Hey, Microsoft! Read my notes
Software giant Microsoft announced the release of an Azure update that will enable users to extract insights from medical data. The Text Analytics for health is available now as a preview and uses Natural Language Processing (NPL) to find and extract medical information from unstructured clinical documents, such as doctor’s notes, medical publications, or electronic health documents.
According to a blog article by Hadas Bitran, Microsoft Healthcare group manager, being able to find and record these snippets of information in healthcare and clinical settings can help create prediction models from historical data, match patients to clinical trials, and improve clinical quality reviews.
One of the COVID-19 resources that I mentioned in a previous newsletter, the CORD-19 open research dataset, was created by the Allen Institute for AI in partnership with Microsoft and leading research groups to provide researchers with a quicker way of gaining insights from the overwhelming amount of COVID-19 information. This highlights the role that these types of AI tools can have in quickly retrieving and assessing information.
If, like me, you start hearing “security” and “privacy” alarm bells ringing in your mind, know that according to Bitran, “Text Analytics for health is currently available in containers. With containers, you can deploy resources in your own development environment that meets your specific security and data governance requirements.” That’s better, I suppose.
CRISPR, CRISPR, baby
Do you remember the CRISPR babies’ scandal? Go on, I’ll wait. Are you back? CRISPR has some targeting issues, meaning that it’s not as precise as you might think. You aim for a certain region of the genome but the gene-editing tool can make ‘off-target’ gene mutations, which are difficult to assess because you’re focusing on the region that you want to change.
Three recent studies available as bioRxviv pre-prints, show the same problems and reveal unwanted changes affecting large regions of the genome. You can read more about the problems surrounding CRISPR gene editing and links to the three preprints in this Nature article.
Now, scientists have discovered a strange bacterial enzyme that is more precise than CRISPR and can enter mitochondria, the cells’ energy-producing structures. This technique is still in its early days, but it has the potential to act directly on DNA without relying on the Cas-9 enzyme to break it.
Another tool in the scientists’ arsenal to potentially correct mutations and treat mitochondrial disorders.
Any drug is better than no drug?
Biogen recently filed an FDA application of its monoclonal antibody (mAb) aducanumab for the treatment of Alzheimer’s disease. If approved, it will be the first drug to reach the market with the potential to modify the disease progression.
According to an article in the Clinical trials Arena website, “Aducanumab is a recombinant human mAb that binds primarily to aggregated forms of Aβ, including soluble oligomers and insoluble fibrils, but reportedly does not bind Aβ monomers. The potential approval for aducanumab is not likely to be the final treatment for AD, but it could open doors for combination research, namely anti-Aβ therapies with anti-tau-therapies”.
Highlight of the week
I’ll be presenting my first EMWA webinar next Thursday! I’ve always had an irrational fear of stages, and while the digital version does provide some relief, I already feel a lump forming in my throat. Send me your good vibes, will ya? That will be all from me this week. Do you have any comments or questions? Leave a reply below 🙂
Other posts you may like:
- New to medical devices? Here are 3 insights about Clinical Evaluation Reports
- Are you an aspiring medical writer? Apply for the Geoff Hall scholarship
- This week, less is more
- How AI and gaming can improve pharma and medtech
- Biopharma this week—(nearly) COVID-free edition
- COVID-19 long haulers, smell tests, and other updates