The field of healthcare has always been a major leader in technological innovation and advancement. As diseases and viruses continue to mutate, medical treatments and procedures must always stay ahead of the curve. Through the use of artificial intelligence (AI) and machine learning platforms, healthcare professionals can better treat their patients and help them live healthier, happier, and longer lives. AI in general is advancing at a surprising pace and according to the Future Healthcare Journal, we will see it increasingly deployed in the healthcare field as time goes on.
So, how does AI itself actually contribute to the identification and diagnosis of diseases and other ailments? The results of various scans and genome analyses can be fed into AI healthcare systems which in turn interpret the readings and make their own diagnosis. This can be used not only to confirm a doctor’s suspicions about the likely illness, but also to identify a potential issue the doctor may have missed. Likewise, this type of AI can be applied to general screenings in the hope of identifying undiagnosed conditions or even spotting patients who are more prone to increased risk of disease.
In fact, a study published by The Lancet Digital Health compared the results of AI detection versus those of a regular healthcare professional, using a sample of over 31,000 studies performed between 2012 and 2019. The thorough study found that within the past few years AI has reached a point where it is now a viable and reliable source of diagnostic information. A comparison of the deep learning models and healthcare professionals studying the exact same samples found the results to be equivalent.
Real world examples of AI saving lives through automated pathology detection are not overly difficult to find. Take, for example, the case of a young woman returning to NYU’s Langone Health center for a routine follow-up concerning her history with brain cancer. The checkup indicated the medulloblastoma had returned. This recurrent form of cancer was found in the same area of the brain as before, and the biopsy appeared to confirm that it was indeed a medulloblastoma. The average doctor would have accepted this diagnosis and initiated a specific course of treatment of radiation and chemotherapy targeted directly at this unique form of cancer. Fortunately, her doctor Matija Snuderl hesitated in this case; ordering a full-genome analysis purely out of an abundance of caution. He then fed the results into an AI system developed by scientists at the German Cancer Research Center, and let the computer make its own diagnosis.
“The tumour came back as a glioblastoma, which is a completely different type,” claims Snuderl. This new cancerous mass had formed as a result of the radiation used to treat her initial medulloblastoma. A different type of cancer means a unique treatment plan including particular forms of radiation therapy and drug regimens. If Snuderl had attempted to treat the wrong form of cancer, not only would the treatment be completely ineffective at actually destroying the cancer itself, it would cause serious harm and side effects to her body. “If I had finalized the case just on pathology, I would have been terribly wrong,” Snuderl says.
The AI platform used by Snuderl in this case is still in its infancy and has already proven that it can save lives. NYU Langone received permission from the state to deploy the AI system for diagnostic purposes only as recently as October of 2019 and it appears to be quite effective. It relies upon its ability to spot patterns far too subtle for the human eye to pick up. This also means that in some cases, the AI is able to spot potential health issues far before any human physician or even before any symptoms begin to show. Catching diseases in their early stages means easier treatment options and higher survival rates. It also means treatments are typically less invasive, have less side effects, and are significantly cheaper. This has the potential to considerably reduce strain on the already overloaded healthcare system while also saving untold amounts of money for patients, healthcare providers, and the government.
Pharmaceutical giant Bayer has begun working in tandem with tech companies and healthcare professionals to create a software that will not only be able to screen for and detect diseases, but also formulate and develop new drugs to help treat them. By partnering with hospitals and researchers, Bayer has continuously refined the data that machine learning uses to make a smart diagnosis. Data points fed into the system include everything from symptom analysis, test results, medical imaging, doctor reports, and more. “We can model how it will behave in a cell in combination with other drugs the patients might be taking. We’re looking at how we can identify the right patients and sites to run our clinical trials. We would be able to run shorter studies and show where the medication is the right one for those patients earlier,” claims Angeli Moeller, head of Artificial Intelligence Projects at Bayer.
Angeli was quick to emphasize that the AI platforms are not meant to take over the treatment process. At the end of the day, the patient and doctor still have full control over their treatment as the software is intended to be used supplementally. She stressed the AI’s primary function as “decision support,” ensuring the patient always feels comfortable and in control.
Bayer isn’t the only company seeking to tackle AI treatments and disease management. According to a report published by BenchSci, there are an excess of 230 startups worldwide using artificial intelligence in drug discovery. A few of the biggest and most successful projects include IBM’s Watson for Health, Google Health, AI-Rad Companion Chest CT, and AI-Pathway Companion.
Watson for Health focuses on improving the decision making process, streamlining workflow, protection from fraud, and enabling low-cost approaches to medical research. Google Health excels in monitoring and measuring personal fitness programs while providing relevant information about their medical conditions. It also tracks medication regimens and provides reminders for dosages and timings. AI-Rad Companion Chest CT is a more specialized AI created by Siemens Healthineers that is able to read and analyze chest CT images, automatically take organ measurements, and compile valuable medical reports composed of clinical images and data points. The AI-Pathway Companion is another platform created by Siemens Healthineers that compiles vast amounts of data from medical imaging, pathology and lab reports, genetic sequencing, and more to optimize treatment along disease-specific pathways. It leverages this data to provide both standardized and personalized treatment recommendations using evidence-based guidelines. It seeks to keep patients at the center of treatment while providing transparent insights for diagnosis and treatment.
It is clear that healthcare AI platforms have a lot to offer when it comes to diagnosing, treating, and managing diseases. Not only can they save lives and extend healthy lifespans, they have the potential to ease strain on the healthcare system while saving both patients and hospitals significant amounts of money. Industry leaders are predicting not only imminent improvements upon existing platforms, but also the introduction of AI into many other facets of medical practice. Exactly what kind of new solutions they will offer is still yet to be seen.
Contact us to see why the brightest companies trust Lithios.
Get in touch
Social
.svg){kind=logo}
©2023 Lithios LLC. All Rights Reserved
