Artificial Intelligence in Cancer Biomedicine: A Comprehensive Survey of Diagnostic, Treatment, and Research Application

Introduction and Context

Artificial intelligence (AI) has increasingly become a fundamental element of contemporary medical research and clinical practice. Its implementation in various medical specialties, including ophthalmology, radiology, and dermatology, has proven to improve diagnostic precision, optimize processes, and facilitate clinical decision-making [1]. The incorporation of AI into biology, pharmacology, and medicine has yielded substantial advancements, with some AI-driven models currently achieving performance levels akin to seasoned human specialists [2].

Initial AI systems were predominantly rule-based, depending on symbolic reasoning and expert knowledge repositories. The advent of machine learning (ML) revolutionized this domain, enabling computers to acquire knowledge from data and enhance their predictive capabilities over time [3]. Deep learning (DL), a subset of machine learning, utilizes neural network designs to identify intricate patterns in extensive datasets, tackling issues that conventional ML methods frequently cannot [3].

The human brain's capacity to comprehend information is limited, whereas AI's ability to rapidly evaluate extensive datasets has established it as a valuable asset for improving medical intelligence [4]. Cancer, distinguished by many genetic and epigenetic modifications, constitutes one of the most intricate diseases in contemporary medicine. Artificial intelligence has novel prospects for the early identification of genetic mutations, anomalous protein interactions, and patterns of disease progression. Furthermore, there has been a growing emphasis on the ethical and secure integration of AI technologies within therapeutic settings [5][6].

AI-driven systems can assist doctors in illness risk evaluation, diagnosis, prognosis, and treatment formulation, hence, enhancing precision medicine. AI-powered systems facilitate real-time data exchange and collaborative research, enabling scientists and healthcare practitioners to expedite discoveries and enhance patient outcomes [4][7].

The availability of open-source healthcare data has accelerated the development of AI systems for cancer detection and prognosis. Machine learning and deep learning models trained on distributed datasets provide efficient, reliable, and scalable options for cancer therapeutic research [5]. Advanced methodologies like federated learning provide secure and collaborative data analysis among universities. Moreover, significant assets in oncology are emerging from novel methodologies—such as virtual biopsies, natural language processing (NLP) for predicting patient trajectories, and blood-based cancer diagnostics employing deep sequencing [8].

Oncology has progressed from basic microscopy to sophisticated, evidence-based techniques that utilize gene expression monitoring and next-generation sequencing for diagnosis, treatment selection, and risk assessment [9]. Artificial intelligence has significantly advanced this progression by facilitating data-driven pharmaceutical development and individualized treatment approaches. With the aid of sophisticated AI infrastructures, researchers lacking significant computing expertise can participate in AI-assisted cancer research. This extensive accessibility indicates a future in which AI-driven data mining, electronic health record (EHR) analysis, and radiologic image interpretation become integral elements of clinical practice [10].

Market estimates forecast that AI integration in the U.S. healthcare system may yield savings of over $52 billion by 2021, highlighting its capacity for substantial economic and therapeutic advantages [11]. The proliferation of digital data in the past twenty years has spurred the comeback of AI, especially in cancer, where the accessibility of high-dimensional molecular data from tumors has expedited model building [12].

Notwithstanding swift advancements, artificial intelligence in oncology continues to be a developing domain. A limited number of AI-driven solutions have received regulatory approval for widespread clinical use. Debates persist concerning whether AI will augment or supplant human knowledge in medicine [13]. Most experts concur that AI is more likely to assist human judgment rather than replace it, offering data-driven insights that improve decision-making. Recent clinical trials employing AI for cancer detection, diagnosis, and treatment optimization have yielded promising results, particularly in digital pathology, radiology, and genomics [14].

This review seeks to encapsulate the fundamental ideas and contemporary advancements of AI applications in cancer research. The text examines the progression of the discipline over the past twenty years, investigates the accessible data sources and analytical methodologies, and underscores the significance of AI in radiation therapy, chemotherapy, surgery, immunotherapy, targeted therapy, and nanotechnology. Moreover, it tackles the principal obstacles in converting AI advancements from theoretical models to practical application. This review aims to draw attention to the advantages of AI in healthcare and encourage additional research for its safe and effective use in cancer treatment [15].

The Role of Artificial Intelligence in Cancer Clinical Research Over the Last Twenty Years

In the past two decades, artificial intelligence (AI) has increasingly integrated into cancer research, exhibiting significant advancements and attaining performance on par with human experts [1]. Commercial organizations and research institutions are increasingly utilizing artificial intelligence to enhance cancer detection, diagnosis, and treatment strategies [16].

In 2014, IBM achieved a significant milestone in clinical AI with the creation of Watson, a system tailored to support cancer research and clinical decision-making [17]. Watson synthesizes extensive medical literature, patient records, and treatment protocols to assist physicians in making informed judgments. Likewise, teams at Microsoft Research have investigated the utilization of machine learning (ML) and natural language processing (NLP) to model biological systems and discover possible cancer treatments [18].

These AI-driven projects seek to enhance diagnostic precision, expedite decision-making, and provide data-informed therapy suggestions. Through the analysis of intricate datasets encompassing genetic profiles, imaging results, and clinical records, AI can enhance medication choices and more accurately forecast patient outcomes [10][19].

Evaluation of Search Methodology

A methodical strategy was utilized to discern pertinent literature for this evaluation on AI applications in cancer research. The objective was to deliver an exhaustive summary of the present condition of the field, accessible datasets, analytical methodologies, and clinical applications pertaining to diverse facets of cancer research [20].

Multiple academic databases, including PubMed, Web of Science, Scopus, and Google Scholar, were queried to obtain research published from 2006 to the present, thereby incorporating the latest findings. The search technique employed targeted keywords like “Artificial Intelligence,” “Cancer Research,” “Precision Medicine,” “Diagnosis,” “Treatment,” “Radiology,” and “Nanotechnology.” This meticulously crafted keyword structure seeks to adequately encompass studies at the convergence of AI and cancer [21].

Articles, reviews, and reports were evaluated based on established inclusion and exclusion criteria. The inclusion requirements mandated that papers be authored in English, document recent advancements in AI applications for cancer research and clinical results, investigate AI integration across all phases of cancer diagnosis or treatment, and be published during the specified period from 2006 to the present. We eliminated studies that failed to match these criteria, including non-English publications or those unrelated to AI applications in oncology [22].

This meticulous search process allowed for the acquisition of a significant and current body of literature, enabling a comprehensive and precise review. The data obtained from the chosen studies were methodically examined and structured to elucidate AI's contribution to the progression of cancer research and its prospective effects on clinical practice (Figure 1) [23].    

Figure 1 - Referring to the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) standards, the PRISMA flow diagram for literature search is depicted in Figure 1.

Various AI Applications in Biomedical Cancer Research

Artificial intelligence (AI) has significantly advanced biomedical cancer research in various domains. It possesses the capacity to expedite drug development, improve early cancer detection via sophisticated image analysis, and customize treatment programs according to individual patient data. Machine learning (ML) and deep learning (DL) techniques enable researchers to analyze extensive datasets and identify nuanced patterns that can enhance early diagnosis and forecast disease progression [1][2][3][4].

Furthermore, AI is helpful in precision medicine by allowing healthcare providers to formulate treatment strategies that take into account a patient's genetic profile and other individualized elements. In drug discovery, AI reduces the duration needed to uncover potential therapeutic candidates and optimizes the preliminary stages of research. These AI-driven methodologies offer enhanced efficiency, personalization, and efficacy in cancer treatment, fundamentally altering the landscape of scientific research and patient care [5][6][7][8][9].

Artificial intelligence has significantly enhanced the key components of cancer detection and therapy, resulting in optimization and advancement in these domains [24]:

Radiodiagnosis

Contemporary medical imaging modalities increasingly utilize machine learning (ML) approaches to improve diagnostic precision, especially in computed tomography (CT) and magnetic resonance imaging (MRI) systems. CT scans offer comprehensive volumetric and structural insights into a patient's anatomy. Advanced CT scanners can do full-body scans with excellent resolution, facilitating thorough viewing of anatomical structures [14][25].

MRI is exceptionally proficient in distinguishing soft tissues, rendering it optimal for the assessment of joints, ligaments, and internal organs. MRI produces intricate images of almost any body region, contingent upon differences in soft tissue density, making it very effective for identifying malignant cells. Moreover, deep learning (DL) models and neural networks are progressively utilized in brain segmentation, providing semi-automated techniques for differentiating healthy tissue from malignant areas in MRI data [26][27].

Breast cancer constitutes around 30% of newly diagnosed malignancies in women. Ultrasonography is frequently employed for the detection of breast cancer. Enhancing the segmentation of breast ultrasound pictures can improve breast density measurements, accurately identify tumor locations, and assess therapy responses. Conventional manual segmentation by radiologists is labor-intensive and necessitates specialized knowledge. Automated approaches, particularly those using convolutional neural networks (CNNs), have demonstrated the ability to accurately classify key tissue types—such as fatty tissue, fibro-glandular tissue, skin, and masses—from 3D ultrasound images. These findings offer potential for enhancing clinical breast cancer diagnoses in the future [28][29].

Radiation Therapy

Artificial intelligence (AI) is assuming a progressively significant role in radiotherapy, enhancing both the accuracy and efficacy of cancer treatment. AI technologies are very effective in precisely defining tumor margins, enabling targeted radiation delivery to malignant cells while reducing exposure to adjacent healthy tissues. This focused strategy mitigates adverse effects and enhances overall therapy safety [10][11][12].

AI-driven algorithms can process extensive amounts of patient data and medical imaging, facilitating optimum treatment planning customized for individual patients. Furthermore, AI can facilitate real-time monitoring throughout therapy, allowing physicians to implement prompt alterations according to a patient's response. Integrating these technologies renders radiotherapy more precise, individualized, and efficient. AI is improving treatment outcomes and facilitating significant advancements in biomedical cancer research [13][14][15].

Chemotherapy

AI is progressively revolutionizing chemotherapy and biomedical cancer research by enhancing and individualizing treatment approaches. Through the analysis of extensive and intricate datasets, AI assists doctors in formulating chemotherapy regimens customized to each individual's unique genetic and molecular characteristics. This personalized strategy can improve therapeutic efficacy while minimizing the likelihood of unwanted side effects [30][31].

Furthermore, AI-driven predictive models can anticipate a patient's response to various chemotherapy regimens, enabling clinicians to choose the most appropriate and targeted treatment. The incorporation of AI into chemotherapy enhances patient outcomes and facilitates the advancement of more efficient, precise, and minimally invasive cancer treatment. These advancements highlight the increasing significance of AI in advancing biological cancer research [32][33][34].

Immunotherapy

AI, utilizing methodologies such as machine learning (ML) and deep learning (DL), is progressively influencing contemporary medicine by executing intricate tasks with limited human involvement. In healthcare, AI facilitates diagnostic validation, risk evaluation, data analysis, outcome forecasting, therapy supervision, and virtual health support. These skills are revolutionizing medical practice and improving patient care [35][36].

In cancer immunotherapy, AI is employed to identify fundamental immune response patterns and to forecast patient responses to particular treatments. Through the analysis of high-throughput genetic data and medical imaging, AI offers insights that facilitate patient selection, therapy optimization, and personalized prognosis forecasting [37][38].

Immunotherapy has emerged as a fundamental component of holistic cancer treatment, frequently augmenting conventional methods such as chemotherapy, radiation, and surgery. Progress in biotechnology has enhanced our understanding of tumor molecular biology, allowing immunotherapy to utilize the patient's immune system to combat various tumors. This technique reinstates normal immune function and activates immunological processes to identify, target, and eradicate cancer cells. Essential stages in this process encompass the liberation and display of tumor antigens, activation of effector T cells, infiltration of T cells into tumor tissue, and the identification and annihilation of tumor cells by these activated immune cells [39][40].

The incorporation of AI in immunotherapy boosts treatment precision and efficacy while expediting the formulation of personalized treatments, a significant advancement in cancer care [41].

Targeted Therapy

AI is increasingly transforming targeted therapies in cancer research by improving the accuracy and customization of cancer treatment by allowing oncologists to formulate therapies that correspond to the distinct biological traits of each patient's tumor. By integrating and analyzing complex clinical datasets and individual patient profiles, AI facilitates the creation of personalized treatment strategies that enhance therapeutic efficacy and minimize adverse side effects, thereby improving patient outcomes and quality of life supporting more effective and patient-centered care [42][43].

Additionally, AI aids in identifying specific molecular targets for therapy, guiding clinicians and researchers in selecting the most suitable drugs or interventions for each patient. By facilitating these data-driven decisions, AI is helping to advance precision medicine and usher in a new era of individualized cancer treatment, which marks a significant step forward in biomedical research [44][45].

Surgery

The removal of tumors is a crucial component of cancer treatment, and artificial intelligence is progressively revolutionizing this domain by enhancing surgical accuracy and results. A significant advancement is the implementation of computer-assisted surgery (CAS), which has become a crucial element in numerous contemporary oncological procedures. CAS improves the precision and efficacy of surgical procedures, thus enhancing patient safety and overall care. Moreover, computer vision (CV) technologies are essential in image-guided navigation systems, which combine preoperative imaging data, such as CT or MRI scans, with real-time intraoperative visuals. This integration allows surgeons to visualize anatomical structures in greater detail, accurately differentiate between healthy and pathological tissues, and meticulously plan surgical procedures. By combining radiological imaging with tracking technologies embedded in surgical instruments, these systems align the instruments with the patient’s anatomy, highlighting structures that may not be visible. This guidance enhances surgical safety and effectiveness, helping surgeons navigate complex procedures with greater confidence [46][47][48].

Currently, image-guided navigation is especially valuable in neurosurgery and orthopedic surgery, where precise targeting is critical and anatomical structures are less affected by tissue displacement or organ movement. The integration of AI in surgery represents a significant advancement in oncology, improving both the accuracy and outcomes of tumor removal [49][50].

In surgical procedures when anatomical structures are substantially modified due to tissue manipulation, conventional computer-based navigation systems have demonstrated limited utility. Significant research endeavors are concentrated on creating AI-driven surgical navigation methods for intricate operations, especially abdominal surgeries where dissecting planes can substantially alter anatomy. These unique methodologies offer significant insights for visualizing concealed or essential structures, hence improving surgical safety and accuracy [51][52].

AI can facilitate the precise mapping of critical structures, such as the aorta and ureter, in relation to surgical instruments during laparoscopic rectal surgery. Likewise, computer-assisted liver mapping has been created to assist in liver cancer procedures. Future developments are anticipated to extend AI-based mapping to other organs, including the spleen, pancreas, and esophagus, utilizing CT imaging to enhance the accuracy of image-guided abdominal surgeries [53][54].

Robotic surgery demonstrates significant potential, with robotic prostatectomy becoming a prominent technique for prostate cancer treatment. The amalgamation of AI-assisted systems with surgical robots is poised to augment the surgeon's proficiency. Computer vision technology will deliver real-time insights into anatomical structures and resection margins by juxtaposing intraoperative data with comprehensive databases comprising millions of reference photos. The integration of AI and robotics is set to revolutionize intricate cancer procedures, enhancing precision, efficiency, and patient results [55][56].

Nanotechnology

The amalgamation of artificial intelligence (AI) with nanotechnology possesses significant promise to transform cancer management. Artificial intelligence can effectively process and analyze extensive datasets, facilitating the development of precision nanomedicines tailored to individual patient profiles. AI boosts early diagnoses and improves treatment planning accuracy through molecular profiling and predictive modeling. It enhances nanomedicine development by refining physicochemical qualities, augmenting medication synergy, and reducing toxicity; hence, it results in more precise and effective therapeutics [57][58].

Conventional cancer therapies frequently encounter obstacles, including significant adverse effects and restricted infiltration of therapeutic agents into solid tumors owing to the aberrant extracellular matrix (ECM). AI-assisted anticancer nanomedicines are being developed to improve targeted delivery and treatment efficacy to surmount these challenges. Nanomedicines, which integrate nanotechnology with medicine, are progressively employed in disease diagnosis, monitoring, and treatment. Numerous nanomedical devices are currently employed in clinical settings for malignancies including breast, ovarian, lung, pancreatic, and hematologic tumors [59][60][61].

Artificial intelligence plays a crucial role in enhancing the formulation and delivery of nanomedicine. Advanced algorithms can forecast interactions between nanocarriers and encapsulated pharmaceuticals, simulate release kinetics, and assess encapsulation effectiveness, thus enhancing targeting and therapeutic results. Nonetheless, numerous obstacles persist. AI systems necessitate extensive, high-quality datasets, and data biases can affect predictive accuracy. Moreover, the opaque nature of AI models generates interpretability issues in clinical applications [62][63].

Notwithstanding these constraints, AI-driven decision support systems have exhibited diagnostic and treatment-planning precision akin to that of expert oncologists. This indicates that the incorporation of AI into nanomedicine may augment precision medicine and patient care. In the future, it will be essential to address data quality, transparency, and ethical considerations in order to optimize the advantages of AI in cancer. AI and nanotechnology collectively represent a significant advancement in personalized cancer detection and therapy, enhancing results, and shaping the future of biomedical research [64][65].