Amazon can now extract data from medical records with AI

This is just one example of how powerful machine learning can be. Amazon’s latest service allows researchers to extract useful information from patients’ medical records. This can potentially help healthcare providers and scientists save both time and money. The service will be called ‘Amazon Comprehend Medical.’ The service was announced by Amazon in a blog post. It will assist healthcare professionals and researchers in identifying key patient data. This data includes patient diagnosis, treatments, dosages, symptoms, and signs, etc. The service does this in quite an extraordinary manner. The data is quite often written or saved in an undecipherable or illegible manner. Using machine learning and text analysis the service can extract data from a variety of different sources. These sources can consist of prescriptions, notes, audio interviews, and test reports, etc. It digitizes the information first and then stores it according to categories that I mentioned earlier. Amazon said in its blog post,

“Identifying this information today is a manual and time-consuming process, which either requires data entry by high skilled medical experts, or teams of developers writing custom code and rules to try and extract the information automatically.”

This service doesn’t just help healthcare providers or researchers. Patients greatly benefit from the service as well. Patients can use Amazon Comprehend Medical to get a deep analysis of their condition, and then prepare a detailed plan accordingly. Scheduling hospital visits and medication, managing treatment options and checking insurance validity becomes much easier this way. Amazon Comprehend Medical complies with the U.S. Health Insurance Portability and Accountability Act of 1996 (HIPAA). This means the data will be portable, will have integrity against fraud and abuse and will be simple to administer. The e-commerce giant says that it will not store any of the processed data on its servers or use it to tweak and teach its algorithms. All patient data will be encrypted and would only be unlocked if the patient has a key for it. The service is being put through its paces by Fred Hutchison Cancer Research Center in Seattle which is using it to identify patients for clinical trials and Roche Diagnostics. The cancer research center used the service to analyze millions of patient notes. Thanks to the software, the center was able to process each document from hours to seconds. Time is a major factor in medical research. Lives can be saved every second. The most important field of medical research is oncology. Cancer research is a time-consuming process. Designing clinical trials and finding suitable candidates for a trial can take weeks and even months. Researchers need to go through piles and piles of unorganized patient data before designing such clinical trials. Amazon Comprehend Medical can cut down this time from hours to just seconds. Time is one of the biggest hurdles in converting research into a practical solution. Researchers will get access to the right information in due time by minimizing the time overhead. This will help scientists create lifesaving cures for not just cancer patients but patients suffering from other diseases as well. Stay tuned for more updates!

---

Related reading

• Ray Kurzweil's Predictions Scorecard: What He Got Right & Wrong
• AI in Healthcare: From IBM Watson to Modern Medical AI
• Viral Customer Service Moments That Built Brand Loyalty in the Social Media Age
• Meet Norman AI; the American Psycho of Digital World