Algorithms or computer programs that use data to determine the course of action or make predictions are referred to as artificial intelligence. In order for the computer to examine data and come to a judgment, scientists may develop a set of rules or instructions for the computer to follow. Machine learning is another artificial intelligence technique where the system trains itself on how to evaluate and understand data. As a result, machine learning algorithms may spot patterns that are difficult for the human eye or brain to recognize. Additionally, these algorithms get better at learning and interpreting the data as they are exposed to more fresh information.
Data Bridge Market Research analyses that the artificial intelligence in healthcare market is expected to undergo a CAGR of 51.37% during the forecast period of 2022-2029. This indicates that the market value, which was USD 6.35 billion in 2021, would rocket up to USD 175.22 billion by 2029. In January 2019, Dartford and Gravesham NHS Trust in the United Kingdom developed an AI-powered wearable technology for patient monitoring when discharged from the hospitals. In October 2019, care.ai and NVIDIA announced a collaboration to provide an artificial intelligence-powered autonomous patient monitoring in healthcare leveraging NVIDIA'S platform.
To know more about the study, visit: https://www.databridgemarketresearch.com/reports/global-artificial-intelligence-in-healthcare-market
Deep learning, a subset of machine learning, has also been applied by researchers in cancer imaging applications. Deep learning refers to algorithms that categorize data in methods similar to the human brain. Artificial neural networks are used by deep learning technologies to simulate how our brain cells receive, interpret, and respond to messages from the rest of our body. In order to determine if a mass is cancerous or not, doctors perform cancer imaging tests. How quickly is it developing if it is cancer? How much of a spread is it? Has it recovered since receiving treatment? According to studies, AI may enhance medical professionals' responses' promptness, precision, and dependability. The application of AI in oncology can be understood at different stages:
Fig.1: The Role of AI in Oncology
- Finding Cancer at an Early Stage- People are routinely examined for indications of cancer or cells that could develop into cancer using procedures such as mammography and Pap tests. The objective is to identify and treat cancer early before it spreads or even grows. To help with breast cancer screening tests and other types of cancer screening testing, scientists have created AI technologies. Over the past 20 years, AI-based computer algorithms have assisted doctors in deciphering mammograms, but the field of study is rapidly developing. An AI system was developed by one team to assist in deciding how frequently women should be checked for breast cancer. The algorithm predicts a person's likelihood of acquiring breast cancer within the following five years based on the results of their mammograms. The model performed better in tests than the current breast cancer risk prediction methods. A deep learning algorithm that can recognize cervical precancers that need to be removed or treated has been developed and tested by NCI researchers. Health professionals in some low-resource situations examine the cervix with a tiny camera to check for cervical precancer. This approach is straightforward and sustainable; however, it is not very precise or dependable. Several AI technologies have been demonstrated in clinical studies to improve the diagnosis of adenomas, which are precancerous growths that can lead to colon cancer. Some specialists are worried that these AI technologies could force many people to undergo unneeded treatments and additional tests because only a small proportion of adenomas develop into cancer.
- Cancer Detection and Diagnosis- AI has the ability to help diagnose cancer earlier in people who are already exhibiting signs. For instance, the AI model created by Dr. Turkbey and his colleagues at the NCI's Center for Cancer Research may make it simpler for radiologists to identify prostate cancer that may be aggressive on a relatively new type of prostate MRI scan known as multiparametric MRI. The AI model developed by the NCI team "may minimize the error rate and make [the learning] curve easier for practicing radiologists," according to Dr. Turkbey. He said that the AI model might act as "a virtual expert" for less-experienced radiologists who are learning to use multiparametric MRI. Many deep learning AI models have been developed to assist clinicians in detecting lung cancer on CT scans. There is a significant proportion of false-positive test results that indicate a person has lung cancer when they actually don't because some non-cancerous abnormalities in the lungs might seem on CT scans to be very similar to cancer. Theoretically, AI might reduce the incidence of false positives and spare some patients from unnecessary stress, follow-up testing, and surgeries by better differentiating lung cancer from noncancerous alterations on CT images. A team of researchers created a deep learning algorithm to discover lung cancer and avoid other alterations that resemble cancer.
- Cancer Treatment Choice- Doctors also use imaging tests to gather crucial data on cancer, such as how quickly it is developing, whether it has spread, and whether it is likely to return following therapy. Doctors can use this information to determine their patients' best course of action. Numerous research indicate that AI may be able to more accurately and comprehensively extract prognostic data from imaging scans than humans currently are. For instance, a deep learning model developed by Dr. Harmon and her associates can predict the risk that a patient with bladder cancer will require additional therapies in addition to surgery. According to medical professionals, clusters of cancer cells that have moved outside of the bladder in around 50% of persons with tumors in the bladder muscle (muscle-invasive bladder cancer) are too tiny to be detected using conventional methods. These undetected cells can keep multiplying after surgery if they aren't eliminated, leading to a recurrence. These small clusters can be eliminated by chemotherapy, stopping the cancer from returning after surgery. However, as demonstrated by clinical trials, it might be challenging to identify whether patients also require chemotherapy, according to Dr. Harmon. The model analyses digital images of original tumor tissue to determine whether there are microscopic cancerous groupings in surrounding lymph nodes. In a study published in 2020, the deep learning model outperformed the conventional method of predicting if bladder cancer has spread, based on several variables including the patient's age and specific tumor characteristics. More and more, the genetic makeup of the patient's cancer is being studied to determine the best course of action. Chinese researchers developed a deep learning algorithm to predict the existence of important gene mutations in liver cancer tissue from photographs of the tissue, something pathologists are unable to accomplish just by looking at the images. The scientists who created the algorithm don't know how it determines which gene changes are present in the tumor, making their tool an example of AI that operates in surprising ways.
- AI in Medical Imaging- The prediction of cancer can benefit from AI and machine learning. Artificial intelligence is able to spot malignancies that have already spread and people who are at a high risk of getting it before it does. This enables medical professionals to closely monitor these patients and act quickly when necessary. A computer scientist at MIT named Regina Barzilay was interested in testing artificial intelligence (AI) for cancer prediction. The MIT team looked at its potential for identifying women at risk of breast cancer before any overt symptoms appear. To find out which patients had cancer, she gathered over 40,000 women's mammograms over a four-year period, totaling about 89,000, and compared the scans to the national tumor registry. Regina then used a selection of these photos to train a machine learning (ML) algorithm, a sort of AI, and then used that algorithm to generate predictions. The algorithm correctly identified 30% of future breast cancer patients as belonging to a high-risk group. AI has various uses in the field of medical imaging. Identifying and categorizing malignant tumors is one of the most obvious. The FDA authorized Paige Prostate, an AI-powered pathology tool for cancer, in September 2021. Together with the FullFocus digital pathology viewer, this AI tool aids in the detection of prostate cancer. The FDA reviewed data from a clinical investigation where 16 pathologists assessed 527 prostate biopsy photos in search of cancer indicators as a prerequisite for this approval.
- AI in Blood Testing- Blood testing with AI enhancements can aid doctors in more accurate cancer detection. According to a study in Cancer Cell International, blood profiling, which analyses plasma ctDNA and miRNA profiles using AI algorithms, is a more effective way to find and monitor cancer than conventional CT scans. A cutting-edge AI-based technique was created by Johns Hopkins Kimmel Cancer Center researchers to detect lung cancer using blood testing. Blood samples from 796 US, Denmark, and Netherlands participants were used to test this method. This blood test was paired by researchers with protein biomarkers, clinical risk factors, and CT scans of the patients. They correctly identified cancer in 91% of people with early illness stages and in 96% of patients with advanced cancer phases as a result.
- AI in Immunotherapy- The primary function of AI in immunotherapy is to assess the outcomes of various therapies and assist physicians in modifying their prescriptions. An AI-powered method was developed by a research team at the MD Anderson Cancer Center and UT Southwestern Medical Center to determine whether neoantigens—peptides made when the genomes of cancer cells are mutated—are recognized by a patient's immune system. Such AI algorithms would make it possible to forecast how cancer cells will react to immunotherapies. T cells in our immune system are always on the lookout for indications of cancer and other invasive organisms. These cells bind to one another when they identify neoantigens. However, some neoantigens go unidentified, which promotes the spread of cancer. This information would make the ability to anticipate patient response to immunotherapies and create individualized T cell-based therapeutics and cancer vaccines possible.
The immuno-oncology (IO) market is expected to witness market growth at a rate of 8.90% in the forecast period of 2022 to 2029. The immuno-oncology (IO) market is segmented on the basis of type, target, indication, end users and distribution channel. Asia-Pacific is projected to observe significant amount of growth in the growing favorable growth rate in cancer immunotherapy adoption. Moreover, the rise in the incidence of the disease and in turn, increasing mortality rate is further anticipated to propel the growth of the immuno-oncology (IO) market in the region in the coming years.
To know more about the study, visit: https://www.databridgemarketresearch.com/reports/global-immuno-oncology-market
- Drug Development- The same medicine may respond differently to various forms of cancer. AI is able to forecast how various drugs would affect malignant cells. This information aids in the creation of new anticancer medications and the timing of their use. For instance, depending on the mutational state of the cancer cell, a research team created a random forest algorithm that can forecast the action of anticancer medications.
Benefits of AI in Oncology
AI generally has lots of advantages in the medical field. Here are the top three benefits of using artificial intelligence in cancer detection and treatment:
Fig.2: Benefits of AI in Oncology
- Personalized Medicine and Therapies - Big data and AI enable medical professionals to examine a variety of data about the patient and the cancer cells to develop individualized treatments. The side effects of this kind of therapy will be less severe. Less harm will be done to healthy cells, but it will have a greater effect on cancer cells. AI aids radiologists in determining which tumors and anomalies are cancerous and require genuine medical intervention. According to a study in the Journal of the National Cancer Institute, AI algorithms can identify precancerous lesions in cervical pictures and differentiate them from other abnormalities to save patients from receiving unnecessary treatment for little problems.
- Elimination of Invasive Procedures- Sometimes, the tumor's benign nature is discovered only after the removal surgery, which would have allowed the procedure to be completely avoided. Such occurrences can be greatly decreased with AI's assistance in the cancer detection process. One study, for instance, found that AI can cut back on breast-conserving procedures by 30.6%. Image-guided needle biopsies can be used to train machine learning algorithms to recognize malignant tumors. A random forest ML system was used to assess 335 potential cancer patients, and the researchers found that it stopped one-third of unneeded procedures.
- Reduction in False Positives and Negatives- AI for cancer detection will increase diagnostic precision and decrease false positives and negatives. We have proof thanks to research on breast cancer detection. One in ten female patients who have mammograms examined by doctors have false-positive results, forcing them to undergo stressful procedures and unneeded invasive testing. The research team at Google created software that uses AI to reduce false positive and false negative mammography readings by 6% and 9%, respectively. Another team of researchers created an AI algorithm for the identification of breast cancer. This algorithm assisted radiologists in lowering false-positive rates by 37.3% during an examination.
Challenges for AI in Oncology and Future Outlook
Complex nonlinear interactions, fault tolerance, simultaneous distributed processing, and learning are all tasks that AI can handle with ease. due to its benefits of self-adaptation, the simultaneous treatment of quantitative and qualitative information, and validated outcomes from numerous clinical studies in numerous domains. There is no doubt that AI is used in clinical care in a variety of ways. It fully exploits the different facets of clinical variability while also addressing the current lack of universality and objectivity in expert systems. Hospitals can train junior doctors in clinical diagnosis and decision-making by using AI. A growing number of academic papers discuss the remarkable diagnostic and prognosis capabilities of ML-based computer systems.
To assure its application in cancer diagnosis and prognosis, AI technology faces some significant difficulties that must be overcome. For instance, raw input data from medical imaging cannot be used. Processing and extracting information from the image data is essential. Further study is needed to interpret the results of the weights coefficient in neural network models, which have been validated, calculated, and have adequate confidence intervals due to technological development and widespread adoption. The field of clinical medicine will probably use ANNs more frequently as a result of greater research into them. Although the value of AI in this industry is acknowledged, computer scientists and medical professionals must work together to ensure that interdisciplinary staff members are trained and collaborate. Medical professionals can then utilize the potential of this technology in a cost-effective and practical manner. Privacy and data security assurances are a major problem with relation to the future of AI in medicine. Although "big data" and ML-based solutions have generated a lot of excitement in recent years, there are currently very few cases that show how AI has affected clinical practice.
Data Bridge Market Research analyses that the cancer diagnostics market is expected to reach the value of USD 28.21 billion by the year 2029, at a CAGR of 7.29% during the forecast period. The rise in the cancer cases provides growth opportunities to the market. Cancer is the world's second leading cause of death, accounting for 10 million deaths by 2020. Cancer accounts for approximately one-sixth of all deaths worldwide (Source: World Health Organization). In 2020, 19.3 million new cancer cases were reported, with that number expected to rise to 30.2 million by 2040. This increase in cancer incidence can be attributed to the growing geriatric population as well as the overall population.
To know more about the study, visit: https://www.databridgemarketresearch.com/reports/global-cancer-diagnostics-market
