AI Improves Cancer Research Significantly

Introduction

Artificial Intelligence (AI) is present everywhere. Today, we have personal digital assistants that can answer our questions, robot-advisors that can trade stocks, and soon we will have driverless cars that can take us anyplace we want to go. AI has indeed entered our lives significantly and its use has increased by leaps and bounds in health care and biomedical research and that also includes all spheres of cancer research, where the potential of AI is quite vast.

Use of AI in Cancer Detection

AI excels at identifying patterns in huge volumes of data, deriving relationships between multifaceted features in the data, and recognizing characteristics in data (comprising images) that cannot be understood or also perceived by the human brain. AI has already generated great results in radiology, where clinicians utilize computers to generate images at a fast pace, thus enabling radiologists to concentrate on aspects where their technical judgment is of paramount importance.

Integration of AI technology in cancer care has the ability to improve the precision and pace of diagnosis, help in clinical decision-making, and also give better health outcomes. AI-guided clinical care has a significant opportunity to play a crucial role in decreasing health disparities, especially in low-resource environments.

Development of AI Protocol

University of Texas MD Anderson Cancer Center and the UT Southwestern Medical Center researchers have developed an AI technique to understand and determine anti-cancer immunity. According to researchers, the pMTnet technique could generate predictive analytic methods to achieve immunotherapy response and cancer prognosis.

“Determining which neoantigens attach themselves to T cell receptors and the ones that do not seems like a challenging task. However, with ML, we are making progress,” states senior author Tao Wang, PhD.

With predictive analytics working to understand which neoantigens are identified by T cells, researchers could create customized cancer vaccines, better t cell-based therapies, or also predict to see how patients react to various immunotherapies.

Upcoming AI Applications in Oncology

NCI-funded research has already given rise to various opportunities for the use of AI:

• Improving Cancer Screening and Diagnosis: Scientists leading the NCI’s intramural research program are leveraging the abilities of AI to expand cancer screening in prostate and cervical cancer. NCI investigators have created a deep learning approach for the automated identification of precancerous cervical abrasions from digital images.
• Aiding the Genomic Characterization of Tumors: AI methods can also be used to recognize and identify explicit gene mutations from tumor pathology images as a substitute for using traditional genomic sequencing. For instance, NCI-funded researchers at New York University have used deep learning (DL) to analyze pathology images of lung tumors that are obtained from The Cancer Genome Atlas.
• Accelerating Drug Discovery: NCI is utilizing the power of AI in various ways to find new treatments for cancer. The Cancer Moonshot is also helping out in two major efforts in partnership with the Department of Energy (DOE) to utilize its supercomputing power and expertise for cancer research.

The potential applications of AI in cancer research and medicine have great promise. Utilizing these opportunities will need substantial investments and addressing of challenges that will come up.

Conclusion

None of the researchers expect AI to completely replace radiologists, physicians, or pathologists. But with an aging population, augmented availability of diagnostic tests and the increasing emphasis on precision medicine, AI can most certainly enable them to do their jobs by recognizing the high-risk cases they should concentrate on, thus, helping them to make decisions about the uncertain diagnosis.