AI in Oncology Market: Growth, Size, Share, and Trends

According to the World Health Organization (WHO)’s cancer agency, the International Agency for Research on Cancer (IARC), in 2022, there were an estimated 20 million new cancer cases and 9.7 million cancer-related deaths globally. The number of people alive five years after a cancer diagnosis was 53.5 million. Around 1 in 5 people are likely to develop cancer in their lifetime, with 1 in 9 men and 1 in 12 women expected to die from the disease. 10 types of cancer were responsible for about two-thirds of new cases and deaths in 2022. The IARC also projects that the number of new cancer cases will reach 35 million by 2050, marking a 77% increase. This rising cancer burden is attributed to factors such as population aging and growth, alongside increased exposure to risk factors like tobacco use, alcohol consumption, obesity, and air pollution

High initial investments remain a significant restraint for AI adoption in the oncology market, particularly due to the costs associated with development, data acquisition, and regulatory approval. According to the FDA, the costs of regulatory processes can be substantial. For instance, the 510(k) process (premarket notification) costs USD 24,335, whereas the premarket approval (PMA) process for Class III devices can run as high as USD 540,783. The De Novo Classification Request, which is necessary for novel devices without existing analogs, costs USD 162,235. These regulatory costs alone are considerable, particularly for small to mid-sized AI startups that must also allocate funds for research, data collection, and clinical trials

One of the primary advantages of AI in radiomics is its ability to handle a vast amount of data far better than traditional methods. AI, especially through deep learning, can analyze both predefined radiomic features and raw imaging data, automatically generating its features from these images. In August 2022, researchers at the NIH developed a deep learning-based model, DeepSurv, leveraging computed tomography (CT) radiomics and key clinical features to predict personalized survival outcomes for cancer patients. The model aims to enhance treatment decision-making by providing more accurate, individualized predictions based on both imaging data and clinical characteristics. Thus, this synergy of AI and radiomics is paving the way for more precise, less invasive cancer diagnostics and treatment planning

The limited availability of datasets is a significant challenge in the AI-driven oncology market, particularly due to data privacy and protection regulations such as the General Data Protection Regulation (GDPR) and the Health Insurance Portability and Accountability Act (HIPAA). These regulations impose strict limitations on the collection, storage, and sharing of sensitive health data, complicating efforts to build comprehensive datasets for AI training. HIPAA governs the US' protected health information (PHI), focusing on ensuring the confidentiality, availability, and integrity of health data. GDPR provides broader protection for all personal data of EU residents, including health-related information. These regulations necessitate careful management of data privacy, and non-compliance can result in significant penalties of up to USD 21 million (€20 million) or 4% of global revenue under GDPR and up to USD 2 million for severe violations of HIPAA