangi, S. Personalized risk prediction for breast cancer pre-screening using artificial intelligence and thermal radiomics. Artif. Intell. Med. 2020, 105, 101854. [Google Scholar] [CrossRef]
  93. Okonkwo, Q.L.; Draisma, G.; der Kinderen, A.; Brown, M.L.; de Koning, H.J. Breast cancer screening policies in developing countries: A cost-effectiveness analysis for India. J. Natl. Cancer Inst. 2008, 100, 1290–1300. [Google Scholar] [CrossRef] [Green Version]
  94. Robles, S.C.; Galanis, E. Breast cancer in Latin America and the Caribbean. Rev. Panam. Salud Publica 2002, 11, 178–185. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  95. Weir, H.K.; Thompson, T.D.; Stewart, S.L.; White, M.C. Peer Reviewed: Cancer Incidence Projections in the United States Between 2015 and 2050. Prev. Chronic Dis. 2021, 18, E59. [Google Scholar] [CrossRef] [PubMed]
  96. Krijthe, B.P.; Kunst, A.; Benjamin, E.J.; Lip, G.Y.; Franco, O.H.; Hofman, A.; Witteman, J.C.; Stricker, B.H.; Heeringa, J. Projections on the number of individuals with atrial fibrillation in the European Union, from 2000 to 2060. Eur. Heart J. 2013, 34, 2746–2751. [Google Scholar] [CrossRef] [PubMed]
  97. Roseline, R.; Manikandan, S. Determination Breast Cancer Accuracy Using Data Mining. Prof. RK Sharma 2018, 12, 253. [Google Scholar] [CrossRef]
  98. Khatib, O.M.; Modjtabai, A. Guidelines for the Early Detection and Screening of Breast Cancer; World Health Organization, Regional Office for the Eastern Mediterranean: Cairo, Egypt, 2006; p. 50. [Google Scholar]
  99. Tajaddini, A.; Pourzand, A.; Sanaat, Z.; Pirouzpanah, S. Dietary resistant starch contained foods and breast cancer risk: A case-control study in northwest of Iran. Asian Pac. J. Cancer Prev. 2015, 16, 4185–4192. [Google Scholar] [CrossRef] [Green Version]
  100. Mamishi, N. Evaluating Predictive Factors for Engaging in Positive Breast Health Behaviours: A Quantitative Study among Iranian Immigrant Women in Montreal. Ph.D. Thesis, McGill University, Montréal, QC, Canada, 2018. [Google Scholar]
  101. Mehrabi, E.; Hajian, S.; Simbar, M.; Hoshyari, M.; Zayeri, F. Coping response following a diagnosis of breast cancer: A systematic review. Electron. Physician 2015, 7, 1575. [Google Scholar] [CrossRef] [Green Version]
  102. Schmauch, B.; Romagnoni, A.; Pronier, E.; Saillard, C.; Maillé, P.; Calderaro, J.; Kamoun, A.; Sefta, M.; Toldo, S.; Zaslavskiy, M. A deep learning model to predict RNA-Seq expression of tumours from whole slide images. Nat. Commun. 2020, 11, 3877. [Google Scholar] [CrossRef]
  103. Azamjah, N.; Soltan-Zadeh, Y.; Zayeri, F. Global trend of breast cancer mortality rate: A 25-year study. Asian Pac. J. Cancer Prev. 2019, 20, 2015. [Google Scholar] [CrossRef]
  104. Fregene, A.; Newman, L.A. Breast cancer in sub-Saharan Africa: How does it relate to breast cancer in African-American women? Cancer Interdiscip. Int. J. Am. Cancer Soc. 2005, 103, 1540–1550. [Google Scholar] [CrossRef]
  105. Guilloteau, P.; Zabielski, R.; Hammon, H.; Metges, C. Adverse effects of nutritional programming during prenatal and early postnatal life, some aspects of regulation and potential prevention and treatments. J. Physiol. Pharm. 2009, 60, 17–35. [Google Scholar]
  106. Courtiol, P.; Maussion, C.; Moarii, M.; Pronier, E.; Pilcer, S.; Sefta, M.; Manceron, P.; Toldo, S.; Zaslavskiy, M.; Le Stang, N. Deep learning-based classification of mesothelioma improves prediction of patient outcome. Nat. Med. 2019, 25, 1519–1525. [Google Scholar] [CrossRef]
  107. DeVries, T.; Taylor, G.W. Improved regularization of convolutional neural networks with cutout. arXiv 2017, arXiv:1708.04552. [Google Scholar]
  108. Shrivastava, A.; Gupta, A.; Girshick, R. Training region-based object detectors with online hard example mining. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, NV, USA, 27–30 June 2016; pp. 761–769. [Google Scholar]
  109. Fu, Y.; Jung, A.W.; Torne, R.V.; Gonzalez, S.; Vöhringer, H.; Shmatko, A.; Yates, L.R.; Jimenez-Linan, M.; Moore, L.; Gerstung, M. Pan-cancer computational histopathology reveals mutations, tumor composition and prognosis. Nat. Cancer 2020, 1, 800–810. [Google Scholar] [CrossRef] [PubMed]
  110. Gerlinger, M.; Rowan, A.J.; Horswell, S.; Larkin, J.; Endesfelder, D.; Gronroos, E.; Martinez, P.; Matthews, N.; Stewart, A.; Tarpey, P. Intratumor heterogeneity and branched evolution revealed by multiregion sequencing. N. Engl. J. Med. 2012, 366, 883–892. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  111. Ståhl, P.L.; Salmén, F.; Vickovic, S.; Lundmark, A.; Navarro, J.F.; Magnusson, J.; Giacomello, S.; Asp, M.; Westholm, J.O.; Huss, M. Visualization and analysis of gene expression in tissue sections by spatial transcriptomics. Science 2016, 353, 78–82. [Google Scholar] [CrossRef]
  112. Chen, K.H.; Boettiger, A.N.; Moffitt, J.R.; Wang, S.; Zhuang, X. Spatially resolved, highly multiplexed RNA profiling in single cells. Science 2015, 348, aaa6090. [Google Scholar] [CrossRef] [Green Version]
  113. Eng, C.-H.L.; Lawson, M.; Zhu, Q.; Dries, R.; Koulena, N.; Takei, Y.; Yun, J.; Cronin, C.; Karp, C.; Yuan, G.-C. Transcriptome-scale super-resolved imaging in tissues by RNA seqFISH+. Nature 2019, 568, 235–239. [Google Scholar] [CrossRef]
  114. Liu, R.; Mignardi, M.; Jones, R.; Enge, M.; Kim, S.K.; Quake, S.R.; Zou, J. Modeling spatial correlation of transcripts with application to developing pancreas. Sci. Rep. 2019, 9, 5592. [Google Scholar] [CrossRef] [Green Version]
  115. Lee, J.H.; Daugharthy, E.R.; Scheiman, J.; Kalhor, R.; Yang, J.L.; Ferrante, T.C.; Terry, R.; Jeanty, S.S.; Li, C.; Amamoto, R. Highly multiplexed subcellular RNA sequencing in situ. Science 2014, 343, 1360–1363. [Google Scholar] [CrossRef] [Green Version]
  116. Kamentsky, L.; Jones, T.R.; Fraser, A.; Bray, M.-A.; Logan, D.J.; Madden, K.L.; Ljosa, V.; Rueden, C.; Eliceiri, K.W.; Carpenter, A.E. Improved structure, function and compatibility for CellProfiler: Modular high-throughput image analysis software. Bioinformatics 2011, 27, 1179–1180. [Google Scholar] [CrossRef] [Green Version]
  117. Yu, K.-H.; Zhang, C.; Berry, G.J.; Altman, R.B.; Ré, C.; Rubin, D.L.; Snyder, M. Predicting non-small cell lung cancer prognosis by fully automated microscopic pathology image features. Nat. Commun. 2016, 7, 12474. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  118. He, K.; Zhang, X.; Ren, S.; Sun, J. Deep residual learning for image recognition. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, NV, USA, 27–30 June 2016; pp. 770–778. [Google Scholar]
  119. Huang, G.; Liu, Z.; Van Der Maaten, L.; Weinberger, K.Q. Densely connected convolutional networks. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, HI, USA, 21–26 July 2017; pp. 4700–4708. [Google Scholar]
  120. Zhang, Z.; Li, J.; He, T.; Ding, J. Bioinformatics identified 17 immune genes as prognostic biomarkers for breast cancer: Application study based on artificial intelligence algorithms. Front. Oncol. 2020, 10, 330. [Google Scholar] [CrossRef] [PubMed]
  121. Behravan, H.; Hartikainen, J.M.; Tengström, M.; Kosma, V.M.; Mannermaa, A. Predicting breast cancer risk using interacting genetic and demographic factors and machine learning. Sci. Rep. 2020, 10, 11044. [Google Scholar] [CrossRef] [PubMed]
  122. Koohestani, A.; Abdar, M.; Hussain, S.; Khosravi, A.; Nahavandi, D.; Nahavandi, S.; Alizadehsani, R. Analysis of driver performance using hybrid of weighted ensemble learning technique and evolutionary algorithms. Arab. J. Sci. Eng. 2021, 46, 3567–3580. [Google Scholar] [CrossRef]
  123. Liu, X.; Collister, J.A.; Littlejohns, T.J.; Morelli, D.; Clifton, D.A.; Hunter, D.J.; Clifton, L. Combining Machine Learning with Cox models for identifying risk factors for incident post-menopausal breast cancer in the UK Biobank. medRxiv 2022, 1–19. [Google Scholar] [CrossRef]
  124. Sharma, D.; Kumar, R.; Jain, A. A systematic review of risk factors and risk assessment models for breast cancer. In Mobile Radio Communications and 5G Networks; Springer: Singapore, 2021; pp. 509–519. [Google Scholar] [CrossRef]
  125. Dembrower, K.; Wåhlin, E.; Liu, Y.; Salim, M.; Smith, K.; Lindholm, P.; Eklund, M.; Strand, F. Effect of artificial intelligence-based triaging of breast cancer screening mammograms on cancer detection and radiologist workload: A retrospective simulation study. Lancet Digit. Health 2020, 2, e468–e474. [Google Scholar] [CrossRef]
  126. Chen, M.; Zhang, B.; Topatana, W.; Cao, J.; Zhu, H.; Juengpanich, S.; Mao, Q.; Yu, H.; Cai, X. Classification and mutation prediction based on histopathology H&E images in liver cancer using deep learning. NPJ Precis. Oncol. 2020, 4, 1–7. [Google Scholar]
  127. Mobadersany, P.; Yousefi, S.; Amgad, M.; Gutman, D.A.; Barnholtz-Sloan, J.S.; Velázquez Vega, J.E.; Brat, D.J.; Cooper, L.A. Predicting cancer outcomes from histology and genomics using convolutional networks. Proc. Natl. Acad. Sci. USA 2018, 115, E2970–E2979. [Google Scholar] [CrossRef] [Green Version]
  128. Kather, J.N.; Heij, L.R.; Grabsch, H.I.; Loeffler, C.; Echle, A.; Muti, H.S.; Krause, J.; Niehues, J.M.; Sommer, K.A.; Bankhead, P. Pan-cancer image-based detection of clinically actionable genetic alterations. Nat. Cancer 2020, 1, 789–799. [Google Scholar] [CrossRef]
  129. Zheng, H.; Momeni, A.; Cedoz, P.-L.; Vogel, H.; Gevaert, O. Whole slide images reflect DNA methylation patterns of human tumors. NPJ Genom. Med. 2020, 5, 11. [Google Scholar] [CrossRef] [Green Version]
  130. Anand, D.; Kurian, N.C.; Dhage, S.; Kumar, N.; Rane, S.; Gann, P.H.; Sethi, A. Deep learning to estimate human epidermal growth factor receptor 2 status from hematoxylin and eosin-stained breast tissue images. J. Pathol. Inform. 2020, 11, 19. [Google Scholar] [CrossRef] [PubMed]
  131. Li, X.; Xiong, H.; Li, X.; Wu, X.; Zhang, X.; Liu, J.; Bian, J.; Dou, D. Interpretable deep learning: Interpretation, interpretability, trustworthiness, and beyond. arXiv 2021, arXiv:2103.10689. [Google Scholar] [CrossRef]
  132. Karatza, P.; Dalakleidi, K.; Athanasiou, M.; Nikita, K.S. Interpretability methods of machine learning algorithms with applications in breast cancer diagnosis. In Proceedings of the 2021 43rd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC), Virtual Conference, 1–5 November 2021; pp. 2310–2313. [Google Scholar]
  133. Available online: https://christophm.github.io/interpretable-ml-book/ice.html (accessed on 12 November 2022).
  134. Da Cruz, H.F.; Pfahringer, B.; Martensen, T.; Schneider, F.; Meyer, A.; Böttinger, E.; Schapranow, M.-P. Using interpretability approaches to update “black-box” clinical prediction models: An external validation study in nephrology. Artif. Intell. Med. 2021, 111, 101982. [Google Scholar] [CrossRef] [PubMed]
  135. Jafari, M.; Shoeibi, A.; Khodatars, M.; Ghassemi, N.; Moridian, P.; Delfan, N.; Alizadehsani, R.; Khosravi, A.; Ling, S.H.; Zhang, Y.-D. Automated Diagnosis of Cardiovascular Diseases from Cardiac Magnetic Resonance Imaging Using Deep Learning Models: A Review. arXiv 2022, arXiv:2210.14909. [Google Scholar]
  136. Shapley, L.S. A value for n-person games. Class. Game Theory 1997, 69, 1–100. [Google Scholar]
  137. Jafari, M.; Shoeibi, A.; Ghassemi, N.; Heras, J.; Khosravi, A.; Ling, S.H.; Alizadehsani, R.; Beheshti, A.; Zhang, Y.-D.; Wang, S.-H. Automatic Diagnosis of Myocarditis Disease in Cardiac MRI Modality using Deep Transformers and Explainable Artificial Intelligence. arXiv 2022, arXiv:2210.14611. [Google Scholar]
  138. Niu, L.; Cong, W.; Liu, L.; Hong, Y.; Zhang, B.; Liang, J.; Zhang, L. Making images real again: A comprehensive survey on deep image composition. arXiv 2021, arXiv:2106.14490. [Google Scholar]
  139. Murtaza, G.; Shuib, L.; Abdul Wahab, A.W.; Mujtaba, G.; Nweke, H.F.; Al-garadi, M.A.; Zulfiqar, F.; Raza, G.; Azmi, N.A. Deep learning-based breast cancer classification through medical imaging modalities: State of the art and research challenges. Artif. Intell. Rev. 2020, 53, 1655–1720. [Google Scholar] [CrossRef]
  140. Cheerla, A.; Gevaert, O. Deep learning with multimodal representation for pancancer prognosis prediction. Bioinformatics 2019, 35, i446–i454. [Google Scholar] [CrossRef]
  141. Gou, J.; Yu, B.; Maybank, S.J.; Tao, D. Knowledge distillation: A survey. Int. J. Comput. Vis. 2021, 129, 1789–1819. [Google Scholar] [CrossRef]
  142. Allen-Zhu, Z.; Li, Y. Towards understanding ensemble, knowledge distillation and self-distillation in deep learning. arXiv 2020, arXiv:2012.09816. [Google Scholar]
Figure 1. The statistics of the selected papers are categorized according to their publishers.
Figure 1. The statistics of the selected papers are categorized according to their publishers.
Healthcare 10 02395 g001
Figure 2. The end-to-end attachment of two networks: 1. Patch-wise CNN, 2. Image-wise CNN.
Figure 2. The end-to-end attachment of two networks: 1. Patch-wise CNN, 2. Image-wise CNN.
Healthcare 10 02395 g002
Figure 3. Chemotherapy in hormone receptor-positive breast cancer patients.
Figure 3. Chemotherapy in hormone receptor-positive breast cancer patients.
Healthcare 10 02395 g003
Figure 4. Whole image classification and prediction of cancer or normal.
Figure 4. Whole image classification and prediction of cancer or normal.
Healthcare 10 02395 g004
Figure 5. Diagrams of the single-view classifier for the “CV test” (top) and “OD test” (bottom).
Figure 5. Diagrams of the single-view classifier for the “CV test” (top) and “OD test” (bottom).
Healthcare 10 02395 g005
Figure 6. Small Tumor-Aware Network (STAN) to improve the performance of segmenting tumors with different sizes.
Figure 6. Small Tumor-Aware Network (STAN) to improve the performance of segmenting tumors with different sizes.
Healthcare 10 02395 g006
Figure 7. Multi-instance learning architecture for classification of breast cancer histopathology images.
Figure 7. Multi-instance learning architecture for classification of breast cancer histopathology images.
Healthcare 10 02395 g007
Figure 8. Proposed model based on graph breast cancer identification.
Figure 8. Proposed model based on graph breast cancer identification.
Healthcare 10 02395 g008
Figure 9. Predict genomic biomarkers—TP53 mutation model.
Figure 9. Predict genomic biomarkers—TP53 mutation model.
Healthcare 10 02395 g009
Figure 10. Proposed TopoTxR Method.
Figure 10. Proposed TopoTxR Method.
Healthcare 10 02395 g010
Table 1. Breast Cancer datasets and their links.
Table 1. Breast Cancer datasets and their links.
Dataset NameLink
Cancer Waiting Timeshttps://data.world/datasets/breast-cancer (access: 11 November 2022)
NKI Breast Cancer Datahttps://data.world/datasets/breast-cancer (access: 11 November 2022)
Ispy1_Trialhttps://data.world/datasets/breast-cancer (access: 11 November 2022)
Mammographic Masseshttps://data.world/datasets/breast-cancer (access: 11 November 2022)
Uta4 Datasetshttps://data.world/datasets/breast-cancer (access: 11 November 2022)
Breast Cancer Wisconsin (Diagnostic) Datasethttps://www.kaggle.com/uciml/breast-cancer-wisconsin-data (access: 11 November 2022)
Breast Cancerhttps://data.world/uci/breast-cancer (access: 11 November 2022)
Seer Breast Cancer Datahttps://ieee-dataport.org/open-access/seer-breast-cancer-data (access: 11 November 2022)
Breast Cancer Histopathological Database (Breakhis)https://web.inf.ufpr.br/vri/databases/breast-cancer-histopathological-database-breakhis/ (access: 11 November 2022)
Atp5b In Breast Cancerhttps://www.frontiersin.org/articles/10.3389/fgene.2021.652474/full#supplementary-material (access: 11 November 2022)
Single-Modality And Multi-Modalityhttps://data.world/datasets/breast-cancer (access: 11 November 2022)
Atp5b In Breast Cancerhttps://www.frontiersin.org/articles/10.3389/fgene.2021.652474/full#supplementary-material (access: 11 November 2022)
Breast Cancer Datasethttps://archive.ics.uci.edu/ml/datasets/breast+cancer (access: 11 November 2022)
Breast Cancer Wisconsin (Original) Datasethttps://archive.ics.uci.edu/ml/datasets/breast+cancer+wisconsin+(original) (access: 11 November 2022)
Sklearn.Datasetshttps://scikit-learn.org/stable/modules/generated/sklearn.datasets.load_breast_cancer.html (access: 11 November 2022)
Cbis-Ddsm: Breast Cancer Image Datasethttps://www.kaggle.com/awsaf49/cbis-ddsm-breast-cancer-image-dataset (access: 11 November 2022)
Breakhis (Breast Cancer Histopathological Database)https://paperswithcode.com/dataset/breakhis (access: 11 November 2022)
Mammography Databasehttp://marathon.csee.usf.edu/Mammography/Database.html (access: 11 November 2022)
http://www.eng.usf.edu/cvprg/Mammography/Database.html (access: 11 November 2022)
Cancer Data https://portal.gdc.cancer.gov (access: 11 November 2022)
Cbioportalhttps://www.cbioportal.org/ (access: 11 November 2022)
Table 2. Summary of reviewed articles.
Table 2. Summary of reviewed articles.
AuthorsDatasetTaskApproachDisease TypeAccuracy (%)
Nikhil Naik et al. [61]WSI-level annotationsClassificationReceptorNetEstrogen receptor
Prediction
92
William Lotter et al. [42]DDSM, OMI-DBClassificationResNet-50detection in mammography94.5
Bryan He et al. [62]Histopathology imagesClassificationST-NETPrediction of local gene expression66
SooYoun Cho et al. [49]HE slidesClassificationLunit SCOPE, CCNMitotic cells in the cancer epithelium50
Li Shen et al. [50]INbreast dataDetectionCBIS-DDSMScreening mammograms95
Farzin Negahbani et al. [53]SHIDC-BC-Ki-67Cell detection and classificationPathoNetNuclear protein Ki-67 and Tumor infiltrating lymphocytes (TILs)95.6
Kamyar Nazeri et al. [48]Breast Cancer Histology (BACH)ClassificationICIAR, CNNTissue, classification microscopy images95
Abhijeet Patil et al. [54]BreakHIS, BACHClassificationA-MILHE slides predicts84.43
Bryar Shareef et al. [52]BUSIS, Dataset BSegmentationSTANImprove the performance of segmenting tumors90.2
Guillaume Jaume et al. [55]BRACSDetectionClass separability computationConventional pixel-wise analysis62.5
Eric Bonnet [59]PCam (Patch CAMELYON)ClassificationShelf deep learning frameworkBreast cancer metastasis tissue89
Fan Wang [60]e I-SPY1ClassificationTopoTxRTissue structures85.1
Zakaria Senousy [63]Histology imageClassificationMCUaBreast histology image classification98.11
Ruchi Chauhan et al. [56]TCGA datasetClassificationGenetic-histologic RelationshipsGenomic biomarkers90.9
Daniel G.P. Petrini et al. [51]CBIS-DDSMClassificationEfficientNet-Based ConvolutionalTwo-View Mammography87.57
Paul Gamble et al. [64]FFPE, TCGA and H&E slidesClassificationTwo-stage deep learning system (DLS),Predict ER/PR/HER2, tissue regions (patches)93.9
Karthik et al. [65]BreakHisClassificationEnsemble of DL modelsBreast cancer99.55
Hao et al. [66]BreaKHisClassificationFeatures based on DenseNet201 fused with gray-level co-occurrence matrix, support vector machineBreast cancer96.75
VR Allugunti [67]A Kaggle datasetClassificationRandom forestBreast cancer90.55
Wang et al.
[68]
Kaggle Histopathologic cancer detectionClassificationHybrid DL modelBreast cancer86.21
Table 3. Articles that have open-source codes.
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Iqbal, M.S.; Ahmad, W.; Alizadehsani, R.; Hussain, S.; Rehman, R. Breast Cancer Dataset, Classification and Detection Using Deep Learning. Healthcare 2022, 10, 2395. https://doi.org/10.3390/healthcare10122395

AMA Style

Iqbal MS, Ahmad W, Alizadehsani R, Hussain S, Rehman R. Breast Cancer Dataset, Classification and Detection Using Deep Learning. Healthcare. 2022; 10(12):2395. https://doi.org/10.3390/healthcare10122395

Chicago/Turabian Style

Iqbal, Muhammad Shahid, Waqas Ahmad, Roohallah Alizadehsani, Sadiq Hussain, and Rizwan Rehman. 2022. "Breast Cancer Dataset, Classification and Detection Using Deep Learning" Healthcare 10, no. 12: 2395. https://doi.org/10.3390/healthcare10122395

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop