Artificial intelligence (AI) has reached new heights in clinical cancer diagnosis.
Cancer is a dreaded disease characterised with its low median survival rate. The cancer treatment process is long and very expensive due to high reoccurrence and high mortality rates. Accurate early diagnosis and prognosis prediction can increase a patient’s survival rate. Of late Artificial intelligence (AI), especially machine learning and deep learning, have found popular applications in clinical cancer research, early diagnosis and prescriptive treatment.
Here are use cases where artificial intelligence is moving forward to decrypt the holy grail of cancer care.
Tissue biopsies have formed the pillar of a cancer diagnosis for decades, but the invasive procedure is not risk-free. Using AI-based algorithms to conduct liquid biopsies is a possible solution that oncologists look forward to
The day is not far when cancer screening could be as simple as taking a blood test. Medical analysts are deigning AI programmes that help to detect circulating tumour DNA in blood samples which could prove a useful tool in the early detection of the cancer disease.
Application to image analysis
Early detection of cancer is the key to saving the lives of affected individuals. Deep learning has revolutionized image analysis, with many researchers and physicians have attempted to harness the power of AI for clinical radiology and pathology trials.
One example of high AI impact is the successful classification of dermoscopy images where Artificial Intelligence is found to be able to annotate skin lesions in the same lines as expert dermatologists would do. Given that smartphones extend the reach of dermatologists outside the clinic, this achievement has the potential to provide universal access to dermatologist‐level diagnoses.
A key challenge in cancer diagnosis lies the development of optimal therapies, which is characterised to achieve a positive outcome of many patients. The current “gold standard” approaches to disease classification in oncology encapsulate the opinion of expert pathologists with the expression of molecular markers such as cell surface receptors at the protein or mRNA level. Given the increasing complexity and the colossal amount of molecular data available, AI has a greater role to play in identifying disease subtypes more comprehensively to accurately predicting future disease behaviour and patient treatment response.
AI and clinical decision making
Clinical trial enrolment, drug development, and biomarker discovery in oncology give a plethora of opportunities for AI to assist in data synthesis that guides decision-making. Several commercial applications leverage deep learning and natural language processing linking patient data to clinical trial databases. AI-based NLP and deep learning approaches match patients to appropriate clinical trials nationwide.
Cancer is a leading cause of death in developed countries and is estimated that the number of cancer disease cases will increase further in ageing populations. Surprisingly Japan, one of the ageing world economies witnesses almost 1 million individuals diagnosed with cancer every year and nearly 400 000 of them die of the deadly disease.
The application of artificial intelligence to the diagnosis of cancer disease based on the classification of radiological or pathological images has demonstrated a future promise that exceeds that of medical experts. With high expectations on Artificial Intelligence, AI-based cancer research will, thus, continue to be a top priority for oncologists across the world in the years to come.