The combination of AI-related technology with cancer research is nothing new, however with companies like DeepMind investing heavily on this subject, there are new innovative grounds being unlocked which never seemed possible before. AI may be a topic that everyone is ready to digest just yet – however the results, however little they may be right now, do point towards a future where AI-based medical science could be a reality.
Recently, with DeepMind opening another branch in Japan, a new journey into the reviewing of thousands of images and scans has begun as a means to test the AI systems and see if they are able to accurate spot cancerous tissue.
DeepMind has also been known for their previous work in the UK, most notably their collaboration with the Cancer Research UK Imperial Center at Imperial College London. Here they analyzed a collective 7500 mammograms. “The involvement of the Jikei University Hospital in a global research partnership will help take us one step closer to developing technology that could ultimately transform care for the millions of people who develop breast cancer around the world every year,” is what professor Ara Darzi, director of the Cancer Research UK Imperial Center had to say about this partnership.
With AI systems, one needs to be extra mindful of how one chooses to train the AI system. As stated by Dominic King, the clinical lead of DeepMind Health, there is a chance for bias to occur when undergoing these processes. This can result in serious problems – and most notably, can lead to variations in results. In some cases, the different answers may be missed, or wrongfully identified, all of which are unsatisfactory results.
From this, it does become certain that AI programming is a largely trial and error-based system. One needs to manually feed thousands upon thousands of images, and even then to make sure that there isn’t any bias or other notable problems in the data fed. This inherent sluggish pace is perhaps the major reason behind why strides in this field are hard to come by. However, with the success that DeepMind achieved in UK, they believe that their new venture in Japan will prove to be equally fruitful.
However, DeepMind isn’t the only one investing in AI-based medical science. Google has shown interest in this field, has been quite strongly focused on creating new AI-related healthcare programs. A new AI system is also set to release this Spring that forecasts the mortality rates at hospitals, with over 90% accuracy.
With new and improving technology, and successes in this field, there truly are no limitations. People can potentially unlock multitudes of features, and with proper backing, companies will not have any restrictions in their research. Healthcare is a constantly evolving force, and it is one that is becoming the focal point of the current generation. With fundamental breakthroughs being done in the fields of Physics, and Computer Science, many believe that it is time for Biology and Medical Sciences to take center stage.