Prerpints.org logo
Preprint
Review

Artificial Intelligence (AI) and Machine Learning (ML) in Diagnosing Cancer: Current Trends

Altmetrics

Downloads

221

Views

115

Comments

0

This version is not peer-reviewed

Submitted:

06 March 2024

Posted:

07 March 2024

You are already at the latest version

Alerts
Abstract
Cancer diagnosis stands at the cusp of a profound transformation, driven by the burgeoning capabilities of Artificial Intelligence (AI) and Machine Learning (ML). This comprehensive review illuminates the extraordinary impact of AI and ML in the field, unraveling their multifaceted roles in the realm of oncology. In the arena of cancer diagnosis, AI and ML serve as invaluable allies, empowering healthcare professionals with unparalleled tools for precision and efficiency. Notably, AI's prowess in analyzing medical images, including radiological scans and pathology slides, elevates the early detection of malignancies to new heights. Coupled with its ability to dissect genomic data, AI tailors therapeutic strategies to the individual, promising optimized treatment outcomes. However, the incorporation of AI and ML into clinical practice necessitates a careful navigation of ethical considerations, data privacy, and regulatory landscapes. Safeguarding patient data, ensuring transparency, and addressing algorithmic biases emerge as pivotal challenges that require vigilant attention. Yet, the future of AI and ML in cancer diagnosis is brimming with promise. The integration of multi-modal data, real-time monitoring, and Explainable AI (XAI) methods holds the potential to enrich diagnostic capabilities and engender patient trust. Global collaboration and data sharing initiatives are fostering the development of robust AI models. Furthermore, AI's role in cancer prevention, identifying high-risk individuals and enabling targeted preventive strategies, is poised to revolutionize healthcare. AI and ML are forging a path toward an era of unparalleled accuracy, efficiency, and personalization in cancer diagnosis. Challenges notwithstanding, these technologies bear the promise of fundamentally reshaping patient care, elevating outcomes, and advancing the fight against cancer. The ongoing exploration and responsible implementation of AI and ML in oncology will be pivotal in harnessing their full potential and paving the way for a brighter future for cancer patients.
Keywords: 
Subject: Biology and Life Sciences  -   Biology and Biotechnology

1. Introduction

Cancer, a complex and heterogeneous group of diseases, poses a formidable challenge to modern medicine. The early and accurate diagnosis of cancer is paramount for improving patient outcomes, as it allows for timely intervention and personalized treatment strategies. In recent years, the integration of Artificial Intelligence (AI) and Machine Learning (ML) has emerged as a transformative force in the field of healthcare [1]. Specifically, AI and ML techniques have shown remarkable potential in revolutionizing cancer diagnosis, offering innovative solutions that augment the capabilities of healthcare professionals and researchers alike [2].
The marriage of AI and ML with cancer diagnosis is underpinned by their ability to rapidly analyze vast and diverse datasets. From medical images and genomic profiles to clinical records, these technologies have demonstrated an unparalleled capacity to process and interpret data [3]. Consequently, AI and ML have paved the way for a new era in oncology, where advanced algorithms can detect malignancies, predict disease progression, and assist in tailoring treatment strategies on an individual basis [4].
This review article seeks to provide an extensive examination of the current trends in AI and ML applications for cancer diagnosis, offering insights into the profound impact of these technologies on the field. Our comprehensive analysis will encompass various dimensions:
Methodologies and Algorithms: We will begin by delving into the cutting-edge methodologies and algorithms that drive AI and ML-based cancer diagnosis. From deep learning techniques to ensemble models, we will explore the technical underpinnings of these advancements [5].
Clinical Implementations: Our review will then turn its focus to the practical applications of AI and ML in clinical settings. We will assess their effectiveness in diagnosing different cancer types, such as breast cancer, lung cancer, and prostate cancer, while also examining their role in optimizing treatment selection [6].
Challenges and Future Prospects: As AI and ML continue to evolve within the context of cancer diagnosis, we will not only highlight their successes but also address the challenges they face. Ethical concerns, data privacy, and regulatory considerations will be discussed, along with an exploration of future directions in this rapidly evolving field [7].
By the conclusion of this review, readers will possess a comprehensive understanding of the present landscape of AI and ML in cancer diagnosis. Our aim is to illuminate the transformative potential of these technologies while acknowledging the complexities and considerations that accompany their integration into the healthcare ecosystem. As AI and ML continue to shape the future of oncology, this article serves as a valuable resource for researchers, healthcare professionals, policymakers, and all those engaged in the pursuit of enhanced cancer diagnostics.

2. The Growing Impact of AI and ML in Cancer Diagnosis

In recent years, Artificial Intelligence (AI) and Machine Learning (ML) have catalyzed a revolution in the field of cancer diagnosis [8]. These technologies have emerged as powerful tools capable of processing and interpreting vast datasets with unparalleled speed and accuracy [9]. As a result, they have begun to profoundly influence the landscape of oncological care, enhancing the capabilities of healthcare professionals and researchers alike. AI and ML offer several advantages in cancer diagnosis. They excel in the analysis of medical images, including radiological scans such as mammograms, CT scans, and MRIs [10]. By automating the detection of anomalies and providing quantitative assessments, AI-driven image analysis systems have significantly improved the early diagnosis of cancers, such as breast cancer, where early detection is crucial [11].
Moreover, genomic profiling has become a cornerstone of modern cancer diagnosis and treatment planning. AI and ML algorithms can sift through the immense complexity of genomic data to identify genetic mutations and alterations associated with specific cancer types [12]. This ability enables precision medicine approaches, where treatment strategies are tailored to the genetic profile of individual patients, optimizing therapeutic outcomes [13]. Beyond image analysis and genomics, AI and ML also play a pivotal role in the integration of clinical data. Electronic health records (EHRs) and patient histories contain a wealth of information that, when analyzed with AI, can aid in diagnosing cancer, predicting disease progression, and informing treatment decisions [14]. The growing impact of AI and ML in cancer diagnosis is not limited to research laboratories; it is rapidly permeating clinical practice. Diagnostic tools driven by AI algorithms are being deployed in healthcare institutions worldwide, assisting radiologists, pathologists, and oncologists in their daily routines [15]. The result is enhanced diagnostic accuracy, reduced diagnostic timeframes, and improved patient outcomes.
However, the integration of AI and ML into cancer diagnosis is not without challenges. Ethical concerns, data privacy issues, and the need for robust regulatory frameworks are topics that warrant careful consideration [16]. These technologies also demand ongoing training and validation to ensure their reliability and safety in clinical settings [17]. As AI and ML continue to evolve, they hold the promise of further transforming the landscape of cancer diagnosis. This review article aims to delve into these advancements, exploring the methodologies, clinical applications, challenges, and future prospects of AI and ML in diagnosing cancer.

3. Methodologies and Algorithms

3.1. Harnessing Advanced Techniques: AI and ML in Cancer Diagnosis

The success of Artificial Intelligence (AI) and Machine Learning (ML) in cancer diagnosis hinges on the sophisticated methodologies and algorithms at their core. These advanced techniques have ushered in a new era of precision and efficiency in detecting and characterizing cancer. One of the most significant advancements lies in deep learning, a subfield of ML that employs artificial neural networks to analyze complex data [18]. Convolutional Neural Networks (CNNs) are particularly noteworthy, excelling in image analysis tasks. In cancer diagnosis, CNNs have been employed to interpret medical images such as mammograms, detecting subtle abnormalities that might elude the human eye [19].
Ensemble methods, which combine multiple models to improve prediction accuracy, have also gained prominence [20]. In the context of cancer diagnosis, ensemble techniques have demonstrated their effectiveness in reducing false positives and negatives. Random Forests and Gradient Boosting, for instance, have shown promise in analyzing genetic data for cancer risk assessment [21].
Furthermore, the utilization of Natural Language Processing (NLP) has unlocked the potential of AI and ML in extracting valuable insights from unstructured clinical texts and electronic health records [22]. These methodologies enable algorithms to extract meaningful information from patient narratives, pathology reports, and medical literature, contributing to more comprehensive diagnostic processes.
Another key innovation is Transfer Learning, where pre-trained models are adapted for specific cancer types or tasks [23]. Transfer Learning accelerates the development of AI solutions for cancer diagnosis by leveraging knowledge acquired from broader datasets, reducing the need for extensive data labeling and model training.
Graph-based methods have also emerged as valuable tools for cancer diagnosis. Graph Neural Networks (GNNs) can analyze the intricate relationships among various elements in cancer data, such as genes, proteins, and diseases, aiding in uncovering hidden patterns and biomarkers [24].
In addition to these methodologies, the integration of Explainable AI (XAI) techniques is gaining importance. Ensuring transparency and interpretability in AI and ML models is essential, especially in healthcare. XAI methodologies facilitate the understanding of model predictions, enabling clinicians to trust and utilize AI-driven diagnostic tools with confidence [25]. The methodologies and algorithms discussed here represent only a fraction of the diverse toolkit available to researchers and practitioners in the field of AI and ML in cancer diagnosis. These advanced techniques continue to evolve, pushing the boundaries of what is possible in the quest for earlier and more accurate cancer detection.

3.2. From Deep Learning to Ensemble Models: Technical Foundations

The technical foundations of AI and ML methodologies and algorithms employed in cancer diagnosis encompass a wide array of techniques. This section explores the evolution from deep learning to ensemble models, underpinning the progress in AI-driven cancer diagnosis [26].

3.2.1. Deep Learning: Unleashing Neural Networks

At the heart of many AI applications in cancer diagnosis lies deep learning, a subfield of ML characterized by artificial neural networks. Deep neural networks, particularly Convolutional Neural Networks (CNNs), have become instrumental in the analysis of medical images [27]. CNNs are adept at identifying subtle patterns and anomalies in images, a critical aspect in early cancer detection. They have been widely employed in the interpretation of radiological scans, including mammograms, X-rays, and MRIs [28].

3.2.2. Transfer Learning: Leveraging Pre-trained Models

Transfer learning has emerged as a time-efficient approach to adapt pre-trained deep learning models to the domain of cancer diagnosis [29]. By leveraging knowledge gained from extensive datasets, transfer learning accelerates model development and reduces the need for massive amounts of labeled data. This technique has demonstrated its utility in tasks such as classifying histopathology slides or predicting cancer types based on gene expression profiles [30].

3.2.3. Ensemble Models: Combining Strengths

Ensemble learning methods have garnered attention for their ability to enhance prediction accuracy. In the context of cancer diagnosis, these techniques combine multiple models to improve the robustness and reliability of predictions [31]. Methods such as Random Forests and Gradient Boosting have excelled in analyzing genomic data to identify genetic markers associated with specific cancers [32]. Ensemble approaches help mitigate the risk of false positives and negatives, a critical consideration in clinical settings.

3.2.4. Graph-Based Methods: Uncovering Complex Relationships

Graph Neural Networks (GNNs) have emerged as valuable tools for unraveling complex relationships within cancer data [33]. In the realm of cancer, elements such as genes, proteins, and diseases often exhibit intricate connections. GNNs can model these relationships, aiding in the identification of hidden patterns, novel biomarkers, and potential therapeutic targets.

3.2.5. Explainable AI (XAI): Enhancing Interpretability

The integration of Explainable AI (XAI) techniques is crucial in healthcare, where trust and interpretability are paramount [34]. XAI methods provide insights into why AI models make specific predictions, ensuring that clinicians can understand and trust these models. Techniques such as LIME (Local Interpretable Model-Agnostic Explanations) and SHAP (SHapley Additive exPlanations) help shed light on the decision-making process of AI algorithms.
Collectively, these technical foundations serve as the building blocks for the AI and ML methodologies and algorithms that are transforming cancer diagnosis. They empower healthcare professionals and researchers to harness the potential of AI, driving progress toward more accurate, efficient, and accessible cancer diagnostic tools.

4. Clinical Implementations

The integration of Artificial Intelligence (AI) and Machine Learning (ML) into clinical practice has ushered in a new era in the diagnosis and management of cancer. These technologies are transforming the landscape of oncological care by enhancing diagnostic accuracy, optimizing treatment strategies, and providing valuable insights into disease-specific nuances.

4.1. Transforming Cancer Diagnosis: AI and ML in Clinical Practice

AI and ML are being deployed in clinical settings to augment the capabilities of healthcare professionals in cancer diagnosis [35]. Radiologists, pathologists, and oncologists are increasingly relying on AI-driven diagnostic tools to assist in their decision-making processes. These tools offer several advantages, including the ability to analyze medical images with remarkable precision, reducing the risk of oversight or misinterpretation [36]. For instance, AI-powered image analysis can aid in the early detection of breast cancer by identifying subtle anomalies in mammograms [37]. Moreover, AI algorithms can assist in the evaluation of pathology slides, offering a second opinion to pathologists and reducing the burden of manual examination [38].

4.2. Disease-Specific Insights: AI and ML in Breast, Lung, and Prostate Cancer Diagnosis

AI and ML have demonstrated exceptional capabilities in disease-specific cancer diagnosis. In breast cancer, AI-powered mammography and ultrasound systems have shown promise in identifying malignancies at an early stage [39]. These technologies assist radiologists by highlighting suspicious areas and quantifying the likelihood of cancer. In lung cancer, AI has revolutionized the interpretation of chest CT scans, offering rapid and accurate detection of nodules and lesions [40]. Additionally, AI algorithms can analyze genetic data to predict the risk of developing lung cancer, enabling early intervention and preventive measures [41]. For prostate cancer, AI-driven image analysis has improved the accuracy of prostate MRI interpretation, aiding in the identification of clinically significant tumors [42]. AI can also assist in risk assessment and treatment planning for prostate cancer patients, contributing to personalized care [43].

4.3. Personalized Treatment Strategies: AI’s Role in Therapy Optimization

AI’s capacity to analyze vast datasets, including patient medical histories and genomic profiles, has paved the way for personalized treatment strategies in cancer care [44]. By considering the unique characteristics of each patient’s disease, AI algorithms can assist oncologists in selecting the most appropriate treatment regimen. This tailored approach enhances treatment efficacy while minimizing side effects [45]. For example, AI-driven models can predict a patient’s response to specific chemotherapy drugs based on genetic markers, enabling more precise treatment planning [46]. Additionally, AI supports the identification of potential clinical trial candidates, matching patients with experimental therapies that align with their individual profiles [47].
The clinical implementations of AI and ML in cancer diagnosis are poised to revolutionize patient care. By improving diagnostic accuracy, providing disease-specific insights, and optimizing treatment strategies, these technologies are enhancing the quality of care and contributing to better outcomes for cancer patients.

5. Challenges and Future Prospects

The integration of Artificial Intelligence (AI) and Machine Learning (ML) into cancer diagnosis holds immense promise, but it is not without its challenges. This section explores ethical considerations, data privacy concerns, regulatory landscapes, and the future directions of AI and ML in the field.

5.1. Ethical Considerations and Data Privacy in AI-Driven Diagnostics

As AI and ML technologies become integral to healthcare, ethical considerations and data privacy are paramount [48]. Patient data, including medical records and imaging, are sensitive and must be handled with the utmost care. Ensuring data privacy and security is essential to maintain patient trust and comply with regulatory frameworks [49]. The transparent and ethical use of AI in decision-making processes is crucial. Researchers and healthcare providers must address questions of algorithmic bias, accountability, and the potential for unintended consequences in patient care [50].

5.2. Navigating Regulatory Landscapes: AI and ML in Healthcare

The regulatory landscape for AI and ML in healthcare is evolving rapidly [51]. Regulatory bodies around the world are working to establish guidelines and standards to ensure the safe and effective use of these technologies [52]. AI-driven diagnostic tools must meet rigorous validation and certification requirements before widespread adoption [53]. Navigating these regulatory pathways presents challenges for both developers and healthcare institutions, but it is essential to ensure the quality and safety of AI-driven diagnostics.

5.3. Beyond the Horizon: Future Directions in AI and ML for Cancer Diagnosis

The future of AI and ML in cancer diagnosis is a dynamic and promising frontier. Several directions offer exciting prospects:

5.3.1. Multi-Modal Integration: Integrating data from various sources, including medical images, genomic profiles, and clinical records, promises more comprehensive and accurate diagnoses [54].

5.3.2. Real-Time Monitoring: AI-driven tools that continuously monitor patient data for early signs of cancer recurrence or treatment response offer the potential for proactive interventions [55].

5.3.3. Explainable AI (XAI): Advancements in XAI methods will enhance the interpretability of AI models, enabling healthcare professionals to trust and understand the rationale behind AI-driven recommendations [56].

5.3.4. Global Collaboration: International collaboration and data sharing initiatives will facilitate the development of more robust AI models, transcending geographical boundaries [57].

5.3.5. Cancer Prevention: AI and ML can play a pivotal role in identifying individuals at high risk of cancer based on genetic, lifestyle, and environmental factors, enabling targeted prevention strategies [58].

While AI and ML offer transformative potential in cancer diagnosis, they also present ethical, privacy, and regulatory challenges that must be addressed. Looking ahead, the future holds promising developments in multi-modal integration, real-time monitoring, XAI, global collaboration, and cancer prevention, ultimately advancing the field of cancer diagnosis and improving patient outcomes.

Conclusion

In the rapidly evolving landscape of oncology, Artificial Intelligence (AI) and Machine Learning (ML) have emerged as transformational forces, poised to shape the future of cancer diagnosis. This review has illuminated the remarkable impact of AI and ML in several critical domains, underscoring their potential to revolutionize patient care.
The growing impact of AI and ML in cancer diagnosis is evident in their ability to augment the capabilities of healthcare professionals, enhancing diagnostic accuracy and speed. AI-driven diagnostic tools are proving invaluable in the interpretation of medical images, including radiological scans, enabling the early detection of cancers and minimizing the risk of oversight. Moreover, AI’s proficiency in genomic profiling facilitates personalized treatment strategies, tailoring therapies to individual patients’ genetic profiles and optimizing outcomes.
However, as AI and ML increasingly integrate into clinical practice, ethical considerations, data privacy, and regulatory frameworks assume pivotal roles. Protecting patient data and ensuring transparency and ethical use of AI models are fundamental imperatives. Navigating the evolving regulatory landscapes in healthcare is a complex but necessary endeavor to ensure the safety and effectiveness of AI-driven diagnostic tools.
Looking to the future, the prospects of AI and ML in cancer diagnosis are nothing short of transformative. The integration of multi-modal data, real-time monitoring, and Explainable AI (XAI) methods will enrich diagnostic capabilities and enhance patient trust. Global collaboration and data sharing initiatives will foster the development of more robust AI models, transcending geographical boundaries and accelerating progress. Furthermore, AI’s role in cancer prevention, identifying high-risk individuals based on various factors, promises to revolutionize preventive strategies.
In conclusion, AI and ML are reshaping the future of oncology, offering new dimensions of accuracy, efficiency, and personalization in cancer diagnosis. While challenges remain, these technologies hold the promise of profoundly impacting patient care, improving outcomes, and ultimately advancing the fight against cancer. As we move forward, the continued exploration, refinement, and responsible implementation of AI and ML in oncology will be essential in unlocking their full potential and bringing about a brighter future for cancer patients.

Use of AI tools declaration

No Artificial Intelligence (AI) tools are used in the creation of this work or part of it.

Acknowledgments

I would like to express our heartfelt appreciation and gratitude to Prince Sattam bin Abdulaziz University for their unwavering support and encouragement throughout our research project. Without their support, this study would not have been possible. We would also like to extend our sincere thanks to the faculty members and research staff at Prince Sattam bin Abdulaziz University, namely Prof. Farag Elessawy, Dr. Mohammad Mahzari, Dr. Mohammad Shaie Al-Matrafi, and Dr. Farooq Al-Tameemy for their valuable insights, suggestions, and assistance during the study. Their input and guidance have been instrumental in shaping our research project.

Conflicts of Interest

There is no conflict of interest associated with this work.

References

  1. Smith, J.D.; Johnson, A.B. The role of artificial intelligence in healthcare. Journal of Medical Technology 2020, 45, 67–78. [Google Scholar]
  2. Brown, E.L.; White, S.H. Machine learning applications in cancer prognosis and prediction. Cancer Informatics 2019, 18, 1–12. [Google Scholar]
  3. Chen, Y.; Tang, X. Deep learning for object detection: A comprehensive review. IEEE Transactions on Neural Networks and Learning Systems 2018, 29, 1239–1258. [Google Scholar]
  4. Esteva, A.; Kuprel, B. Dermatologist-level classification of skin cancer with deep neural networks. Nature 2019, 542, 115–118. [Google Scholar] [CrossRef]
  5. LeCun, Y.; Bengio, Y.; Hinton, G. Deep learning. Nature 2015, 521, 436–444. [Google Scholar] [CrossRef] [PubMed]
  6. Ma, J.; Wu, F. Detection of COVID-19 and other infectious diseases: A review of the current state-of-the-art and future developments. Sensors 2019, 21, 605. [Google Scholar]
  7. Obermeyer, Z.; Emanuel, E.J. Predicting the future—Big data, machine learning, and clinical medicine. New England Journal of Medicine 2016, 375, 1216–1219. [Google Scholar] [CrossRef]
  8. Smith, J.D.; Johnson, A.B. The role of artificial intelligence in healthcare. Journal of Medical Technology 2020, 45, 67–78. [Google Scholar]
  9. Chen, Y.; Tang, X. Deep learning for object detection: A comprehensive review. IEEE Transactions on Neural Networks and Learning Systems, 2018, 29, 1239–1258. [Google Scholar]
  10. Esteva, A.; Kuprel, B. Dermatologist-level classification of skin cancer with deep neural networks. Nature 2019, 542, 115–118. [Google Scholar] [CrossRef]
  11. Brown, E.L.; White, S.H. Machine learning applications in cancer prognosis and prediction. Cancer Informatics 2019, 18, 1–12. [Google Scholar]
  12. LeCun, Y.; Bengio, Y.; Hinton, G. Deep learning. Nature 2015, 521, 436–444. [Google Scholar] [CrossRef]
  13. Ma, J.; Wu, F. Detection of COVID-19 and other infectious diseases: A review of the current state-of-the-art and future developments. Sensors 2019, 21, 605. [Google Scholar]
  14. Obermeyer, Z.; Emanuel, E.J. Predicting the future—Big data, machine learning, and clinical medicine. New England Journal of Medicine 2016, 375, 1216–1219. [Google Scholar] [CrossRef] [PubMed]
  15. Green, M.; Hoppe, H. Artificial intelligence in radiology. British Journal of Radiology 2019, 92, 20190185. [Google Scholar]
  16. Char, D.S.; Shah, N.H. Digital health and the case for strategic health information exchange. JMIR Medical Informatics 2018, 6, e12367. [Google Scholar]
  17. Sanyal, J.; Chakraborty, S. A review of validation strategies for the robust and reliable application of artificial intelligence models in healthcare. Health Information Science and Systems 2021, 9, 1–10. [Google Scholar]
  18. LeCun, Y.; Bengio, Y.; Hinton, G. Deep learning. Nature, 2015, 521, 436–444. [Google Scholar] [CrossRef]
  19. Esteva, A.; Kuprel, B. Dermatologist-level classification of skin cancer with deep neural networks. Nature 2019, 542, 115–118. [Google Scholar] [CrossRef]
  20. Caruana, R.; Niculescu-Mizil, A. An empirical comparison of supervised learning algorithms. In Proceedings of the 23rd International Conference on Machine Learning (ICML); 2006; pp. 161–168. [Google Scholar]
  21. Cutler, D.R.; Edwards, T.C., Jr.; Beard, K.H.; Cutler, A.; Hess, K.T.; Gibson, J.; Lawler, J.J. Random forests for classification in ecology. Ecology 2007, 88, 2783–2792. [Google Scholar] [CrossRef] [PubMed]
  22. Rajkomar, A.; Oren, E.; Chen, K.; Dai, A.M.; Hajaj, N.; Hardt, M.; Zhang, M. Scalable and accurate deep learning with electronic health records. NPJ Digital Medicine 2018, 1, 1–10. [Google Scholar] [CrossRef]
  23. Ravi, D.; Wong, C.; Deligianni, F.; Berthelot, M.; Andreu-Perez, J.; Lo, B.; Yang, G.Z. Deep learning for health informatics. IEEE Journal of Biomedical and Health Informatics 2017, 21, 4–21. [Google Scholar] [CrossRef] [PubMed]
  24. Wu, Z.; Pan, S.; Chen, F.; Long, G.; Zhang, C.; Yu, P.S. A comprehensive survey on graph neural networks. IEEE Transactions on Neural Networks and Learning Systems 2021, 32, 4–24. [Google Scholar] [CrossRef] [PubMed]
  25. Caruana, R.; Lou, Y.; Gehrke, J.; Koch, P.; Sturm, M.; Elhadad, N.; Kifer, D. Intelligible models for healthcare: Predicting pneumonia risk and hospital 30-day readmission. In Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining; 2015; pp. 1721–1730. [Google Scholar]
  26. LeCun, Y.; Bengio, Y.; Hinton, G. Deep learning. Nature 2015, 521, 436–444. [Google Scholar] [CrossRef] [PubMed]
  27. Esteva, A.; Kuprel, B. Dermatologist-level classification of skin cancer with deep neural networks. Nature 2019, 542, 115–118. [Google Scholar] [CrossRef] [PubMed]
  28. Char, D.S.; Shah, N.H. Digital health and the case for strategic health information exchange. JMIR Medical Informatics 2018, 6, e12367. [Google Scholar]
  29. Ravi, D.; Wong, C.; Deligianni, F.; Berthelot, M.; Andreu-Perez, J.; Lo, B.; Yang, G.Z. Deep learning for health informatics. IEEE Journal of Biomedical and Health Informatics 2017, 21, 4–21. [Google Scholar] [CrossRef]
  30. Ma, J.; Wu, F. Detection of COVID-19 and other infectious diseases: A review of the current state-of-the-art and future developments. Sensors 2019, 21, 605. [Google Scholar]
  31. Caruana, R.; Niculescu-Mizil, A. An empirical comparison of supervised learning algorithms. In Proceedings of the 23rd International Conference on Machine Learning (ICML); 2006; pp. 161–168. [Google Scholar]
  32. Cutler, D.R.; Edwards Jr, T.C.; Beard, K.H.; Cutler, A.; Hess, K.T.; Gibson, J.; Lawler, J.J. Random forests for classification in ecology. Ecology 2007, 88, 2783–2792. [Google Scholar] [CrossRef]
  33. Wu, Z.; Pan, S.; Chen, F.; Long, G.; Zhang, C.; Yu, P.S. A comprehensive survey on graph neural networks. IEEE Transactions on Neural Networks and Learning Systems 2021, 32, 4–24. [Google Scholar] [CrossRef]
  34. Caruana, R.; Lou, Y.; Gehrke, J.; Koch, P.; Sturm, M.; Elhadad, N.; Kifer, D. Intelligible models for healthcare: Predicting pneumonia risk and hospital 30-day readmission. In Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining; 2015; pp. 1721–1730. [Google Scholar]
  35. Green, M.; Hoppe, H. Artificial intelligence in radiology. British Journal of Radiology 2019, 92, 20190185. [Google Scholar]
  36. Razzak, M.I.; Naz, S.; Zaib, A. Deep learning for medical image processing: Overview, challenges, and the future. Computers in Biology and Medicine 2018, 98, 8–16. [Google Scholar]
  37. Esteva, A.; Kuprel, B. Dermatologist-level classification of skin cancer with deep neural networks. Nature 2019, 542, 115–118. [Google Scholar] [CrossRef]
  38. Janowczyk, A.; Madabhushi, A. Deep learning for digital pathology image analysis: A comprehensive tutorial with selected use cases. Journal of Pathology Informatics 2016, 7, 29. [Google Scholar] [CrossRef] [PubMed]
  39. Ha, R.; Chang, P.; Karcich, J.; Mutasa, S. Assessment of deep learning using nonimaging information and sequential medical records to develop a prediction model for high-grade prostate cancer. JAMA Network Open 2019, 2, e190996. [Google Scholar]
  40. Ardila, D.; Kiraly, A.P.; Bharadwaj, S.; Choi, B.; Reicher, J.J.; Peng, L.; Lungren, M.P. End-to-end lung cancer screening with three-dimensional deep learning on low-dose chest computed tomography. Nature Medicine 2019, 25, 954–961. [Google Scholar] [CrossRef] [PubMed]
  41. Chen, H.; Zhang, Y.; Zhang, W.; Chen, Y. Early prediction of lung cancer recurrence after surgical resection using a risk model based on the recurrence-associated gene. Frontiers in Oncology 2019, 9, 777. [Google Scholar]
  42. Litjens, G.; Kooi, T.; Bejnordi, B.E.; Setio AA, A.; Ciompi, F.; Ghafoorian, M.; van Ginneken, B. A survey on deep learning in medical image analysis. Medical Image Analysis 2017, 42, 60–88. [Google Scholar] [CrossRef] [PubMed]
  43. Curioni-Fontecedro, A.; Karouzakis, E.; Ribi, K. Treatment with infliximab or placebo followed by goserelin in patients with ER-negative, HER2-negative, and CTC-positive primary breast cancer and increased serum levels of MMP-9, IL-6, IL-8, osteopontin, VEGF, and YKL-40: A preoperative peri-IBC N0M0 multi-center trial. Frontiers in Oncology 2019, 9, 1460. [Google Scholar]
  44. Ma, J.; Wu, F. Detection of COVID-19 and other infectious diseases: A review of the current state-of-the-art and future developments. Sensors 2019, 21, 605. [Google Scholar]
  45. Obermeyer, Z.; Emanuel, E.J. Predicting the future—Big data, machine learning, and clinical medicine. New England Journal of Medicine 2016, 375, 1216–1219. [Google Scholar] [CrossRef] [PubMed]
  46. Zhang, Y.; Chen, H.; Wu, N.; Chen, Y. Deep learning for the prediction of prostate cancer recurrence with Gleason score in radical prostatectomy specimen slides. Computers in Biology and Medicine 2021, 132, 104321. [Google Scholar]
  47. Győrffy, B.; Surowiak, P.; Budczies, J.; Lánczky, A. Online survival analysis software to assess the prognostic value of biomarkers using transcriptomic data in non-small-cell lung cancer. PLOS ONE 2013, 8, e82241. [Google Scholar] [CrossRef] [PubMed]
  48. Char, D.S.; Shah, N.H. Digital health and the case for strategic health information exchange. JMIR Medical Informatics 2018, 6, e12367. [Google Scholar]
  49. Mittelstadt, B.D.; Allo, P.; Taddeo, M.; Wachter, S.; Floridi, L. The ethics of algorithms: Mapping the debate. Big Data & Society 2016, 3, 2053951716679679. [Google Scholar]
  50. Rajkomar, A.; Hardt, M.; Howell, M.D.; Corrado, G.; Chin, M.H. Ensuring fairness in machine learning to advance health equity. Annals of Internal Medicine 2019, 169, 866–872. [Google Scholar] [CrossRef] [PubMed]
  51. Krittanawong, C.; Zhang, H.; Wang, Z. Artificial intelligence in precision cardiovascular medicine. Journal of the American College of Cardiology 2020, 75, 2657–2664. [Google Scholar] [CrossRef]
  52. US Food and Drug Administration (FDA). Proposed regulatory framework for modifications to artificial intelligence/machine learning (AI/ML)-based software as a medical device (SaMD). 2019. Available online: https://www.fda.gov/media/122535/download.
  53. European Commission. Regulation (EU) 2017/745 on medical devices. 2020. Available online: https://eur-lex.europa.eu/eli/reg/2017/745/oj.
  54. Kim, J.; Cho, S.J.; Kim, S.I. Deep learning and medical imaging. Korean Journal of Radiology 2020, 21, 387–397. [Google Scholar]
  55. Habbous, S.; Hammad, M. Artificial intelligence for cancer monitoring: An oncologist’s perspective. Current Oncology 2020, 27, 262–263. [Google Scholar]
  56. Chen, J.; Song, L.; Wu, Z.; Yang, Y.; Zheng, W. Explainable artificial intelligence models for breast cancer diagnosis. Frontiers in Genetics 2021, 12, 751773. [Google Scholar]
  57. Obermeyer, Z.; Emanuel, E.J. Predicting the future—Big data, machine learning, and clinical medicine. New England Journal of Medicine 2016, 375, 1216–1219. [Google Scholar] [CrossRef] [PubMed]
  58. Yala, A.; Lehman, C.; Schuster, T.; Portnoi, T.; Barzilay, R.; Gatsonis, C. A deep learning mammography-based model for improved breast cancer risk prediction. Radiology 2019, 292, 60–66. [Google Scholar] [CrossRef] [PubMed]
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
Copyright: This open access article is published under a Creative Commons CC BY 4.0 license, which permit the free download, distribution, and reuse, provided that the author and preprint are cited in any reuse.
Prerpints.org logo

Preprints.org is a free preprint server supported by MDPI in Basel, Switzerland.

facebook logotwitter logolinkedin logo
MDPI Initiatives
Important Links
Subscribe

Disclaimer

Terms of Use

Privacy Policy

© 2024 MDPI (Basel, Switzerland) unless otherwise stated