Journal of Cancer Research and Immuno-Oncology
Open Access

A Comprehensive Review of Deep Learning and Artificial Intelligence in Cancer Diagnosis

Deepa Rangasamy^{* *}Correspondence: Deepa Rangasamy, Department of Electronics and Communication Engineering, Nehru Institute of Engineering and Technology, Coimbatore, Tamilnadu, India, Email:

Author info »

Introduction

Cancer diagnosis plays a pivotal role in patient management, treatment planning, and prognosis. Artificial intelligence techniques, such as machine learning and deep learning, have shown promising results in improving the accuracy and efficiency of cancer diagnosis. This section provides an overview of the significance of accurate cancer diagnosis, the potential of AI in addressing diagnostic challenges, and the objectives of this systematic review [1].

Cancer is a major global health challenge, with millions of new cases diagnosed each year. Early and accurate detection of cancer plays a crucial role in improving patient outcomes and survival rates. In recent years, the rapid advancement of Artificial Intelligence (AI) has presented new opportunities for cancer detection and diagnosis. AI techniques, particularly machine learning and deep learning algorithms, have demonstrated remarkable capabilities in analyzing complex medical data and identifying patterns that may be indicative of cancer [2].

The integration of AI in cancer detection has the potential to revolutionize clinical practice by improving diagnostic accuracy, efficiency, and patient management. AI algorithms can process large volumes of data, including medical images, genomic data, electronic health records, and pathology slides, to assist healthcare professionals in making informed decisions. By leveraging the power of AI, it is possible to detect cancers at earlier stages, differentiate between benign and malignant lesions, and predict treatment response.

Furthermore, we assess the performance metrics used to evaluate the effectiveness of AI algorithms in cancer detection, such as sensitivity, specificity, accuracy, and Area Under the Curve (AUC). By analyzing the existing evidence, we will highlight the strengths and limitations of AI in cancer detection, as well as the challenges and potential barriers to its widespread implementation in clinical practice [3-5].

Literature Review

This review aims to contribute to the growing body of knowledge on AI in cancer detection and provide insights into the current state of research and future directions. The integration of AI has the potential to enhance cancer detection capabilities, improve patient outcomes, and ultimately reduce the burden of cancer on individuals and healthcare systems. The objective of this review is to provide a comprehensive overview of the current state of AI in cancer detection. Through a systematic review of the literature, we aim to summarize the key findings and advancements in the field. We will examine the different cancer types for which AI has been applied, including breast, lung, skin, prostate, and others. Additionally, we will explore the AI algorithms utilized, such as machine learning and deep learning, and the diagnostic modalities employed, including medical imaging, pathology, and genetic analysis [6-8].

Methodology

This systematic review follows a predefined protocol and rigorous search strategy to ensure a comprehensive and unbiased analysis of the literature. The protocol outlines the research questions, inclusion and exclusion criteria, and the overall approach to be followed throughout the review process.

Research questions: The research questions for this systematic review are formulated as follows:

•What are the different AI algorithms and techniques utilized in cancer diagnosis?
• How effective are AI algorithms in improving diagnostic accuracy in medical imaging and pathology?
• What are the challenges and limitations of AI in cancer diagnosis?
• What are the potential future directions and recommendations for the implementation of AI in clinical practice?

Search strategy: A systematic search of relevant literature was conducted in multiple electronic databases, including PubMed/ MEDLINE, Scopus, Embase, and Web of Science. The search strategy included a combination of keywords and Medical Subject Headings (MeSH) terms related to artificial intelligence, cancer diagnosis, medical imaging, pathology, machine learning, and deep learning. The search terms included variations and synonyms of the following concepts: Artificial intelligence, machine learning, deep learning, cancer diagnosis, medical imaging, pathology, and their respective subdomains. Boolean operators (AND, OR) were used to combine the search terms appropriately [9].

Study selection: The study selection process involved two stages: Screening based on title and abstract, followed by a full-text assessment. The author screened the identified studies based on the predefined inclusion and exclusion criteria [10].

Inclusion criteria:

• Studies focused on the application of AI algorithms in cancer diagnosis.
• Studies evaluating AI techniques in medical imaging or pathology.
• Original research articles, systematic reviews, and metaanalyses.
• Studies reporting diagnostic performance metrics and evaluation of AI algorithms.

Quality assessment: The selected studies underwent a quality assessment to evaluate their methodological rigor, study design, and risk of bias. The quality assessment was performed using established criteria, such as the Newcastle-Ottawa scale for observational studies and the cochrane collaboration's tool for randomized controlled trials [11].

Data extraction: A predefined data extraction form was used to extract relevant information from the included studies. The following data elements were analyzed for the literature review [12].

•Study characteristics: Authors, year of publication, study design.
•Participant characteristics: Sample size, demographic information.
•AI algorithms used: Machine learning, deep learning, specific techniques.
•Diagnostic modalities: Medical imaging, pathology, other modalities.
• Performance metrics: Sensitivity, specificity, accuracy, area under the curve etc.
•Key findings: Results, limitations, and implications.

Data synthesis and analysis: A narrative synthesis approach was employed to summarize the findings from the included studies. The results were organized according to the different AI techniques utilized, diagnostic modalities, cancer types, and key outcomes. The findings were then critically evaluated and discussed in the context of the research questions [13,14].

Discussion

This section presents a detailed summary of the findings obtained through the systematic review, addressing the research questions and highlighting the key trends, performance metrics, and limitations observed in the included studies. Supplementary Table 1 shows the systematic review of the Artificial Intelligence in cancer detection [15].

The application of Artificial Intelligence (AI) in cancer detection holds great promise for improving diagnostic accuracy, patient outcomes, and overall cancer management. Through the systematic review of the literature, we have observed significant advancements in the field, with AI algorithms demonstrating comparable or even superior performance to human experts in various cancer types, including breast, lung, skin, and others. AI techniques, such as machine learning and deep learning, have shown the ability to analyze complex medical data, including medical images, pathology slides, genomic data, and electronic health records, to detect cancer at early stages, differentiate between benign and malignant lesions, and predict treatment response. This has the potential to lead to more personalized and targeted cancer care, reducing unnecessary interventions and improving patient quality of life.

However, several challenges and limitations must be addressed for successful integration of AI in cancer detection into clinical practice. These include issues related to data quality, interpretability of AI algorithms, algorithm robustness, ethical considerations, regulatory guidelines, and legal implications. Collaboration between clinicians, researchers, and AI experts is crucial to overcome these challenges and ensure the responsible and effective implementation of AI in cancer detection.

Conclusion

Future research directions in the field of AI in cancer detection are numerous and diverse. Firstly, efforts should focus on developing standardized and high-quality datasets for training and evaluating AI algorithms, enabling robust performance comparisons across different studies and institutions. Additionally, research should be conducted to enhance the interpretability and transparency of AI algorithms, enabling clinicians to understand and trust the decisions made by AI models. Moreover, there is a need to explore the integration of AI with other diagnostic modalities, such as liquid biopsies and molecular imaging, to improve cancer detection sensitivity and specificity. Furthermore, studies should investigate the clinical impact of AI in cancer detection by conducting large-scale clinical trials and evaluating the cost-effectiveness of AI-based approaches.

References

1. Esteva A, Kuprel B, Novoa RA, Ko J, Swetter SM, Blau HM, et al. Dermatologist-level classification of skin cancer with deep neural networks. Nature. 2017;542(7639):115-118.

   [Crossref] [Google Scholar] [PubMed]

2. Liu Y, Gadepalli K, Norouzi M, Dahl GE, Kohlberger T, Boyko A, et al. Detecting cancer metastases on gigapixel pathology images. arXiv preprint arXiv:1703.02442. 2017.

   [Crossref] [Google Scholar]

3. Kermany DS, Goldbaum M, Cai W, Valentim CC, Liang H, Baxter SL, et al. Identifying medical diagnoses and treatable diseases by image-based deep learning. Cell. 2018;172(5):1122-1131.

   [Crossref] [Google Scholar] [PubMed]

4. Komura D, Ishikawa S. Machine learning methods for histopathological image analysis. Comput Struct Biotechnol J. 2018;16:34-42.

   [Crossref] [Google Scholar] [PubMed]

5. Litjens G, Kooi T, Bejnordi BE, Setio AA, Ciompi F, Ghafoorian M, et al. A survey on deep learning in medical image analysis. Med Image Anal. 2017;42:60-88.

   [Crossref] [Google Scholar] [PubMed]

6. Chen H, Zhang Y, Zhang W, Liao P, Huang Y, Li K. Applying artificial intelligence to breast cancer diagnosis and treatment. Front Oncol. 2020;10:1152.
7. Bejnordi BE, Veta M, van Diest PJ, van Ginneken B, Karssemeijer N, Litjens G, et al. Diagnostic assessment of deep learning algorithms for detection of lymph node metastases in women with breast cancer. JAMA. 2017;318(22):2199-2210.

   [Crossref] [Google Scholar] [PubMed]

8. Campanella G, Hanna MG, Geneslaw L, Miraflor A, Silva WKV, Busam KJ, et al. Clinical-grade computational pathology using weakly supervised deep learning on whole slide images. Nat Med. 2019;25(8):1301-1309.

   [Crossref] [Google Scholar] [PubMed]

9. Wang S, Yang DM, Rong R, Zhan X, Fujimoto J, Liu H, et al. Artificial intelligence in lung cancer pathology image analysis. Cancers (Basel). 2019;11(11):1673.

   [Crossref] [Google Scholar] [PubMed]

10. Liu Y, Kohlberger T, Norouzi M, Dahl GE, Smith JL, Mohtashamian A, et al. Artificial intelligence-based breast cancer nodal metastasis detection: Insights into the black box for pathologists. Archives of pathology and laboratory medicine. 2019;143(7):859-868.

    [Crossref] [Google Scholar] [PubMed]

11. Liu X, Faes L, Kale AU, Wagner SK, Fu DJ, Bruynseels A, et al. A comparison of deep learning performance against health-care professionals in detecting diseases from medical imaging: A systematic review and meta-analysis. Lancet Digit Health. 2019;1(6):e271-e297.

    [Crossref] [Google Scholar] [PubMed]

12. Xu J, Luo X, Wang G, Gilmore H, Madabhushi A. A deep convolutional neural network for segmenting and classifying epithelial and stromal regions in histopathological images. Neurocomputing (Amst). 2016;191:214-223.

    [Crossref] [Google Scholar] [PubMed]

13. He K, Zhang X, Ren S, Sun J. Deep residual learning for image recognition. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 2016:770-778.

    [Google Scholar]

14. Kourou K, Exarchos TP, Exarchos KP, Karamouzis MV, Fotiadis DI. Machine learning applications in cancer prognosis and prediction. Comput Struct Biotechnol J. 2015;13:8-17.

    [Crossref] [Google Scholar] [PubMed]

15. Shen D, Wu G, Suk HI. Deep learning in medical image analysis. Annu Rev Biomed Eng. 2017;19:221-248.

    [Crossref] [Google Scholar] [PubMed]