Many people may be worrying about how to restrict the effects of artificial intelligence (lack of privacy, copyright difficulties, job loss, global dominance, etc.). That does not, however, negate the fact that AI has a great potential to enhance human life on Earth.
Health care is one of these applications. The use of AI could result in major improvements in predictive diagnostics, including the early diagnosis of cancer, because of its capacity for processing large data sets. One of the most recent studies in the area indicates encouraging results for lung cancer AI-assisted diagnosis, albeit additional research is required.
According to doctors and researchers at the Imperial College London, the Institute of Cancer Research, and the Royal Marsden NHS foundation trust, an AI algorithm has been developed.
In the OCTAPUS-AI trial, researchers developed and tested machine learning (ML) algorithms to determine how well the models could predict recurrence using imaging and clinical data from over 900 patients from the UK and the Netherlands who had undergone curative radiation.
The study specifically examined whether AI could assist in determining a patient's likelihood of developing non-small cell lung cancer (NSCLC) again. In order to create an AI algorithm employing radio mics, researchers employed CT scans. This quantitative method uses medical imaging to extract new data and prognostic biomarkers.
85% of lung cancer cases are NSCLC instances. Even though the condition is frequently curable when discovered early, about a third of people have cancer recurrence. According to the study, practitioners may potentially be able to detect recurrence early in high-risk patients using the method.
To determine how well the model detected cancer, the researchers employed a statistic known as area under the curve (AUC). A model that was merely guessing 50/50 would receive 0.5, while a model with flawless accuracy of 100% would receive a score of 1. The research team's AI program received a 0.87 in the study. This can be contrasted to the 0.67 scores of the existing technology.
Dr. Sumeet Hindocha, a clinical research fellow at Imperial College London and a clinical oncology specialist registrar at The Royal Marsden NHS Foundation Trust, said, "We want to explore more sophisticated machine learning techniques, such as deep learning, to see if we can get even better results." Then, "We want to test this model on newly diagnosed NSCLC patients and follow them to see if the model can accurately predict their risk of recurrence."
Support for practitioners – and patients
Most people today see AI in health tech as a tool that will help practitioners in giving the best treatment possible, including greater bedside manners, rather than as something that would replace doctors. The healthcare AI market is anticipated to increase from just under $14 billion in 2023 to $103 billion by 2028, despite investors becoming progressively less risk-averse over the past year.
There are a tonne of AI health tech startups in the UK. Many are concentrated on the development of pharmaceuticals, genomic analysis, or more consumer-focused wearables and telemedicine symptom checks. Nevertheless, some people are working hard to improve disease detection and diagnosis. These include businesses like Mendelian, which recently got close to £1.5 million from the government to implement an AI-based solution for diagnosing rare diseases in the NHS.
Diagnostic AI startups are also widely spread throughout the rest of Europe. Radiomics, based in Liége, is one of them. The company's main areas of focus are solid tumor phenotypic quantification and imaging based on standard of care. DoMore Diagnostics is enhancing the prognostic and predictive value of cancer tissue biopsies in Norway by utilizing AI and deep learning. The company's founders also claim that it might aid in directing the choice of therapy to prevent over- and undertreatment.
Although they may be crucial for the sick person, a few percentage points of more accurate diagnosis may not be the sole advantage AI might have on our healthcare systems.
"The greatest opportunity offered by AI is not reducing errors or workloads, or even curing cancer: it is the opportunity to restore the precious and time-honored connection and trust—the human touch—between patients and doctors," writes Eric Topol, author of Deep Medicine: How Artificial Intelligence Can Make Healthcare Human Again.
No comments:
Post a Comment