Brandon Park - Section Editor, Medical Technology Department
In “Machine intelligence in non-invasive endocrine cancer diagnostics”, Nicole Thomasian et al. discuss how endocrine cancers, which develop from hormone-producing cells, need to be properly diagnosed in order to be effectively treated. However, the present techniques to do so, such as biopsies and imaging, can be intrusive, expensive, and time-consuming. Through the analysis of vast volumes of information including imaging, genetics, and clinical records, machine intelligence (MI) has the potential to enhance the diagnosis of endocrine cancer. MI can employ radiomics, which utilizes algorithms to extract information from medical pictures, to find patterns suggestive of malignant tumors. Genomic analysis can also be used to find mutations or cancer-related biomarkers. However, before being used in the clinic, MI needs to obtain a large representative dataset and give doctors the ability to interpret complicated algorithms. Despite this, MI offers the potential to non-invasively, precisely, and quickly diagnose endocrine malignancies.
In low-resource situations, where there may be a shortage of medical personnel or insufficient infrastructure for diagnostic tests, MI has the potential to increase access to healthcare and can be applied to other areas of medicine. By analyzing vast amounts of data and forecasting the spread of the disease, MI can also assist in the diagnosis and management of infectious disorders like COVID-19. The article also discusses potential ethical issues with MI's application in global health, including data privacy, algorithmic bias, and the possibility that MI would worsen already-existing health disparities. As a result, the authors stress the significance of developing and using MI responsibly, including making sure that technology is planned and applied in a way that is equitable, transparent, and respects human rights. If all the ethical concerns are addressed, MI has the power to significantly enhance access to healthcare for underprivileged groups and have a positive impact on global health.
References:
Thomasian, N. M., Kamel, I. R., & Bai, H. X. (2022). Machine intelligence in non-invasive endocrine cancer diagnostics. Nature Reviews Endocrinology, 18(2), 81–95. https://doi.org/10.1038/s41574-021-00543-9.
Comments