AI In Cancer Care: How It's Making a Difference In Treatment And Care
One of AI's key strengths is that it is able to process vast and complicated data in short amounts of time, and help automate routine tasks to reduce the level of human intervention needed
Cancer incidence continues to grow worldwide. As per the latest figures from WHO's cancer database GLOBOCAN, 19.3 million new cases of cancer were reported in 2020.This figure is expected to rise to 27.5 million new cancer cases diagnosed each year by 2040. Thus, cancer will remain a key global health issue and utilize a significant chunk of our healthcare resources.
Many countries such as India face challenges in terms of limited healthcare resources available to treat the swelling number of cancer cases. The ratio of patients seeking care for cancer to the availability of cancer specialists is very high compared with developed countries. Indian oncologists, on average, treat a much larger number of cancer patients than their Western counterparts. The pathologist-patient ratio is also highly skewed, and hence the cancer care infrastructure in India faces enormous time pressure, with doctors having to examine a staggering amount of information to make treatment decisions, for every single cancer patient. The result is overcrowded health care facilities and long waiting periods in hospitals equipped to deal with cancer.
This is where artificial intelligence (AI) can be a game changer. One of AI"s key strengths is that it is able to process vast and complicated data in short amounts of time, and help automate routine tasks to reduce the level of human intervention needed.
Application to image analysis allowing better screening and more efficient diagnosis
Early detection of cancer is the key to saving the lives of patients. A group at MIT developed a new deep learning-based prediction model that can forecast the development of breast cancer up to five years in advance. Their model was trained on mammograms and patient follow-up data to identify patterns that would not be obvious to or even observable by human clinicians. The results have so far shown to be far more precise, especially at predictive, pre-diagnosis discovery.
Whole‐slide imaging is becoming routine in developed countries, which has resulted in the accumulation of digital pathology images and allowed the application of deep learning to pathological diagnosis. A deep learning convolutional neural network, or CNN, developed by a team from Germany, France and the US can diagnose skin cancer more accurately than dermatologists. In a recently reported study, the software was able to accurately detect cancer in 95 per cent of images of cancerous moles and benign spots, whereas a team of 58 dermatologists was accurate 87 per cent of the time.
By applying AI and machine learning to multiple data sources—omics data, electronic health records, sensor/wearables data, environmental and lifestyle data—researchers are taking first steps toward developing personalized treatments for diseases from cancer to depression. Here, AI is in action today and making great strides in cancer treatment by leveraging patient medical history and tumour characteristics to help generate multiple treatment options.
Various AI/ML models for breast cancer prognosis have successfully transitioned to clinical use. These models help accurately determine the risk of a patient suffering from a relapse, based on which treatment can be personalized. If a breast cancer patient has a low risk of relapse, then they could potentially avoid chemotherapy and all its side-effects. Other localized cancer treatments such as radiation are also increasingly relying on AI. Radiation oncologists are already using AI-driven software to create plans for personalized radiation therapy.
In the near future, AI/ML can mine large datasets (scans, blood work-up, patient follow-up from thousands of patients at one time) to detect early signs of patients who are responding to treatment, and those who are not.
AL/ML can be applied in multiple stages of new drug discovery including designing the chemical/protein structure of drugs, target validation, investigating drug safety and managing clinical trials. The hope is that use of AI/ML in drug discovery will not only help significantly reduce the cost of introducing new drugs to the market, but also make the drug discovery process faster (currently 10-15 years including clinical trials) and more cost-effective (currently costs almost $1 billion per new drug). Companies today use deep learning software to sift through millions of possible molecules in a day or two, which would normally take months via traditional methods.
In conclusion, some AI solutions have already been deployed in clinical practice, but the industry has a long way to go.
In my opinion, AI should be seen as something that can help cancer specialists spend less time on routine tasks, reduce variability and human error rather than replace the specialists altogether. AI can sift through large data sets and aid in decision making rather than be a standalone tool for diagnosing or treating patients in a completely unsupervised environment. Also, we must ensure that the data going in to make the AI based algorithms must be of the highest quality/standards to ensure we get the most accurate algorithms in return.
With all the fears about AI making many jobs redundant, oncology could be a great example of how humans and technology can work together rather than against each other.