Revolutionising Cancer Care: AI's Role in Detection and Diagnosis

  • Elena Paspel Master of Science in Engineering (Digital Health) - Tallinn University of Technology, Estonia

Introduction to AI in cancer care

Every year, 19.3 million new cases of cancer are diagnosed globally. As such a prevalent issue, many of us will have been personally affected by cancer.1 Thankfully,  new artificial intelligence (AI) technologies are emerging as a promising way to significantly improve prevention, early detection, and treatment outcomes.

AI refers to advanced computer software or ‘algorithms’ that can analyse data and make predictions about diseases.1 One key subset of AI is machine learning, which allows computer systems to ‘learn’ from data patterns and improve their performance over time without explicit programming or supervision.1 Deep learning techniques take this even further by imitating the human brain’s layered approach to processing data.

Image of three nesting dolls, with words written in the middle of each doll. "Artificial Intelligence (AI)" is written on the largest doll. "Machine Learning" is written on the middle doll. "Deep Learning" is written on the smallest of 3 dolls. The text underneath the dolls says "Artificial Intelligence (AI) is like a set of Russian nesting dolls, with each layer revealing more specialized technology. At the broadest level, AI is a collection of computer techniques that imitate human thought processes. Nestled within AI is machine learning, which focuses on using statistical methods to uncover hidden patterns in data. Deep inside machine learning is deep learning, which delves deeper with its complex, multi-layered networks for more intricate data analysis."

Source: Image generated using AI (DALL-E) 

These AI and machine-learning methods are revolutionising cancer care by enabling much more accurate risk assessments, diagnoses and prognoses and can even allow personalised treatment.1 

  • AI technologies, including computer-aided detection (CAD) systems, support radiologists with detecting and locating cancer in the body, especially in early cancer stages where detection is crucial for recovery​​.2
  • AI's ability to learn complex data patterns and make predictions is set to revolutionise early cancer diagnosis, aiding doctors through analyses of health records, medical images, biopsy samples, and blood tests​​.3

However,  limitations around aspects like patient data privacy and difficulties in interpreting the models remain barriers to real-world implementation on a wider scale.1 

Still, real-life applications of AI in oncology, such as FDA-approved AI tools that aid in diagnostics, provide concrete evidence of how these technologies are saving and improving lives.4

Understanding AI's role in cancer detection and diagnosis

Research shows that interest in using artificial intelligence (AI) to detect cancer continues to increase. Between 1986 to 2022, over 6,000 articles were published on AI for finding cancer, indicating major attention on this topic.2

  • Analysing complex information to detect cancer: AI tools, especially convolutional neural networks (CNNs) – a type of deep learning algorithm good at analysing visual information like medical scans and images, can effectively analyse complex data like scans, biopsy images, clinical notes, and genetic, metabolic and radiomic information. They have successfully found early-stage cancers by checking scans and biopsy slides, significantly impacting how doctors evaluate tests.​3
  • Identifying cancer types for precise diagnosis: CNNs can find out the type of a tumour and detect useful information like unique molecular patterns and receptors on the surface. This can help with designing a personalised treatment plan to target each specific kind of tumour. AI can also automate assessments of cancer grade and stage.3 For example, an FDA-approved generative AI for pathology provides automated prostate cancer analysis. Another AI method called deep learning autoencoder, applied by researchers on The Cancer Genome Atlas datasets, has identified distinct cancer subtypes, aiding in more accurate diagnosis and understanding of various cancer forms.3

How AI assists broader cancer care 

  • AI can help to predict if cancer will return after treatment and detect earlier recurrence.​3
  • Before treatment starts, AI can identify high-risk patients needing intensive treatment and low-risk patients needing less intensive treatment, ensuring everyone gets the right kind of care in a timely manner.3
  • After treatment, AI can tailor patient monitoring based on the risk of cancer returning,  enabling earlier treatment and diagnosis of new cancers.3

Current AI techniques to find and diagnose cancer

Artificial intelligence and machine learning tools are making major inroads into cancer detection, diagnosis, and decision-making in the clinic. Some major developments in digital pathology and medical imaging analysis include:1

  • Recognising cancerous tissue patterns - Deep learning analysis of medical scans and images like CT scans, a type of scan frequently used to detect cancer, can spot differences between harmless and cancerous tissue based on texture and shape, which are not visible to the human eye. AI models are trained on either proprietary or publicly available cancer imaging datasets
  • Automated screening for skin cancer - AI programs can now categorise melanoma and other skin cancers with accuracy, equaling expert dermatologists by examining images of skin lesions.
  • Spotting prostate cancer in biopsies – An AI system recently achieved 98% accuracy in finding prostate cancer cells in images taken from core needle biopsy, which involves using a special needle to extract small tissue samples from the body for examination.
  • Distinguishing brain tumours - AI methods can be used to diagnose brain tumours while brain surgery is taking place by differentiating cancerous tissue from normal brain tissue, which typically needs to be done using a microscope.

These examples show how AI is progressing to improve and support the diagnosis of cancer in the clinic, automating screening and making it faster, helping clinicians by providing second opinions on pathology tests, and potentially enabling earlier treatment through timely detection.

AI advancing the detection of common cancers

With an expected 2.3 million new cases (11.7%), female breast cancer is now the most commonly diagnosed type of cancer, followed by lung (11.4%), colorectal (10.0%), prostate (7.3%), and stomach (5.6%). Artificial intelligence (AI) tools are making major strides in finding common cancers earlier and more precisely. Let's explore some key examples.

Breast cancer detection with AI

AI has progressed greatly in finding breast cancer, especially in analysing mammograms. Mammograms are images of the breast routinely used to screen for and detect breast cancer. Convolutional neural networks (CNNs) – deep learning models good at processing images, achieve particularly high accuracy.5 Deep learning programmes are better than radiologists at spotting breast cancer in mammograms. They demonstrate higher sensitivity and specificity, meaning fewer missed cases or false alarms.​5

AI programmes can help radiologists catch early breast cancer signs, improving diagnosis and reducing workload. These systems can prioritise high-risk cases, meaning they will get seen by a doctor quicker.​5 One study with over 3000 slide analyses from 1500+ patients reduced the workload of clinicians by 57.2% whilst still maintaining accuracy.​3 Integrating AI with breast screening could boost precision and efficiency, making it more likely that someone with breast cancer gets quickly and accurately diagnosed.​

Also, AI can assess digital mammography and digital breast tomosynthesis images for future breast cancer risk, which allows doctors to see numerous pictures of the breast rather than the customary single image acquired with traditional mammography.5 AI techniques have also been used to spot when typical breast texture hides lesions in screening mammograms, indicating that another screening method is necessary.​5

Lung cancer and AI's diagnostic precision

Deep learning-based AI systems for lung cancer imaging can enhance accuracy and efficiency in screening, and help clinicians distinguish different cancer types and minimise false-negative results.6

Lung tissue sample slides are typically prepared from biopsies or surgical specimens taken from a patient suspected of having lung cancer. Usually, pathologists examine these slides under a microscope to observe the cellular details, structure, and any abnormalities indicative of cancer. When using a machine learning algorithm to analyse tissue samples, researchers obtained a 97% accuracy rate in detecting two primary types of lung cancer. These types were adenocarcinomas and squamous cell carcinomas, which are generally difficult to differentiate, even for experts.1 

AI models based on the Electronic Health Record (EHR), a directory of patient health records, effectively predict lung cancer without the need for imaging. One model analysing over 6,505 patients with lung cancer and 189,597 controls improved on the existing eligibility criteria for screening. This aids early diagnosis by selecting those most likely to have lung cancer for targeted screening.3

AI also assists pathologists in characterising lung cancer subtypes and predicting how the patient will respond to treatment, significantly improving efficiency and reducing the chance of misdiagnosis.​ AI also improves the staging accuracy of tumours in PET scans, which are a useful form of imaging for finding tumours.​6 Though AI diagnostics are more likely to produce false positive results than radiologists, this is changing with advances like precise segmentation.​6

Colorectal cancer and AI-enhanced screening

AI-guided care plays an important role in improving screening, diagnosis, and treatment for colorectal cancer, reducing the number of missed adenomas and the risk of the cancer growing and developing​​.7

AI can match or surpass human performance in finding colorectal cancer and can be used in various screening approaches like colonoscopy, capsule endoscopy, and laboratory tests for efficient risk evaluation.7

AI techniques in routine screening cut incidence by advancing detection, diagnosis, and robotic surgery.​7

Bladder cancer identification via AI technology

AI algorithms for bladder cancer focus on detecting, staging, grading, and predicting recurrence and survival. Predictive models using artificial neural networks (ANN) have shown high accuracy when detecting bladder cancer using images taken from a cystoscopy, a procedure which involves looking inside the bladder using a thin camera.8 Machine learning models can also accurately classify the stage of bladder cancer using pathology information.8

Pancreatic cancer: AI for early discovery 

AI aids early pancreatic cancer detection by finding at-risk groups via images and lab tests, predicting outcomes like recurrence and response to treatment.​9 By mining electronic health records (EHRs), AI has been able to screen and detect cancer earlier, even years before diagnosis.​9

How AI is transforming patient care

Though impressively accurate, AI systems are not meant to replace doctors and nurses. These technologies can:1

  • Provide automated second opinions to help doctors and pathologists come to a conclusion.
  • Determine a patient’s risk by spotting trends clinicians might miss
  • Reduce backlogs by processing high volumes of data at a time.
  • Supply evidence to help with difficult treatment decisions.

Whilst AI is a promising development in cancer care, input from real doctors and clinicians will continue to be vital in the development, validation, and interpretation of AI models. This is to ensure safety, equity, and applicability in the clinic. It is also still important to treat patients with care and empathy during this difficult time, and talking with a real person can make them feel more comfortable and secure than an AI. 

Training healthcare teams on AI

With the influx of new AI tools, training clinicians on appropriate uses, limitations, and interpreting outputs is necessary to inform the correct decisions and provide the most benefit to patients. This might involve workflow changes to incorporate AI second opinions. Training and supervision will also be essential to prevent over-reliance on AI, as it can be worrying to think that clinicians may become complacent and assume the AI is always correct when this is not the case. 

When used correctly and with the correct training, AI will continue to have a place in cancer care. Since 2018, AI's role in cancer diagnostics has received approval from the Federal Drug Administration (FDA).10 The advantages AI-based devices offer in clinical practice, particularly in diagnosing breast, lung, and prostate cancers are increasingly being recognised.4

Understanding the challenges and considerations in cancer care AI

When we talk about using artificial intelligence (AI) in cancer care, it sounds exciting and full of potential. However, there are certain challenges and considerations we still need to think about to make this technology work best for everyone.

Ensuring AI works for everyone

AI needs to be thoroughly checked and proven effective, which is done through a process called 'model validation.' This is like giving the AI a test to make sure it works well. These tests and their results should be shared in well-known and trusted medical journals for everyone to see. However, we face challenges like keeping patient data private ('data anonymization'), the cost of storing large amounts of data, and sometimes not having enough information about all groups of people (like their race or ethnicity). This lack of information can make it hard to know if the AI will work well for everyone​​.In fact, many AI models reported in the scientific literature have only been tested on a single set of data.

Patient education: how does this affect you?

As a patient, it's important to understand that AI in cancer care is still evolving. It offers great promise for personalised treatment and a better understanding of your condition. If your doctor is using AI technologies, feel free to ask how it helps in your treatment and what benefits you can expect.


Artificial Intelligence (AI) is significantly transforming the landscape of cancer detection and diagnosis, which can help ensure people have quicker access to the treatment that works best for them. In summary:

  • Revolutionising early detection: AI enhances early cancer detection, particularly in the initial stages, which is crucial for successful treatment. It analyses medical images, health records, and biopsy data for early cancer signs.
  • Improving diagnostic accuracy: AI's advanced algorithms, such as deep learning, offer greater precision in identifying different cancer types, helping clinicians make the correct diagnosis.
  • Advancements across cancer types: AI shows notable progress in detecting common cancers like breast, lung, and colorectal cancers, leading to earlier and more precise identification.
  • Supporting clinicians in decision-making: AI tools provide valuable support to healthcare professionals, aiding in quicker and more accurate interpretations of diagnostic data.
  • Overcoming challenges: Despite its advancements, AI in cancer care faces challenges like data privacy, the requirement of sufficient training and the need for widespread, equitable access.

In essence, AI stands as a groundbreaking tool in cancer care by enhancing detection and diagnosis, paving the way for more effective treatment strategies and providing patients with the best outcome.


  1. Zhang B, Shi H, Wang H. Machine learning and AI in cancer prognosis, prediction, and treatment selection: a critical approach. J Multidiscip Healthc [Internet]. 2023 Jun 26 [cited 2023 Dec 11];16:1779–91. Available from: 
  2. Karger E, Kureljusic M. Artificial intelligence for cancer detection-a bibliometric analysis and avenues for future research. Curr Oncol. 2023 Jan 29;30(2):1626–47. 
  3. Hunter B, Hindocha S, Lee RW. The role of artificial intelligence in early cancer diagnosis. Cancers (Basel) [Internet]. 2022 Mar 16 [cited 2023 Dec 11];14(6):1524. Available from:
  4. Luchini C, Pea A, Scarpa A. Artificial intelligence in oncology: current applications and future perspectives. Br J Cancer [Internet]. 2022 Jan [cited 2023 Dec 11];126(1):4–9. Available from: 
  5. Sechopoulos I, Teuwen J, Mann R. Artificial intelligence for breast cancer detection in mammography and digital breast tomosynthesis: State of the art. Semin Cancer Biol. 2021 Jul;72:214–25. 
  6. Pei Q, Luo Y, Chen Y, Li J, Xie D, Ye T. Artificial intelligence in clinical applications for lung cancer: diagnosis, treatment and prognosis. Clin Chem Lab Med. 2022 Nov 25;60(12):1974–83.
  7. Mitsala A, Tsalikidis C, Pitiakoudis M, Simopoulos C, Tsaroucha AK. Artificial intelligence in colorectal cancer screening, diagnosis and treatment. A new era. Curr Oncol. 2021 Apr 23;28(3):1581–607.
  8. Borhani S, Borhani R, Kajdacsy-Balla A. Artificial intelligence: A promising frontier in bladder cancer diagnosis and outcome prediction. Crit Rev Oncol Hematol. 2022 Mar;171:103601. 
  9. Huang B, Huang H, Zhang S, Zhang D, Shi Q, Liu J, et al. Artificial intelligence in pancreatic cancer. Theranostics. 2022;12(16):6931–54. 
  10. Shimizu H, Nakayama KI. Artificial intelligence in oncology. Cancer Sci. 2020 May;111(5):1452–60. 
This content is purely informational and isn’t medical guidance. It shouldn’t replace professional medical counsel. Always consult your physician regarding treatment risks and benefits. See our editorial standards for more details.

Get our health newsletter

Get daily health and wellness advice from our medical team.
Your privacy is important to us. Any information you provide to this website may be placed by us on our servers. If you do not agree do not provide the information.

Elena Paspel

Master of Science in Engineering (Digital Health) - Tallinn University of Technology, Estonia

Bachelor of Laws - LLB (Hons), London Metropolitan University, UK

An experienced professional with a diverse background spanning law, pricing, and eHealth/Digital Health. Proficient in copywriting, medical terminology, healthcare interoperability standards, and MedTech regulations. A strong foundation in scientific research methodologies and user experience research supports the creation of compelling content for the biopharmaceutical, CROs, medical technology, and eHealth sectors.

Proven expertise in driving product vision, synthesizing complex information, and delivering user-centric solutions. Adept at streamlining workflows and processes, and drafting documentation and SOPs. Always open to collaborations and eager to connect with like-minded professionals. presents all health information in line with our terms and conditions. It is essential to understand that the medical information available on our platform is not intended to substitute the relationship between a patient and their physician or doctor, as well as any medical guidance they offer. Always consult with a healthcare professional before making any decisions based on the information found on our website.
Klarity is a citizen-centric health data management platform that enables citizens to securely access, control and share their own health data. Klarity Health Library aims to provide clear and evidence-based health and wellness related informative articles. 
Klarity / Managed Self Ltd
Alum House
5 Alum Chine Road
Westbourne Bournemouth BH4 8DT
VAT Number: 362 5758 74
Company Number: 10696687

Phone Number:

 +44 20 3239 9818