Artificial Intelligence (AI) has cemented itself into a very robust role in the world in the past couple of decades. However, most do not consider the many ways we interact with AI every day. Just by using your social media, emails, or the various apps on your phone, you are involved with AI algorithms.
One of the biggest areas AI is developing in is the healthcare industry. In many critical disease cases, treatment is much more effective if the disease is detected early. In an ideal world, symptoms would present themselves early enough and allow for ample time to seek help. However, that is not the case and often we may not know we are sick until the disease is advanced.
In our post-Covid-19 world more people are interested in their health however, short visits to your doctor may not show you the full picture of your health as a comprehensive health assessment would. That is because they look at statistical health risks rather than focus on you as an individual.
As a result, various technological advancements and AI can help not only with the first levels of screening – small changes that may highlight underlying issues – but research has shown that they can detect serious diseases such as heart disease or lung cancer earlier. These combined would offer people a better chance of effective treatment.
In this blog, we will be exploring some of the ways in which artificial intelligence can improve healthcare and disease detection.
Artificial Intelligence in major disease areas
According to Our World in Data, the top three causes of death in the world are cardiovascular disease, cancers and respiratory disease (Ritchie, 2018). With these diseases consistently being at the top, we must put resources into these areas and improve early detection, diagnosis, and treatment.
AI can provide many benefits in early detection by being able to show any risks that an individual may have.
A study from 2017 involving patients at risk of stroke used AI algorithms based on symptoms and genetic history to place them into an early detection stage. The study found that the early detection alert from the algorithm provided 87.6% accuracy in diagnosis and prognosis. It allowed for earlier implementation of treatment and prediction of whether a patient had a higher risk of future stroke (Jiang et. al., 2017).
In the UK, only about 5% of people diagnosed with stage 4 lung cancer will survive their cancer for 5 or more years, compared to 55% of those diagnosed with stage 1 (Cancer Research UK, 2020). Artificial intelligence has been studied in relation to this disease and several breakthroughs have now shown that a computer was able to find tumours in scans of patients with more accuracy than professional radiologists (Ardila et. al., 2019). The team reported that the system correctly detected early stages of lung cancer 94% of the time.
Further expansion into cancer detection through AI has shown that AI can diagnose colon cancer more accurately than a trained pathologist. Researchers gathered images of colon cancer from 8,803 participants from various independent centres and used them to train a machine learning program.
After the development of a performance measurement tool, they were able to compare machine learning to the work of real pathologists. It was found that the average pathologist scored 0.969 for accuracy of colon cancer identification compared to 0.98 for machine learning (Yu et. al., 2021).
One of the bigger advantages of AI or machine learning systems is that they will be the drivers of population screening. This is currently more aimed at lung cancer screening and there is evidence that volume screening programmes would be effective. This opens opportunities for volume screening for other types of cancer and other chronic diseases.
Artificial Intelligence and diabetes prevention
Other medical areas where artificial intelligence has great helping potential is diabetes. Over the past few years, researchers have delved into various methods to utilise AI for diabetes management. Some strategies include self-management, wearable devices, and remote monitoring.
Research from 2020 investigated information from many continuous glucose monitor insulin pumps using AI and big analytics among people with Type 1 diabetes. People with type 1 diabetes must test their blood sugar levels often to determine how much insulin they have and if they need more.
However, by using continuous glucose monitors they can get a much better idea of their insulin levels. The data gathered by the AI allowed researchers to create models that are much better at predicting the effect of meals and insulin on glucose levels which led to better control of blood sugar levels.
Various AI-backed health tools have shown great improvement in diabetes management including the reduced need for in-person appointments (Navarantha, 2020).
Artificial intelligence has been shown to be important in diabetes prevention. As Bhardwaj et. al., (2018) found “the improvement in diagnostic accuracy and risk prediction and reduction of hospital readmissions has resulted in a significant decrease in health care cost. Big data analytics shows initial positive impact on quality of care, patient outcomes and finances, and could be successfully implemented in chronic disease management”.
As more and more research is carried out into the medical capabilities of artificial intelligence, technology is growing as a great tool for highly effective chronic disease management and prevention.
Preventive health assessments
While fully AI-backed health assessments are not yet available to the public, we still have the technology available to give you a comprehensive picture of your health.
Echelon Health have access to the best imaging technology available today. This imaging technology, through CT, MRI, and ultrasound scans, in combination with a complete blood profile allows us to detect up to 92% and 95% of preventable causes of death among men and women, respectively.
For example, a team in the Netherlands studied the impact of volume CT screening. They followed over 15,000 current and former smokers over 50 years of age for around 10 years. By the end of this study, they found that those who went through regular screening were about 25% less likely to die of lung cancer. The rate of false positive and false negative results also greatly decreased (de Koning et. al., 2020). This shows that the right technology used at the right intervals is extremely likely to reduce your risk of certain diseases.
With this in mind, Echelon Health have built their flagship health assessment – the Platinum assessment – which looks at your body from head to toe and leaves no stone unturned. The following tests and scans are included in the Platinum assessment package:
- Blood Tests
- CT Aorta
- CT Heart
- CT Coronary Angiogram
- CT Chest
- CT Pelvis
- CT Virtual Colonoscopy
- CT Bone Density
- CT Upright Skeleton
- MRI Brain
- MRI Cerebral Artery Angiogram
- MRI Carotid Artery Angiogram
- MRI Prostate
- Ultrasound Thyroid
- Ultrasound Testes/Ovaries
- Digital Mammogram
- Full Body Mole Screen
At Echelon Health we are focused on preventive health screenings and how they can help people better understand their health and how to improve it. It is far easier to take care of yourself and maintain your health when you have complete peace of mind.
If you would like more information about the assessments provided by Echelon Health, please do not hesitate to contact us at any time.
Ritchie, H. (2018). What do people die from? Available at: https://ourworldindata.org/what-does-the-world-die-from (accessed 26/07/2022)
Jiang, F., Jiang, Y., Zhi, H., Dong, Y., Li, H., Ma, S., Wang, Y., Dong, Q., Shen, H. and Wang, Y., (2017). Artificial intelligence in healthcare: past, present, and future. Stroke and vascular neurology, 2(4).
Cancer Research UK, (2020). Lung Cancer survival. Available at: https://www.cancerresearchuk.org/about-cancer/lung-cancer/survival (accessed 26/07/2022)
Ardila, D., Kiraly, A.P., Bharadwaj, S., Choi, B., Reicher, J.J., Peng, L., Tse, D., Etemadi, M., Ye, W., Corrado, G. and Naidich, D.P., (2019). End-to-end lung cancer screening with three-dimensional deep learning on low-dose chest computed tomography. Nature medicine, 25(6), pp.954-961.
Yu, G., Sun, K., Xu, C., Shi, X.H., Wu, C., Xie, T., Meng, R.Q., Meng, X.H., Wang, K.S., Xiao, H.M., and Deng, H.W., (2021). Accurate recognition of colorectal cancer with semi-supervised deep learning on pathological images. Nature communications, 12(1), pp.1-13.
Navarathna, P., (2020). Artificial Intelligence to Improve Blood Glucose Control for People with Type 1 Diabetes. Rensselaer Polytechnic Institute.
Bhardwaj, N., Wodajo, B., Spano, A., Neal, S., & Coustasse, A. (2018). The impact of big data on chronic disease management. The health care manager, 37(1), 90-98.
de Koning, H.J., van der Aalst, C.M., de Jong, P.A., Scholten, E.T., Nackaerts, K., Heuvelmans, M.A., Lammers, J.W.J., Weenink, C., Yousaf-Khan, U., Horeweg, N. and van’t Westeinde, S., (2020). Reduced lung-cancer mortality with volume CT screening in a randomized trial. New England journal of medicine, 382(6), pp.503-513.